Speedrun Community Survey Analysis

Initial survey sent: 6/26/2017

Survey analysis completed: 10/22/2017

Link to this document: https://goo.gl/5msNXE

Link to YouTube audio recording of this document: https://goo.gl/7AcfLQ

Link to original survey description: https://goo.gl/6RbR9e

Link to questions & responses from participants: https://goo.gl/wEJykD

Table of Contents

  1. Opening thoughts

  1. Survey responses & analysis
  1. General
  2. Resources
  3. Decision making/game specifics
  4. Timing methods
  5. Hardware differences / conversions
  6. Leaderboards
  7. Cheating
  8. Community Events
  9. Community Content
  10. Other

  1. Closing thoughts

I. Opening thoughts

This survey was created out of personal curiosity about some baseline community issues and how community members generally approach them. Speedrunning is a large community, which both exists in a ‘greater’ form through large marathons, discussion forums and aggregated leaderboards, and also through its many specific and individual game/series communities. A goal of this survey and its analysis is to tap into some of the mindsets, histories, and motivations of how community members feel about a range of topics.

In my time in this community, I’ve witnessed many ideas and debates on how various problems should be handled & resolved. I believe there to be many topics for which community members think “yes, obviously this issue should be handled way”, but perhaps don’t have a good understanding why others who disagree similarly have the same sentiment of ”yes, obviously this issue should be handled this way”. I have not seen a resource that has attempted to tap into understanding the breadth of opinions out there for a set of given topics, and that is another primary goal of the survey.

Before getting into the analysis, a few ideas should be conveyed up front:

Major thanks to everyone who completed the survey and spread the word. Special shoutouts to:

Please enjoy the survey’s analysis. The topics start off fairly light with brief commentary, then get more and more granular and detailed as the topics become more complicated. As the reader, please view each discussion point in a positive light - at no point whatsoever do any of the discussion topics have the goal to create divisiveness or inspire separation among pools of thought. The idea is to understand what and how different community members think, and how we can bridge the gaps. I hope through review of these topics, we can promote ideas & efforts for discussion and community-building within speedrunning.

- cleartonic

II. Survey responses & analysis

A. General

A1. Where do you _primarily_ go for community resources & communication for a game/series?


This question asks respondents to discuss briefly where they primarily interact and communicate with their communities. Communication platforms (social interaction, like Twitter) & resource platforms (a wiki-style page/forum) can certainly be different things, but can also certainly be the same thing for communities (Discord), which is why multiple selections option was granted for the question.

Let’s briefly review the options:


This question allowed respondents to select multiple choices. The goal is to gain a general sense of what respondents are using for communication, rather than to attempt to define one method of communication per respondent.

Overall, these results are representative in the shift of how community members communicate since the rise of Discord. Both SpeedDemosArchive (SDA)’s forums and SpeedRunsLive (SRL)’s IRC channels were at one time major hubs for many communities to actively discuss speedruns, and are likely where many of communities attribute their roots to. But over time as Discord began to blossom as a more UI-friendly version of SRL’s IRC channels and has the potential for historical discussion logging similar to forums, naturally forum & IRC based platforms diminished in usage.

As this question is tailored for both resources & communication, through the write-in’s it appears that respondents leverage speedrun.com (SRC)’s organization for resources (VODs of personal bests, guides & tutorials, wikis, etc.) and not necessarily rely upon their forums (evident if you click around SRC’s forums), and use other means of platforms (Discord, Twitch Chat) as their primary source to communicate.

Write-in & Analysis

Write-in responses were summarized below among some shared sentiments:


In current times, Discord is the king, which rose very quickly as a top aggregator of both resources (via pins & links) and discussion (individual game channels among series). SRL IRC and SDA forums are commonly considered mostly deprecated, but are still very much in use among (presumably) older speedrun community members (and of course, SRL IRC is still used for races). Still, the necessity for solid resource pages outside of Discord certainly still exists, where some communities leverage SRC’s resource options for organization, whereas others have more specific or customized webpages tailored to their needs.

A2. When do you first consider that you joined the speedrun community, either as a runner, a researcher or a fan?


The response categories are organized annually, with the exception of pre-2011 responses being combined.


The data suggests the spread among most post-2010 years is fairly even, with a larger pool of newcomers in 2013, then 2014. Pre-2010 is roughly tied with 2013 for the largest share, and 2016/2017 have the lowest shares.

Write-In & Analysis

The pre-2010 category should generally be taken with a grain of salt. It seems like some of the responses (through review of the write-in’s) try to reflect ‘earliest date of speedrun exposure’ versus ‘earliest date of joining the community’, which perhaps should have been clearer in the question.

Major rises in community participation likely correlated to similar rises among streaming platforms, where ustream & justin (older streaming platforms) had specific and niche audiences, then eventually grew into Twitch & other more modern audiences. General Twitch growth, YouTube videos, and GDQ exposure are commonly cited write-in reasons for respondents getting exposure & becoming invested.

An overall common theme among responses is that community members seem to say that they knew about speedrunning, but took some amount of time to actually become involved with the process & community:

The concept of speedrunning is appealing to many, but likely community members have needed time to fully ingest what the process is about and how to actually do it, especially before the days of modern GDQ (which helps package the hobby for a viewer to understand). With the advent of more readily available resources (tutorials, guides, or communication platforms), the time delay between ‘hearing about speedrunning’ to ‘becoming a speedrunner’ has lessened, compared to older days where information was more scarce and/or fostering an understanding of speedrunning was generally harder.

Some respondents reflect upon their time with ILs in games/series such as Super Meat Boy, Sonic, Goldeneye 007, racing games (e.g., Mario Kart) far in advance of joining the modern speedrunning community.

Some games happen to have very old communities that largely existed before any sort of greater speedrun network (e.g., Quake, Super Metroid, Metroid Prime). Super Metroid is a good example of a community that grew alongside modern speedrunning from its initial roots (a community with a longstanding history over different websites).


Growth has been fairly steady over the last few years, and likely will continue to grow as major speedrun events and major streaming platforms both continue to similarly grow. The grassroots approach of finding speedrunning through some obscure medium may mostly be a relic of the past (although certainly might still happen), and more mainstream avenues will bring in newer community members.

A3. How did you first hear about speedrunning, and what made you join? (optional)

Write-In & Analysis

Larger streams or popular videos are commonly cited starting points from respondents for their first exposure:

Similar to question A2, many respondents reflect on knowing about speedrunning for awhile, but having some sort of rekindling and rediscovery to really dive in.

A handful of respondents reflect on not even considering something existed, and immediately being hooked as soon as the idea of speedrunning was clear to them:

Certainly a case-by-case experience reflection for each respondent. Reviewing the data set for this question is recommended - it’s interesting to view who (the respondent) was inspired by what videos/runners as their first exposure.

B. Resources

B1. How generally helpful do you think the greater speedrun community is for helping newcomers currently?


This question was designed as an ‘overall’ approach made by speedrunners to provide help to newcomers, but naturally many respondents wrote-in their experiences with specific communities to help paint the broader picture. This question comes at the intersection of having resources available and having community members available to discuss with newcomers. Both are important aspects but certainly an either/or approach is viable (one could easily use resources without talking to community members, or one could easily not look up resources & just talk to a community member). Another important consideration for this is how communicative the newcomer is with their concerns, which is generally not any fault on the community if newcomers lack the proper communication to get the help they need.

The purpose of the write-in section was to allow for reviewing these engagements. However, the ranking of 1-7 may have been difficult for some respondents to adhere to.


The data suggests a fair spread for the assessment helping newcomers, weighted much more towards the 4 through 6 range. The steep decline in 7 responses suggests that most respondents are willing to say “good, but not perfect”, meaning somewhere between 4 through 6 is probably interchangeable among respondents (e.g., if someone put down a 5, they probably could’ve put a 4 or a 6 without thinking much about it). Some respondents reflect on a bad experience they’ve had or heard of, likely swaying the data away from the perfect 7.

Write-In & Analysis

Write-in responses for this question are helpful for understanding what shapes a community member’s assessment of how they perceive the greater community approaching helping newcomers.

The phrase “general speedrun community” seemed to be critically analyzed by respondents. The question is reflecting on how one perceives the general practice of speedrunning and its ability to accommodate newcomers. Yes, the community is quite large and at times fragmented, but one’s individual experiences still comprise the whole picture, and thus it is expected respondents have some sense of how the bigger picture looks (either narrow or broad, depending on the respondent).

Qwertylol says:

“A majority of the community is very helpful, but a select few do not help at all, or very little.”

To pivot off this idea, it is a burden of any given community to have only a few individuals who are organized and willing to help with auxiliary organization of resource & content availability. In my experience as the analyst of this report, many communities I’ve seen are strung together by one or a few individuals who are designated experts about a game/series, and the rest of the community pivots on them for knowledge. It is not necessarily a bad thing to have only a few experts or auxiliary content providers, but the burden being on them to put together resources due to their knowledge base can be problematic if they decide to stop being interested or pursue other things. Likely more community members don’t get involved with auxiliary content for incentivization reasons (sometimes community members just want to speedrun and not put together tutorials, guides, wikis and other videos), and expert community members may grow tired or uninterested in being the pivotal resource that pulls everyone together, especially newcomers. This definitely is a case-by-case occurrence.

Bony highlights:

“Would prefer if almost everything was more centralized. More content creation (e.g. video tutorials and proper organization of said tutorials) would also be ideal, although this has improved dramatically in recent years from what I can tell. Accommodation is generally very solid when you know how to locate it.”

The last sentence is a good assessment of what many of the respondents have wrote-in. Resources are sometimes available, but can be challenging to locate. Centralization seems like a buzzword, yet is a simple concept - making resource and communication standardized so that when a newcomer comes along, they don’t waste unreasonable amounts of time trying to pin-down what they need to know. Any random given community might have a very robust set of resources on SRC, and any other random community might exist on an unlinked set of documents that someone would never know how to find unless they asked the right person.


The outlook for availability for communities to help newcomers is generally positive. Many community members reflect that things have generally improved over time in this regard. The take-away is that on a community-by-community basis, resources range from being scarce and hard to find, to abundant with an army of community members ready to answer a newcomer’s questions. Individuals make up the specific community, and specific communities make up the whole of speedrunning - it would be wise for communities to assess what kind of information they’ve made available to the wayward newcomer and what they could do to help/communicate with them.

B2. Do you have any examples of communities that are either have robust resources available to people, or communities that lack resources but could benefit greatly from them? (optional)

Write-In & Analysis

Respondents typically reflect upon positive experiences using resources and commonly citing other communities’ examples, rather than reflecting upon negative experiences and commenting on how a community doesn’t have resources. Another point- a community made up of just a few community members shouldn’t reasonably be responsible for a robust set of resources.

This section will be broken down by popular examples for robust resources. This is a general reflection of what respondents wrote - data mining this style of question would be both difficult to achieve and not meaningful. We’ll briefly highlight these examples:

To no surprise, many games here are among the most popular speedruns. But certainly there are games missing from this list too that one might expect (certainly could be that respondents simply didn’t think about it, as part of the sample.) But it very well could be that part of a game’s popularity may have to do with its resource availability for newcomers.

This list is certainly not comprehensive, but illustrative of the sample of respondents. Some communities with less organized/available resources were criticized & listed in responses, but for purposes of this analysis it seems fair to highlight those who have been given praise rather than call out those with lesser available resources. Reviewing the data set will help illustrate the negative cases to those readers who are interested in the responses.


There are no major surprises among these results. Larger western communities have more notoriety among their community members and other outside members when they have good resource pools. Still, a robust set of resources is of course very beneficial to newcomers (and even existing community members), and generally looks positive on a given community to have them well organized and available from speedrun community members both inside and outside that given community.

B3. Have you ever felt like you wanted to learn a speedrun, but couldn't find the resources or get the help you needed? If so, what kind of resources do you think would have been helpful?


This question allowed respondents to reflect on their personal experiences with respect to trying to learn and run a new game. This is a fairly specific question and is strongly affected by the specific sampling of this survey, but nonetheless offers some baseline intuition about whether or not community members feel the need to have resources available to them in order to learn a game.


Roughly 2/3rds of respondents have no issue with wanting to learn a game and figuring out how to approach or learn what they need to. The write-in section below will be helpful for understanding what difficulties encountered by those who answered “Yes”.

Write-in and analysis

Below will point out specific reflections that help paint the picture as to why the majority of respondents don’t feel like they couldn’t progress with a game:

“No” answers:

Some common No case sentiments:

Naturally some games are much more complicated than what one run will demonstrate (for example, some RPGs have many scenarios that you must be prepared for, but watching a video of the fastest time only shows you one of them), which is why being able to ask community members for help is an important factor. Many respondents generally reflect on their ability to ask community members for help & independently look up videos without too much issue.

Some responses highlight how doing your own set of research and analysis is a good (and perhaps preferable) way to learn a game in the absence of resources:

Other responses include a general reflection that whatever games they have learned to play, they were able to benefit from resources available to them.

“Yes” answers:

The range of what respondents reasonably ‘expect’ to be readily available is fairly wide. Many respondents here hoped to at least have a video available, then further some notes, video tutorials, wikis, then even getting into specific tutorials and step-by-step approach guides & specific forums.

Yes cases list a few types of resources that respondents say would have been helpful:

The recurring common theme among these responses is that there are hidden bits of information that a video cannot provide. Though it ranges from game to game, there are techniques, tricks and overall understanding of a route that may be lost on newcomers without someone to ask or a resource to glean from.

Some respondents reflect that as time goes on, information and resource availability is far superior than that of the past:

Lastly :



The ratio of yes/no in this scenario is not particularly useful, due to how respondents’ experiences with particular games and interests would directly influence their answers, and those interests are likely incredibly varied. However, the sentiment that community members value & appreciate good resources is very much echoed throughout both “yes” and “no” answers. By no means is it necessary for a community to have them, but generally respondents reflect positively on a specific game/series community if they have great resources & the community members are willing to help, which is likely a catalyst for some game/series communities’ growth over others.

Something that is at times taken for granted is the amount of time and energy that goes into the creation of these resource materials. Some community members are incredibly selfless in their preparation of community materials, and likely put them together as an act to help others and grow their community rather than any self-furthering effort. And sometimes others don’t understand how much work these take, and somewhat expect resources to be available to them. As the analyst of this report, my opinion is that it’s really of no fault of a community (especially if it’s just one or a few community members) to not have the most robust resources. It is nice to have them, but realistically some individuals in the speedrunning community just want to speedrun, and spending hours putting together information & resources is not how they want to spend their time. Upon reflection of the responses and my general experience, many community members are willing to help others learn by sharing information & answering questions through discussion, regardless of what formal resources are out there.

The take-away is that well organized resources are massively beneficial, and may or may not be part of the reason why a community may grow. One should be aware of the time investment and effort to maintain such resources by speedrun community members, as it can be significant. But someone can always just ask another person for help, which is (usually) not a problem.

C. Decision making/game specifics

C1. When considering changes to category definitions and timing methods, do you think specific communities have any responsibility to consider outside speedrunners'/viewers' opinions?


The motivation for asking this question was to investigate both how members of the speedrun community who actively speedrun feel about outside opinions, and (to a lesser degree) how outsiders feel their influence and opinions matter. It is worth noting that speedrunners commenting/having an opinion on other (unrelated to them) speedrunning communities’ rulesets is equally interesting and worth understanding.

Some respondents pick apart the word “responsibility”, which gets into how respondents interpret the question. The idea was to convey whether or not communities should be open & mindful of external parties and should consider their input as valuable/necessary*.

*Part of the difficulty of writing a survey question is commanding the power of a particular word like ‘responsibility’ without the implications that community members HAVE to do anything, like a strongly suggested nice-to-have for community benefit.


Readers should view the blue and yellow together, versus the red & green. This means that 60.2% of respondents favor No (to not consider outside opinions), whereas 39.8% say Yes. To some degree, both “occasional exceptions” may be able to swing either way, but these answers were given to get a sense of respondents more on one side vs. the other.

The split between these is fairly even, with the slight edge to No cases. On paper, this is an issue respondents are divided on, but the results via write-in give helpful commentary as to how respondents generally feel.

Write-in and analysis

The following will demonstrate at first strong “No” and “Yes” cases, then the “occasional exceptions” together. This will be commented at the end, but even responses among the “No” and “Yes” show many signs of mitigation of similar ideas.

Strong “No” cases (should not consider outsiders):

Some overall No case sentiments:

Most stronger No cases emphasize how primarily runners (and to some degree their co-community members who may or may not be regular active runners) should be the driving force behind those who make decisions. Someone who plays a game often knows the ins-and-outs of how a game works, why categories are defined the way they are, and what generally makes most sense for speedrunning.

The sentiment among stronger No cases seems to be that respondents feel outside viewership, especially those who simply watch and don’t participate, are generally unimportant in the scope of defining how speedruns should be defined. Speedrunners should have the control to specify what a speedrun of their game should look like, given they are the most knowledgeable, and therefore outsider opinions are, by likelihood of familiarity of those games, less informed and less meaningful.

Strong “Yes” cases (should consider outsiders):

Some Yes sentiments:

There were the least strong Yes write-ins out of the four categories. Most stronger Yes cases reflect on category definition specifics and general consistency among the speedrun community as a whole (see the anonymous ‘any%’ reflection above). It seems as though respondents would like to avoid seeing speedrunning slip into a highly subjective, highly arbitrary (more so than it already is) set of rules and definitions, and generally value outside opinions on larger scale definitions.

Middle-ground cases:

Before breaking down these responses, the sentiments of the below will highlight the idea that many respondents are reasonably in the same line of thinking via their write-in responses, regardless of which option they chose above (i.e., a “No with exceptions” response may have the same general commentary of a “Yes with exceptions” response via write-in).

Some of the respondents, highlighted by the above sentiments, hint at the idea that speedrunning is a somewhat defined process that shouldn’t be tinkered or re-defined with too much. Certainly new movements take place to re-evaluate timing methods (e.g., SDA timing’s* obsolescence over time), but many games have fairly historied pasts that have their reasonings behind them for category segregation & definition.

* SDA timing refers to starting timing on ‘first control’, when the player actually starts moving. It’s generally been a subject of criticism over the years, and most runs now use some other more objective measure like “Game Start” or on console reset, which (in theory) is consistent among all players. More on this in a later section.

These two sentiments highlight some sort of reasonableness and reality check that some communities should undergo, among other similar responses. The idea that runners & their communities should be the authoritative force behind developing standards is fairly common and well-accepted. However, posited through some responses, the spread of a community ruleset that may be highly arbitrary can become the norm over time if general outside opinion is not necessarily reflected in a community’s decision making.

Some additional shorter commentary to help paint the landscape:

Darkman’s lengthy reflection well summarizes the common sentiment among respondents:

“It's a fairly complicated question with no black / white answer. Yes the runners of the community are typically the most knowledgeable about their run and as such should hold considerable weight in these discussions. Dismissing an outsider's thoughts is in my opinion extremely misguided as someone introducing their viewpoint doesn't harm discussion. Having outsider perspective on any situation is extremely important because the people that are the most passionate about something are often the ones that are the most jaded, and it is important to give a different perspective when discussing anything. That being said, outsider viewpoints do lack that intimate knowledge and experience that community members have been able to formulate based on what their audience enjoys and what they and their fellow community members join, so it becomes an interesting case of where do you draw the line? I don't think there is a single universal answer to this, and the only safe thing to say is never automatically dismiss a viewpoint just because someone isn't a runner. You never have to accept a viewpoint, but it's always important to acknowledge and debate the merits of it.”


Although the data results may seem to show a fairly robust split among answers, it’s evident in many of the write-in discussions that respondents feel pretty similar on the idea of considering outside opinions. The data in this case helps build the sentiment that some community members may lean more one way than the other, but generally that split is fairly ill-defined and likely respondents may go one way or the other (answering “Yes” vs. “No”) without feeling like they’ve compromised their ideas (via write-in).

Some respondents reflected on specific cases where they felt a community was being unreasonable. Certainly, some of those fringe cases still exist today among category definition and timing methods. But overall, community members seem to generally defer decision making to those who know best (the runners themselves and their core community of researchers), but are also sensitive to the idea that viewers and otherwise reasonable community members may have a general opinion on a topic and should be considered. It seems as though community members want to implicitly protect the time-tested process of speedrunning, without being either too archaic (not adapting to common standards) or too eager to change (considering outside opinions too heavily). The take-away is that communities should strive for an open-minded culture of considering outside opinions and having strong responses to why or why not some idea should be incorporated.

C2. When a new major skip, strategy or route is found by someone or a collection of people, do you think it's in speedrunning’s best interest to communicate to the related community immediately?


This question was expected to be one of the more divisive questions. It stresses whether or not 1) if community members think an obligation to communicate findings exists, and 2) do community members respond to the idea of ‘speedrunning’s best interest’ and think it's important or irrelevant. The polar-opposite answers choices avoid grey area type answers to help understand how respondents generally side with the topic. However, as evidenced by the below write-in’s, a fair amount of respondents actually do feel strongly about it being a giant grey area, and for that the question should have been designed able to accommodate that.

Part of the motivation for this question comes with sensitivity among some communities (from my understanding & experience in speedrunning) where community members have felt like their efforts have been a waste of time without valuable information. Some community members see it as a non-issue, others take offense to it.

Another word choice sensitivity with ‘obligation’ in one of the answers. Respondents seemed to have been hyper-critical of such word choices- certainly a lesson for the survey creator to understand the respondent’s viewpoints given the choices they had to choose from.


Respondents certainly favor sharing information as it is discovered in a 73% to 27% split. However, certainly explanations for respondents’ viewpoints will give color as to why they may answer Yes or No, and are very important for a topic such as this one. But overall, the sentiment leans in favor of sharing, without question, when respondents are forced to choose one or the other.

Write-in & Analysis

Most write-in’s are helpful for understanding what comes across as a question intended to make the respondent choose from a black & white answer pool(which, by design, it does). There were fairly lengthy responses via write-in to help clarify some viewpoints about this, however.

Stronger No cases:

A fair amount of respondents reflect on the ‘exciting’ factor of unveiling strats. Rather than thinking of it as ‘strat hoarding’, it’s more along the lines of pushing a new time with a cool new strat or discovery. Theoretically, once the video is available, it’s open to the public, and perhaps that is the discoverer’s way of concisely communicating their findings.

Some cases simply highlight that the person who underwent all the effort to discover something should be the first to reap the rewards from it:

Most stronger “No” cases aren’t suggesting ideas along the lines of that the discoverer is entitled to solely reap the benefits, but moreso that they have no “obligation” to do anything (and perhaps the respondent implicitly doesn’t see it as a problem, else they’d answer “Yes” to the question.)

Stronger Yes cases:

A handful of respondents posit that for runs to be considered wholesome & verifiable, major findings should probably be communicated. It’s worth noting that these answers likely are for major skips and not a minor time save:

Many respondents simply believe the idea of public knowledge to be a critical concept to speedrunning:

Speedrunning does thrive upon strat sharing, investigation of ideas & techniques, and combined-manpower route analysis. A subtle concept that recurs throughout this analysis is how some respondents view the idea of pushing a game to its limit for its speedrun - the concerns about competition and current record holders are secondary to the idea of a game being as figured out and as fast as possible. The idea of withholding information, as posited by this question, would certainly not be in the best interest of pushing a game as far as possible.

Pinkpajamas writes a somewhat cryptic message that’s worth highlighting:

“Sometimes it's not worth writing up a route when you're already working on obsoleting it.”

This respondent works on fairly intricate RPGs, typically rife with many routing options. The time and investment into a route that could be saved (or wasted) upon the drop of a major source of information could be very significant to a speedrunner and their route, especially for long & complex games. Sometimes strats, skips & techniques need to be communicated and fully fleshed out and understood before the sometimes long and arduous process of routing a game takes place, and that hinges upon community members knowing about it in the first place.

Ppotdot1 shares a reflection that adequately reflects a stronger sentiment to share information, driven by motivation to curb strat-hoarding:

“As soon as possible all current and previous runners should be notified of any new large discovery. This promotes a friendly cooperative and competitive environment. Keeping strats secret just for the sake of having an "edge" on everyone else for a short period feels underhanded and manipulative, and long term lowers the competitiveness of a game by reducing the quality of runs across a leaderboard.”

Many stronger “Yes” cases simply reflect of what’s the right thing to do for their community, and that a selfless approach in both helping others and helping the game progress is a common sentiment.

Middle-ground cases:

Middle-ground cases vary fairly heavily, but a large number of responses pick apart both the lack of ‘obligation’ and the non-immediacy of ‘immediately’ as reasonings behind their grey-area sentiment. Some say a few days, some say a few weeks, some say a few months, but overall, similar to the no-cases above, the idea still persists that strats will eventually be shared via video or some documentation.

The ‘immediately’ part of the “Yes” answer was written black-and-white, so many respondents in their write-in noted the above sentiment. The general reflection is that it’s the right thing to do.

But what about cases where not telling others what findings you have isn’t some sort of strat-hoarding, but more “time in the oven” kind of deal to understand and present itself properly?

The above sentiment is an interesting one, which gets into strategy findings and how much community members rely upon others’ information vs. how much subsequent discovery takes place. There may be cases where multiple hands on deck with a new glitch/skip might be for the best. There may be cases where the discoverer first fully fleshing out the new glitch/skip might be for the best. The goal of a community should be to take such findings and fully explore them, rather than consider something solved and figured out upon arrival, but that is certainly a case-by-case, community-by-community basis.

Quite a few respondents discuss about how speedrunners can basically do whatever they want without consequence:

Certainly, as of the writing this analysis, there is no formal ruleset to adhere to nor is there any oversight / authoritative group of individuals concerned with this sort of issue. We have a community, for which community members will be remembered if they do something wrong or ill-intended for their community and for speedrunning as a whole, but ultimately no real hard consequence exists.

Finally, a number of respondents comment on essentially the ‘hype’ factor:

Creating memorable moments for big events is certainly pure in intention. It might be considered unfair if you were to strat-hoard a major finding before a big race or event going into a major marathon, but otherwise, especially if it's a solo-run, a one-time only opportunity at a major marathon with a big unveiling can be exciting and non-harmful (if anything, perhaps accretive to generating community interest.)


The main take-away for this question is that responses are certainly way less black-and-white than the question allows for. Although the survey was designed to not allow for generic “depends tbh” kind of answers, this question is certainly one where respondents see a middle-ground between communication window and overall community benefit. There are certainly many implications for unveiling new findings, such as effect on a route for applicable categories, subsequent category definition, motivation from the community to investigate, and overall sentiment in cases where someone unfairly hid information. However, almost unanimously do respondents who wrote-in communicate that strat-hoarding for any sort of personal benefit is certainly frowned upon.

D. Timing methods

D1. With respect to different timing methods (real time attack (RTA), power-on timing, timing through credits, in-game time (IGT), removing loads), do you believe there are certain methods that are universally superior to others?


The motivation for this question was to uncover any sense of the norm that respondents feel about timing methods. The three answers provided allow respondents to cover both sides of the spectrum - true ‘case by case’ where anything flies, and then also allowing respondents to say that some timing methods are inherently flawed.

For purposes of readability among those reading this report who are not intimately familiar with speedrunning, it may be helpful to discuss some of the timing methods briefly:

* Some additional commentary, RTA tends to be seen as the common norm because it, in all actuality, represents how long it takes to beat a game in real time. Meaning, on the clock. Game-time might be very accurate for a given game, but may simultaneously be completely meaningless for knowing how long a game takes to beat in terms of your time. For example, say a game only measures inputs during gameplay (no cutscenes, loading), and does so very accurately. This would allow runners to wholly compare among each other without problems. But say that same game’s game-time measure is around 25% of how long it takes to beat the game in real time. Is it meaningful to anyone (especially for marathon scheduling) to use game-time, to say “My game time is 15 minutes but it actually takes me an hour to beat this game”? Maybe- it’s a complex issue, more below.

This question was one of the main motivations for creating the survey. Although many different games & series have very unique and arbitrary definitions, such as categories definitions, glitch definitions and bug/glitch banning/allowing, timing is a universal phenomenon that must be decided upon for each game. Real time attack (RTA) is by definition using real time instead of any in-time measure, which is in most cast comparable among all games. Certainly timing differs for in-game time (IGT) per game, but the concept is universal (use whatever the game time is, and not some outside method). And mixed methods (removing loads, some cases of IGT & RTA together) have some presence too.

The sensitivity of this topic lies in how speedrunners interpret the smartest way for their game to be measured in time, and then how other communities & viewers assess that interpretation.


The “Yes” case here is certainly the most rigid interpretation of timing - to those respondents, there are methods that are inherently flawed and not worth even considering, at 17.9%. The first two answers tend to lean towards commenting that some timing methods are flawed, but the first “No” case posits that games should still be timed individually (for example, do not apply RTA timing to a game that thrives upon IGT for the best accuracy), which is at 60.7%, the clear majority. Together over 3/4ths of respondents (78.6%) believe that some timing methods are flawed in some way. Then finally, 21.4% of respondents indicate that a pure case-by-case basis must be available.

The “Yes” case implies that some respondents have a serious concern with some timing methods. Breaking down the types of comments will be helpful in understanding the motivation for the answers for all three.

Write-in & Analysis

All responses will be weighed together for this question, rather than splitting up by response. This is mostly due to common ideas among all of the answers, regardless of the multiple choice.

Some respondents reflect that timing methods don’t matter much, as long as all who play the same game use the same method (given its consistent, which is a separate issue):

The “Yes” case was designed to allow these respondents to show their strong beliefs in timing consistency and comment on flawed methods. Most of the respondents fell into two categories with this question and the “Yes” case - some believe that some timing methods should be universally applied to all games, and other respondents agree that some methods are flawed, but moreso comment on why some methods are flawed rather than emphasize any universality of timing application. Even among stronger “Yes” cases, very few respondents assert the idea that there’s a one-all-be-all approach to timing, but moreso that some timing methods conceptually make much more sense than others.

 Examples of those who believe in a more concise, universal timing method:

PJ shares his view of an older timing approach:

“I'm of the (apparently rather-outdated) opinion that the timing method should have a consistent definition. RTA timing means vastly different things to different people and different games, so not just stating your run is timed by RTA timing is essentially meaningless. I've always preferred the SDA method (character control approach) because it is at least consistent, but still not a lot of people know what it is. In general it doesn't matter too much, because a really interested runner/viewer can just re-time a video himself if he wants a different timing scheme for comparison.”

Perhaps more than any other method, SDA timing (timing from first control to last input) is subject to backlash among many community members. However, the concept of attempting to cleanly define among multiple games & series is an understandable one. Some issues lie in how character control and endings play for any particular game, and to some respondents some meaningless inputs occur later (for example, a RPG where RTA may time to the defeat of a final boss, but in SDA/other methods the timing would include any post-fight dialogue and perhaps any inputs leading up to the credits). The respondent above does point out above that as long as one can compare consistently to other players, then it doesn’t matter much.

As a quick counter to this, Salzkorn says:

“First possible input to last possible input is a method can be extremely inconsistent depending on the game. The Batman: Arkham community uses it and it means that Asylum runners have to wiggle their mouse or camera stick to determine the first frame of input. “

Which certainly shows its potential flaws (should one have to do meaningless inputs for the speedrun to be able to reflect the timing method?). Similarly, a fair number of respondents reflect upon SDA not being useful/meaningful.

The above sentiments are a great segue into the concept that perhaps timing methods wouldn’t matter if you wanted to re-time any given run after the fact with a different timing standard, but what would happen if the timing method chosen influences the way a speedrun takes place? This issue primarily revolves around IGT vs. RTA, where there may be specific strategies that only IGT may do to save IGT time.

For example, imagine there’s a cutscene that plays, and ordinarily for RTA you would skip the cutscene (naturally, you would lose time if you watched it). But if you were to let it play out, perhaps the character model moves a few pixels/location that moves you further into a favorable position than if you skipped the cutscene (e.g., Metroid Prime cutscenes). Further, depending on the game, IGT sometimes allows the player to ‘pause’ without losing IGT time, so effectively you could take an hour break and come back later (something that would certainly be frowned upon from an RTA perspective). So for IGT timing, you would watch the cutscene, lose significant RTA time, and gain IGT time. Therefore when a game becomes subject to this, it may cause disarray among those who don’t agree upon timing methods, and then the routes & speedruns are no longer comparable universally.

Some comments on this topic:

* The “Sonic” example is the classic counter argument to IGT being useless by those who prefer RTA. Most members who commonly use RTA in their communities and understand RTA well, concede that Sonic’s timing is legitimate. The way older Sonic games work is that upon finishing a stage, the player will earn a score (in points) based on how fast they completed the stage. The faster they go, the more points they earn. However, earning points (meaning, literally watching the game increment the score counter as your reward for going fast at the end of a stage) actually burns real-time, which would imply that perhaps earning a lower score/going slower might be faster real time overall. So the Sonic community ignores this problem by using the IGT per stage, which actually measures how fast you go per stage, then they add up the stages to get the final time. As discussed immediately below, this is all well-and-good except for the fact that older Sonic games only count full seconds.

IGT is only as precise as the game’s timing system allows, leading to situations where more general times can be tied and not distinguishable from each other. For example, in RTA you can technically time everything down to the precise frame, but in IGT you will be bottlenecked (e.g., older Sonic games and GoldenEye 007 gives precision only to the full second per stage). Respondents say:

Another separate issue- timing through credits has always been a sensitive subject. Long before speedruns were common in the west, they were common in the East. (This is mostly true, there certainly are western community members that have been active since mid 2000’s, but speedrunning as a whole most certainly blossomed much later, with major growth somewhere around 2012 onwards, while Japan had a fairly robust (RPG centric) community from late 2008 on). Specifically in Japan, the main timing method used is RTA, and more specifically timing from power-on to the end of the credits. Theoretically, it is an objective measure that loses no element of the entirety of a playthrough. Further, many games typically have a randomness factor, which may be present and generating random values from the instance of power-on, and the Japanese community wanted to perhaps protect against manipulation of that (for example, you might sit at the title screen for 5 minutes if it helps influence the run because the randomness is different. But resetting from console power-on eliminates that.)

The main sensitivity with timing this is much less so power-on timing (a fairly uncommon practice among the west, but still not a major difference than starting on “new game” or some equivalent), but moreso a problem with timing through credits. Some community members honor the Japanese method out of respect, some community members enjoy the credits roll or see no issue with it, and then some community members feel very strongly that it’s a waste of time and should be abolished. Thoughts:

* Referring to timing through SMRPG credits, where “the D” is the end of the credits roll, and the end of commonly accepted timing

Another issue with timing has to do with hardware. Consoles in the ~late 90s onwards began to have different specifications that influenced loading times. For example, PlayStation 2 has a plethora of different models, all of which have different loading times (and sometimes different for individual games). Later consoles, and especially for any PC, allow you to have specifications upgrades such as SSDs (solid state disks, which are computer storage drives that allow you to load & access data faster compared to older and more standard hard disks). Neerrm says:

“This is a huge hot button topic in speedrunning right now, especially with the advent of consoles with SSDs vs not, improving hardware causing faster loading times, and other such shenanigans. Should hardware advantage really be a factor in determining who the "best speedrunner" is? It's a tough call.“

Speedrunners need to be compared on a consistent basis among their own games to be able to actually reasonably be competitive in their times. There comes a point with the above sentiment that yes, you may be able to submit a run then specify all of their computer/console specs, but at some point the lines become very blurred as to what constitutes the fastest run in terms of execution and playing the game vs. the hardware setup you have, comparatively to others.

Finally, many respondents simply reflect that a community should make the decision together, and do their best to vet out the opposition, because each game & series may have incredibly specific attributes that cannot be generalized.


Coming up with a take-away for this question is very difficult. There really isn’t a common thread among many of these answers - some respondents feel very strongly that certain timing methods are good and others are bad. Based on pure observation (somewhat backed by these responses), the community tends to overall prefer RTA over IGT, but the proponents of IGT strongly back it. My guess is that IGT supporters have significant hands-on experience with some games they run that use (or need) IGT timing to be accurate, and then rely upon that perspective for their overall view, whereas many RTA runners may not be able to fully understand nor appreciate the value of IGT (or conceptually disagree with some aspects about IGT, such as pausing) and much moreso stick to the realistic practicality of RTA (timing how long it takes you to beat a game).

A fair amount of respondents do reflect on a general “to each game their own” approach, but still feel that one method (typically RTA) should be the norm and exceptions for specific games (typically IGT) should be made. There are many ideas among the responses here- some of them will be clarified in the next section with discussion of specific timing examples that may help provide context.

As the analyst of these responses, the take-away I would posit would be for communities to take a hard look at how they time their games, strip away any past methods that they’ve done out of either respect or historical stake, and decide collectively if how they time their game makes sense and is as least arbitrary as possible. Getting community members together to discuss can be difficult, but it certainly can be done (look for Pokemon Gen I timing, as a semi-recent example.) *

* Pokemon Gen I refers to how the Pokemon community came together after multiple years of using IGT as their comparative basis, and decided to switch to RTA. Hundreds of times had to be converted for comparability, but the community felt the switch was strong enough.

D2. Do you believe any games or series are inappropriately timed for speedruns, whether it is your own community or another's, and should consider an alternative?


This section was designed to allow respondents to provide some clarity for the above question through a specific example. For the survey and this analysis, it will be helpful to briefly break apart some of these cases to understand some of the issue games, and why they’re commonly cited as problems.


This data set is highly dependent on the sampling, and is not particularly useful for an overall community assessment. It helps paint the picture through the write-in responses that a fair number of respondents who answered “Yes” continued on to provide a write-in section to clarify their thoughts.

Write-In & Analysis

Naturally, listing a game as part of this section is a ‘call-out’ of sorts to a community, but the intent isn’t at all to stir drama, it’s to develop an understanding and cause the readers of this analysis to think about the issues at hand. As a reminder, the analyst of this survey does not intend to provide bias in any of the below sentiments, and the goal is to have the reader simply understand the complications of timing methods.

Commonly cited games among respondents include:

Let’s take a look at Metroid Prime (MP) as the primary example, because it is commonly recited example for this question’s responses among non-MP community members. MP is a very classic speedrun- there’s evidence of runners performing MP speedruns long before speedrunning in its current format was a commonly known practice. This was before RTA was considered the norm, long before any sort of standard for timing was developed. As a reminder, MP is chosen here not at all to pick apart the community’s decisions, but moreso to understand and discuss the complexity of the issues.

A few things about MP and similar IGT timing games:

Although the IGT specific strategy problem is a concern for comparability among different types of MP runs, the timing being rounded to the whole minute appears to be the most cited problematic aspect of MP speedruns.

To illustrate the concept for those who may not clearly understand*:

* This example is abstract in nature

Under RTA timing, Player A is the undisputed best. And even if there were instances in which someone else got a time of 42 minutes and 12 seconds, those runs could be timed down to the frame (1/60th of a second) to decide if it was truly a tie or not. But with IGT, Player A and Player B are tied because the game only tells the player to the minute how they performed, even though there’s a 37 second difference in real time.

One might say- okay, well in cases of an IGT tie of 42 minutes, then why don’t you use RTA timing as backup? You can’t do that, because of the whole issue #1 listed above - RTA and IGT can’t be comparable in this case. Why?- because routes among two similar runs may or may not reflect similar comparable gameplay (i.e., RTA vs. IGT specific strategies)*. Meaning, one player may be penalized for their playstyle because they were prioritizing IGT strategies but being compared by RTA standards (or vice versa). So you’re left with a one-or-the-other issue for timing, so IGT may be chosen (as in the case of MP) and therefore that 42 minute time is simply a tie. Even if Player A might be the better player. The only time the threshold would be pushed would be if Player A would somehow squeeze out ~12.1 seconds of time save to get into the 41 minute range, which they’d be temporarily the best until the same issue happened again.

*As the analyst, I’ve never seen a game have its specific route be part of a timing comparison method to verify that RTA and IGT strats were the same among two comparable runs, and therefore were comparable by RTA when IGT was tied. As a reader, if you know of an example of this situation, I’d be interested in hearing about it.

Why would a community willingly opt to have a timing method that will group players by the entire minute? Perhaps it has to do with historical respect of what their predecessors have done. Some of it may have to do with the community fully accepting the ideology of having tied times, and the concept of pushing past a huge barrier such as a full minute is appealing (the same concept exists for Goldeneye 007 speedrunners of individual levels, though their times are measured to the second.) But most importantly, likely there are very game-specific reasons that the community members defend for their choice of timing.

As the analyst of this survey & its responses, I did not observe many MP-pro responses. Certainly their community must have their reasons for keeping their system alive, and it would be unfair to posit one side over the other (the above discussion only lays out the abstract about how the timing works). But undoubtedly, MP is a very commonly cited example of timing method criticism from an outsider perspective, among these survey responses and from my experience in discussions.

To discuss some other brief examples for popular games:

Here, there appears to not yet be a solution to the issues, which consist of some combination of load inconsistencies (even among the same console), version differences, console differences, peripheral usage. Among responses for this question and others, some respondents show dismay regarding the concept of having to specifically go out and purchase specific versions or setups to gain time & be competitive. Yet, it is an unfortunate reality when speedruns are naturally compared by time, which in most cases is an objective measure. One could qualify their console/version setup with notes, but time is ultimately the comparable measure.


The take-aways from the above are that communities may have complications for implementing changes to their time methods, and communities may also feel strongly about keeping their timing methods regardless of what outsider inputs are. Timing method is a complicated beast per game/series, and even if one were to posit either universal timing or strict case-by-case bases, the fact remains that there are certainly differences in opinion among the whole community. Most game/series specific communities have some sort of reasoning for their methods, and the general community should, in my opinion, do their best to understand those reasons.

E. Hardware differences / conversions

E1. How strongly do you feel a game should be played on its original console with original hardware only in order to be considered a competitive speedrun?


Another strong incentive for the creation of this survey, this question tests the time-old question of what constitutes a valid platform to play & compete on. To give some background, SpeedDemosArchive was one of the first ‘validators’ of speedruns, and had very strict console-only requirements (and an intense verification system). As time went on, and especially as streaming became more prevalent, the need for SDA’s verification process to have a run commonly accepted in the speedrun community gradually diminished (and the SDA verification arm for modern speedruns and leaderboards is not seen as a priority). Both runners & community members (viewers, supports, etc.) essentially stopped caring about such rigid methods, and as long as a run had video proof, simple public verification was enough.

However, just because a runner no longer needs to send in their VHS tapes to be verified any longer, there still is some disarray in what community members consider ‘legit’. We are in an age of ‘accurate’ emulators as an alternative to original consoles, and many options to customize one’s console & PC setup to tweak for the preferred playstyle. And as echoing an earlier discussion, there’s no authoritative governance over these types of issues in speedrunning, so the problem is largely dealt with in a passive manner, by community, case-by-case. The goal of this question is to understand how respondents view this problem- if they think that original consoles & hardware is the gold standard and should be sought out by nature, or if less stringent standards are more acceptable for the sake of availability & competition.

Further concerns may be present, such as the choice of community chosen competitive version based on its merits vs. its origins (is it the original? Or is it the VC version which loads faster?), and touching upon topics of cheating / verifiability.

One note: This is the analyst’s interpretation of the word ‘official’, but my understanding is that ‘official’ relates to any hardware or peripheral that was released by a standard licensing/developer/software company with the accompanying console/system. For example, anything Nintendo has released and was available for standard purchase (in stores, online) is ‘official’, whereas 3rd party controllers that perhaps offer different advantages are not ‘official’. The lines may get blurry in some contexts (especially with PCs), but the concept should be fairly understandable.


For purposes of clarity, this section defines the following cases in line with the answers:

Starting with the most rigid Case 1, 10.6% of respondents have the strictest interpretation. Together with the next answer reaches just over 50% of the answers, where 39.7% believe in further extension of Case 1, except allowing for essentially anything in the ‘official’ world. Case 2 fully encapsulates the Case 1, so take note of how many community members strongly believe that even VC/official ports beyond the very original version are not as good of a standard (10.6%).

The remaining 49.7%, roughly half, of respondents essentially have no rigid standard, or Case 3. Certainly among this category there may be varying levels of rigidness, but at its core, these respondents don’t have nearly the strictness the other two answers do, and rather primarily are concerned with the availability of a competitive landscape for those who may not be able to have an original console/’official’ setup.

It would be wrong to say that Cases 1 & 2 are not concerned with a competitive landscape in comparison to the way Case 3 is phrased. In many ways, they may be particularly more concerned with the competitive landscape, and wish to preserve it by following more rigid rules. It’s all interpretation of how accuracy and comparability intersects with reasonableness and competitiveness. In any case, reviewing the write-in’s below will help demonstrate this balance.

Write-in & Analysis

This analysis will break down major ideas rather than a split by Case choice. Similar to other questions, major ideas are communicated among all responses, but some respondents’ interpretations cause them to lean one way or the other with respect to which Case they chose.

Two respondents provide some context to what this whole question is about:

The majority of these questions specific to setups, cheating & motivation have to do with the competitive aspect of speedrunning (i.e., objectively measuring performance among different players & their runs). It’s common sentiment that community members don’t see emulators as an inherently bad thing- they can be fantastic practice tools, and perhaps acceptable for competitive times, depending on a community’s decision. But this survey question is designed around the more objective, comparative aspects of measuring speedruns, for which these reflections help serve as a reminder.

Many respondents reflect upon the ‘fastest official release’ being the gold standard:

Simply, if a game was released with faster completion options (version differences, ports, etc.), then use that (discussed a bit more later).

Some respondents wrote about how emulators and any setup is fine, as long as the lines of performance are not blurred:

Respondents reflect in these cases that they hope to see emulators separated into different categories, if runners choose to use them:

Some responses comment on accurate emulators being allowed:

This begs the question of what an accurate emulator is, and unfortunately, even the most seasoned speedrunners typically can’t answer. In my experience as the analyst, few community members are involved enough to know very specific console vs. emulator differences, due to the nature of emulator creation & measurability being an incredibly specific programming-related concern that few community members are qualified to comment on. For example, it’s commonly accepted that Bizhawk (a modern SNES emulator, which to my understanding, has become the mainstay gold standard for tool-assisted speedrunning (TASing) and used by many speedrunners) is an accurate emulator. As the reader of this analysis, what does ‘accurate’ mean to you? There’s a decent chance that you & your peers have difficulties fully understanding what ‘accurate’ means, and hope to rely upon the general notion that Bizhawk is indeed accurate from its inception. Sure, you might say that it generally plays the same, but are you specifically aware that a loading screen takes exactly X number of frames on both console and emulator? Or that in the case of lag frames in a laggy section, there’s exactly Y number of lag frames on both console and emulator? The point of this discussion is that there’s no governing body, there’s no authority on this kind of issue, no one’s defining ‘accurate’ with respect to speedrunning. Community members commonly accept X/Y/Z because they have to rely upon others’ knowledge- and speedrunners will invariably keep using emulators in the current age (that’s likely not changing). Some thought should be given as a community member to how one interprets these kinds of issues and flesh out the distinguishing characteristics of what an accurate emulator means to them & their game, if emulators are a significant part of their community.

On the emulator front, typically emulators & some PC games allow for control remapping. By nature of using an emulator for many consoles, you have to use a non-official controller, whether that's a USB controller, mouse & keyboard, or some other peripheral (e.g., in order to play NES on emulator, you must either use a NES controller with a USB adapter, or use a completely different USB-enabled controller). This topic will be covered more in the next question, but the idea persists that controlling for these variables in an emulator environment is a challenge.

Sometimes emulator is the only option- accessibility is a concern for some games that have major barriers for capturing video (or even just to play):

Some respondents tend to focus on the accessibility of speedrunning, and generally prefer to not have community members be gated or blocked by original console accessibility (typically a Case 3 response):

Finally, let’s look at some Case 1, 2 and 3 responses that help draw the bigger picture:

Kirua gives a lengthy explanation that mirrors what many of the Case 1 respondents communicate:

“For console games, my stance is that the original release should always be the norm for the speed, lag, possible glitches etc. Then, "any" emulator (deemed accurate enough, VC included) is fine AS LONG AS THEY ARE NOT FASTER. For example, playing a lagless version of a SNES game on an emulator should never be mixed with the console competition. Playing on VC because it's faster or allows additional glitches is terrible. People can obviously play whatever they want but communities should have standards for LBs and all at least. Not having the original version as the norm also brings an additional problem which is that any broken rerelease of the game would make previous versions outdated. For example, if there's a new "faster" OoT released on the switch, players would have to go to that. The SM64 community got it right on the other hand and they care primarily about the N64 version.”

Most of the Case 1 responses (which appear to be a rigid ‘official’ interpretation) are actually more rooted in the motivation for strictness of competitive platforms & comparisons, and communities choosing properly to segregate or accommodate for version differences. Case 1 respondents do typically frown upon communities & newer ports (of the same game, so like N64 SM64 vs. VC SM64, and not like Twilight Princess vs. Twilight Princess HD) capitalizing upon release-specific changes/glitches/time-saves as the basis of their preferred version, rather than sticking to the original release.

Many Case 1 responses are not dogmatic in nature at all- many of the write-in responses comment on how it is their personal preference. What they may wish out of other community members is to play on original hardware, but realize that sentiment is not universally applicable.

A pro-Case 2 (anything official, any port) response, Deuceler writes:

“In my opinion it should be official hardware - but it should also be separated by that hardware. SM64 is a perfect example N64 / VC / Emulator are all separate; no one gets deterred from playing but the real competition is on the original version.”

This perspective helps demonstrate that communities can sometimes easily segregate among official releases, even if it's not the original version. It may be very difficult for communities to distinguish between emulator runs consistently, but in the case of Super Mario 64 (SM64) cited here, at least N64 & VC are easily distinguishable. But the distinguishing factor here (from the above) is that there’s less of a concern that runners willingly play VC, it’s just important that VC is separate from console.

And finally a pro-Case 3 (any competitive landscape)  response from ChristosOwen helps demonstrate why accessibility is mostly maintained by proper leaderboards & community decision-making:

“So long as things are consistent then I would consider them competitive. Different game regions (JP vs US) or hardware (console vs emu) can have huge impacts on different games. Leadboards go a good job of splitting these out already based on my experience. If an original console version is fastest then I think it is fine to combine things into one leaderboard and people who player on inferior/slower versions, so long as they are aware, should not be penalised for doing so. However if an emulator version is faster (due to framerate, less lag etc) then this should not be considered an improvement on the original version and should either be banned or at minimum separated on a leaderboard.”

Most ideas among responses are not radically different, but the above stance largely lets platform separation & banning of emulators/peripherals dictate how runs are controlled.


Many respondents across all Cases do feel that each game, its hardware/software & its community is a case-by-case basis, but there are certainly respondents who don’t, which is the motivation for this question. Some responses for strict ‘original’ release (Case 1) simply state it’s their preference, and most other responses have a sense of attempting to bridge the middle ground and lean one way or the other.

Much of the sensitivities around not using original hardware do not appear to stem from accessibility concerns (i.e., most respondents state they don’t care if community members use emulators if they cannot use original hardware), but moreso whether or not those emulators & setups (controllers/peripherals) are verifiable and accurate, which is a challenge (as discussed above).

The take-away is that in absence of an authority that dictates run verification & decision making, the specific game communities are individually responsible (no surprise).

The challenge comes when communities/individuals strongly disagree with how other communities handle a decision. Yes, as echoed by many respondents, this is primarily a case-by-case issue, but the larger point is that even with case-by-case treatment, there’s certainly disagreement among such treatments fundamentally (evidenced by the general Case 1/2/3 split) that have yet to be bridged.

E2. In cases where the ruleset/standard does not specify, do you think using controller adapters (convert one controller to another console's format, i.e. PS2 > SNES) or modified/custom controllers constitutes a red flag/necessary qualification to the run?


This question explores one example of a grey area in speedrunning - the concept of how speedrunners interpret regulation over controllers. There are many grey area topics that this survey could have dove into, but choosing one and viewing the types of responses is helpful for understanding the bigger picture of why some topics in the speedrunning realm are not clear cut.

As the analyst and a runner who uses a controller converter for a specific game due to (almost) necessity, but also someone who typically believes a standard unmodified controller should be used in most situations, this question was somewhat personal.

Some respondents did not appear to understand what ‘qualification’ means. Qualification means to essentially asterisk the run with a note somewhere in the comments/description that something abnormal exists about the run. Imagine a leaderboard with only console runs (by nature) with 100 runs exist, but one of them is on emulator. To qualify that run would be to mark the submission with a * (asterisk) somewhere, denoting that it’s an emulator run. That’s all the concept means, so for this question, it means to specifically call out that someone used a modified controller.

A concept to consider is not only if runners use a modified controller, but further if they willingly communicate so, which is somewhat of a separate integrity problem. Still, write-in responses perhaps will discuss.


Roughly only 1 out of 5 respondents believe that runs with modified controller setups should be treated/viewed differently is somewhat notable, whereas the rest of respondents either don’t view it as a problem or have undecided feelings.

Some respondents struggled with the direction of this question, which perhaps should have been made clearer.

Write-in & Analysis

Before analyzing responses, a few discussion points about what constitutes a different controller may be helpful. Some controllers/controller setups can/offer:

Each of these are reflected in some way by some of the responses, and generally any run that uses one of the above in a manner that allows for unique and specific comparative advantage is not considered fair.

Many (if not most) of the write-in responses reinforce the idea of no clear advantages:

Some others talk about how an alternate controller is perfectly fine, given it doesn’t allow any inputs or abilities the original could not:

A few respondents highlight enforcement issues:

Many responses discuss turbo being available would cause a major problem. There certainly are game communities (rare in the west, common in the east) that allow turbo controllers (mostly JRPGs). However, it is very rarely, if ever, the case where the blending of two worlds is allowed. Based on these responses, the idea of using a turbo controller for an advantage over a standard non-turbo run is heavily frowned upon (if not blatantly cheating):


Not only are these very specific, grey area concerns related to controller setups, but these responses highlight the self-policing aspect of speedrunning. Many of the responses state how most don’t feel like X is a major issue, but Y would be an issue (Example 1, using a converter isn’t a problem, but using a converter to remap to a more comfortable button layout is unfair. Example 2, using custom controllers for emulators is fine because one has to do so to play in the first place on a PC, but using custom controllers for console setups is unnecessary and unfair).

The problem lies in how community members could possibly know what others’ runs are using, and what say they have to actually influence any given set of runners to change their methods. If the majority of the community agrees that a modified controller should still generally mirror the intended design of the original controller, what can any individual do to possibly understand every individual run’s setup? Do community members even care about identifying controller schemas unless something was exposed in a run? All of this leads into this bordering of accessibility vs. advantages standoff, and (as discussed in a later part of this report) may promulgate cases of cheating or unfair advantage capitalization.

Unsurprisingly the take-away here is that respondents feel that clear, distinct advantages from a runner using a setup that another runner may not have/know about it not are unfair and should not be allowed. But respondents also show signs of murkiness in decision making when fringe cases come up. It’s one thing to say “case-by-case” basis, but it’s another thing when there’s still really no clear cut answer in the first place even on a case-by-case basis. Community members must try to put their heads together to figure out what’s reasonable and what’s not.

F. Leaderboards

F1. In cases where other major leaderboards exist outside of speedrun.com (SRC), do you think administrators on both SRC and the other leaderboards have an obligation to integrate to SRC?


This is a big question. We’ll have to discuss some things about SRC, because it is clearly a both an exalted resource & point of contention.

Before we discuss the next set of questions, which are questions about leaderboards, it should be noted that there is no intended bias, stance or agenda in favor of or against speedrun.com. Simply, speedrun.com is a major player in speedrunning, and is a prevalent part of most leaderboard discussions. Speedrun.com will be referenced often in the next few questions because of its prevalence.

Speedrun.com (SRC) is a universal leaderboards site that houses individual pages for any game, given a community decides to make a page for it (an important distinction covered later). Any game can have a page created for it, with a leaderboards, resources, forum (specific to the game), and other information sections nicely laid out on the navigation panel. It’s designed to be a one-stop-shop for all your needs related to one specific game.

Long before SRC became a mainstay (~2014*), the idea of a universal leaderboard site was a holy grail. There was no central repository in any form. Either communities had to form their own specific specific leaderboard pages (e.g., MMRTA (Mega Man), deertier (Super Metroid), ZSR (Zelda)), had to maintain their leaderboards on a wiki-style site (e.g., Donkey Kong Country 2), or simply the leaderboard content existed in the form of public knowledge or forum posts. If anyone wanted to know the fastest time for any given game, unless it was a popular game/series, likely you couldn’t find an accurate answer without asking around or knowing the right resources. Community members thought about a central repository and its benefits, but were turned off for a long time because many knew the arduous process of verification & difficulties of universal oversight.

* SRC doesn’t list its inception date, but April 2014 was the oldest forum post I could identify.

Still, after enough time, SRC was created, and became that central repository. To this day, it’s uncontested (no other speedrun aggregators exist of any anywhere near comparable magnitude). SRC’s approach was to leave communities responsible for maintaining community pages & resources (as if each game/series had its own individual leaderboard set of rules & style that collectively build up to SRC’s database), and SRC would be responsible for providing the back-end to aggregate the individual pages. Theoretically community issues are self-contained and would not reflect upon the greater website.

SRC grew quickly over time. With the growth came some decision making on the overall hierarchy of moderation and design that has polarized some community members. In short, many community members feel strongly SRC is a great resource, while many community members feel strongly it is a detrimental reflection on how speedruns should be organized and viewed (evidenced by responses herein). Many external leaderboards maintain specifics to their game/series that SRC may not be able to offer, on top of other factors (well-established, verification processes, historic sentimental value.) Many of the issues with both SRC itself and its integration of other leaderboards will be covered in the write-in responses to the next few questions.

Before we get into the data section, let’s review the question. This asks whether or not major leaderboards that currently sit outside of SRC should be obligated to be maintained similarly within SRC. The word choice ‘obligated’ is a very strong one, and is purposefully chosen here. If a respondent thinks ‘not obligated, but would be nice’, they still (should have) answered “No.” The “Yes” case is a strong belief that those leaderboards should be fully integrated into SRC.


It’s glorious, the most pure pie graph I’ve ever seen.

1 in 3 respondents believe that SRC should effectively merge and sync with any external leaderboards. Keeping SRC up to date and fully in sync with other leaderboards is no easy task, but these respondents believe that the benefits of the central repository are that strong.

Write-in & Analysis

Because of the strong difference of Yes/No cases, they will be separated here:

“Yes” cases:

Many respondents chime in that they feel it’s simply an ease of access benefit, and thus consolidation of the speedrun community is both helpful and convenient with its leaderboards:

Some respondents believe it is on SRC to be the driving integration force to other major leaderboards, if their goal is to become that sought-after central repository:

Respondents comment on the messiness of maintaining two leaderboards. Although the above Premise discusses how sometimes external leaderboards exist, there are instances in which SRC reflects some but not all of the proper leaderboard entries of another site. For example, ZSR may have the latest Zelda runs for a given game (its leaderboard is accurate), but perhaps due to lack of attention from their community, SRC’s page may not be as up-to-date and accurate, but still may exist. Therefore any person who goes to SRC instead of ZSR may type in any given Zelda game, and actually WILL find a page there, but then accesses out-of-date information. It’d be one thing if the game page didn’t exist and the searcher was told to go to ZSR instead, but in cases like this where its muddied what information is up to date, respondents reflect:

“No” cases:

A large number of “No” responses comment that they believe it would be in the best interest of SRC and external leaderboards to integrate, but have no obligation to do so. Many of these types of responses discuss that individual communities likely have their reasons for not integrating:

Some responses highlight that having a central speedrun information hub is helpful and important, but less-so that the leaderboards function itself is centralized. With the above examples of external leaderboards, this would functionally be a redirect to an external site. (For example, the Donkey Kong Country (DKC) series has no runs on SRC. If you search DKC, you will find a page on SRC (meaning they acknowledge the game exists), but it will simply tell you to go to their specific website instead.)

Fundamentally some respondents disagree with the above, too:

Many “No” responses comment on the burden of how much effort full integration would be, given an automatic API or data-feed option isn’t available (most likely the case for many websites as of this report):

Ppotdot1 adequately summarizes why external leaderboards have benefits that SRC may not, and therefore should not be forced to integrate:

“An outside, well maintained and functional leaderboard should have no obligation to integrate into SRC just for integration's sake. I think that SRC is a good site for generally keeping speedrun information, and is an extremely helpful tool for smaller communites who don't know how to make a separate leaderboard and who also would like to have some recognition from the greater speedrunning community. However, outside well developed major leaderboards generally have more resources, more users who care about those boards, and more features on those boards than SRC does currently, and therefore should not be obligated or pressured into joining SRC just becuase its the most well known site.”


Naturally, SRC being the central repository that may or may not eventually integrate other leaderboards, much leaderboard decision-making and power to SRC management. Again, the idea is that community problems are self-contained, but when the site’s design and moderation hierarchy (discussed later) has limitations, even individual communities are at the mercy of SRC’s staff and administrators’ decision making.

There’s a fair amount of contention among “No” responses. Some believe it’d be nice to integrate. Some do not approve of SRC’s leadership and organization style. Some believe external leaderboards are worthwhile and valuable, whereas others think although no obligation exists to merge and sync them, that external leaderboards should be deprecated.

Leaderboards may seem like a trivial thing to an outsider of speedrunning (it’s just a list of times, right?), but the fact of the matter is that the greater community went from being largely disorganized and spread among many websites, to fairly quickly being (mostly) aggregated into one major website, and with that aggregation comes complications about how specific communities feel about their timing methods and organization style.

The take-aways here are that most respondents seem to support the idea of integration, but have varying levels of force behind their beliefs. As the analyst of this report, my opinion is that being on the sidelines in the “it would be nice but whatever” bucket means that real changes one way or the other won’t ever occur, and both speedrun community members and outsiders (looking to join, or looking for information) will still be burdened with the same problems listed above. Areas of contention won’t change unless communities take a look at what their frustrations are, and adequately summarize and communicate them to other parties, then seek a solution.

F2. Should SRC be the only central repository for leaderboards? If no, please elaborate on any reasoning and/or briefly discuss functionalities of other leaderboard sites


This question is a more theoretical/concept driven version of the previous question. The previous question is a practical one: should SRC and other leaderboards integrate and work together when they’re separate? This question asks: should SRC be the end-all-be-all for leaderboards where communities work with SRC admins to get the leaderboards to function like they want & need for their games, or is it perfectly acceptable to have other leaderboards exist outside of SRC?

In many ways, SRC is like a general department store with “all your needs”, and other leaderboards are specific and unique mom-and-pop stores. Chances are a SRC page could probably get you by as a community, but in some cases it can’t, and specific outside leaderboards have their specific and unique benefits. The caveat to this analogy is that there really isn’t competition per-se, SRC and other leaderboards have typically been hands off of each other and don’t threaten or interact with each other much. However, a problem comes from the demand side, where both new users and community members deal with a muddiness of ‘which store to go to’. For example, if one wants to look up a Donkey Kong Country time, what’s their approach as a newcomer? Should they say “Oh yeah, SRC should have a page, let’s check there”, or should one say “Let me look up the specific leaderboard page for DKC, if it’s on SRC, go there, if it’s elsewhere, find it somehow”? The ‘competition’ stems when those two worlds collide, and there’s mismatch of information at the ‘stores’, and end users may wind up confused or with misinformation.

Side note: Strict interpretation of this question is that the question is asking about if other websites should rival SRC as a central repository (meaning, rivaling what SRC currently does). Responses to both the above and this will be reviewed (i.e., should there be other leaderboards for specific communities outside of SRC | OR | should there be other leaderboards that rival SRC’s current functions).

This question will help explore why respondents think SRC should conceptually be the only place to go, or why outside leaderboards are perfectly acceptable (among other concepts).


45%/55% : Yes/No should simply be an indication that respondents are divided on the topic. There are some responses that indicate “Undecided” or “Don’t care”. Also as with most questions, depending on how strict the question was interpreted, perhaps some respondents may swing one way or the other but likely would keep the ratio here similar.

Write-in & Analysis

Because these are strong opposites, Yes and No cases will be segregated:

“Yes” cases:

Many respondents comment that ease of accessibility of one central repository is important:

Some comment that SRC has grown large enough that its influence cannot be superseded any longer:

Some respondents put the onus on the community to work with SRC admin/staff to fix their problems, rather than maintain a separate leaderboard:

And respondents who comment on whether or not other sites should exist to do what SRC currently does (as a central repository) typically say that SRC has established itself as the central repository, and others need not exist:

Most Yes cases highlight the overall benefits of ease of information for end users, and posit focusing on helping SRC grow and improve rather than turn to maintaining duplicative or separate efforts. Some users do feel here that separate boards have their benefits, but the majority of these responses posit that SRC can change to accommodate specific niche requirements, given their communities agree and want to do so.

“No” cases:

It should be noted that “No” cases had many more write-in responses than “Yes” cases (293 “No”, 76 “Yes”, where the overall No/Yes split was fairly even).

Many respondents conceptually don’t believe SRC should be a requirement for communities to adhere to:

Some respondents don’t trust the landscape of SRC to be able to freely accommodate the greater speedrun community:

A few respondents discuss how other leaderboards have very specific functions that SRC (in its current form) can’t accommodate:

Some respondents generally do not agree with SRC’s leadership and structural organization (not highlighted here). Although this may be a point of contention, it is worthwhile to think about drawbacks of having a central repository being just that, central, and therefore decision making and power to make changes and influence the community ultimately lies in a few members’ hands.

Connecting the above, many respondents highlight their concerns with the idea of a “monopoly”, where only one website/governing force has all the power:

Garrison brings up a specific point that comments about how there’s somewhat of a split between specific community members (speedrunners & those involved with speedrun specifics) and viewers:

“SRC strikes me as something that is more appealing to outsiders while individual leaderboards are almost a celebration within their respective communities.”


Finally, Chlorate offers a lengthy response that fairly well summarizes the benefits / challenges of a central repository such as SRC (and some SRC specific commentary):

“On the fence about this one. I think ideally it should be (it would be less headaches if it was), but SRC came too late when major communities had already established their own leaderboards. Why SRC is good: 1) less effort to set up leaderboards for any game, 2) one, go-to site for everything. Why SRC is bad: 1) all eggs in one basket: does SRC have a good backup process? what happens if a bug/hack causes data loss? 2) community-specific features that SRC doesn't have (version time conversions come to mind), 3) lack of updates (changelog has no updates this year), 4) not open source; if a community has an idea for a new feature, they have to convince SRC to build it rather than building it themselves. There's also nothing stopping anyone from building a site that competes with SRC (this is the internet which is decentralized).”

Certainly there are many factors at play. But given the responses to this question, most respondents feel fairly strongly one way or the other.


There are some very big concepts at play here. Should a community grow to revolve around a central database and organization medium, adhere to its methodology while also investing resources to help it grow? Or should there be a free-form approach to speedrunning, where communities can do what they individually please? And how does that all factor into accessibility of information, reliability of information, and the overall connectedness of what the greater speedrun community is doing?

This is perhaps one of the most divisive topics in the survey, because each respondents’ specific ideas and approach to any small leaderboard/community related topic rolls up into this large amalgamation of ideas. However, based on observation, many individual communities tend to agree with each other internally (e.g., most Mega Man runners agree that their own leaderboards page, MMRTA, is a fine resource that caters to their needs while SRC may not accommodate well enough).

The take-away is, given that it's important to someone (some community members simply don’t care, and that’s fine), to consider how one thinks speedrunning should be conceptually organized (more/less governance, centralized/decentralized), and to understand that many other community members feel strongly about an array of concerns. Changes will only happen when community members either individually or collectively communicate their concerns and push for changes & adaptation. The concept of one leaderboard may seem to some like a trivial part of the greater picture, but it is a major point of contention and is one of the few unifying resources that exists as a common landing point for both outsiders and community members.

F3. How should moderators be decided for a game's leaderboard on SRC, and what qualities do you think are important (or necessary) for a good leaderboard moderation team? (check multiple)


Moderation and decision-making in the speedrun community realm has always been a major grey area. Who’s qualified to judge a run? Someone who knows the game well? How can one possibly gauge that? And who’s qualified to make decisions about categories & leaderboard treatment? How can the general community for any given game know moderators understand both bigger picture and specific detail concepts to be able to perform as a moderator?

Further, who should be involved, based on experience and involvement? This question aims to see if a moderator’s affiliations and interests fundamentally coincide with what community members expect out of their game’s moderation staff. Do communities think specifically that runs need to be maintained by active runners, or does that not matter at all, and should the moderators be “analytic TAS tools savvy driven warriors” with thorough understandings of how speedruns work, regardless of their affiliation? And what’s actually feasible - do non-runners even care about this stuff?

There are many concepts at play with this type of question. This question focuses on moderators given the leaderboard context of these questions, but similar concepts can be applied to important decision-making (Next time you, the reader,  hear someone say “the community decided”, take a step back and think what that means? Who? Who decided X, Y, or Z?)


Let’s review the possible answers:

The first two are the most involved, day-to-day members of a community. Runners clearly have experience with a given game, and community researchers may not actively run but offer insights that runners may not. Sometimes, runners don’t really care about the routing & “speed-science” of a given game, they just want to play & execute. And sometimes community researchers have no interest in competition and playing, but are much more intrigued by the routing & mechanical time-saving aspects of a speedrun. Nevertheless, both are typically closely connected to the inner-workings of a game, and it’s no surprise their likelihood to get chosen for this question is very high.

The next answer, unaffiliated members, gets into the territory of how speedruns should be judged, and who should judge them. As seen in the responses below, some respondents would like to see an unbiased third-party approach to judging high level runs. Some unaffiliated members believe they have the organization and wherewithal to moderate leaderboards, and offer to engage their interests so. Whether they’re qualified or not is certainly a case-by-case basis. This answer and its considerations is by far the largest point of contention for moderation. Non-community members start having a say in leaderboard issues specific to a game, and some community members may feel strongly to discourage such a normality.

The last option is sort-of a mix of the second and third- those individuals on the outskirts of a speedrun community, who know enough about a game by watching and understanding through commentary, but aren’t specifically involved with routing or performing. They may offer insights from seeing enough runs to catch suspicious activity or adequately judge a fair run that perhaps even others who regularly play the game in their own style may not. For example, a runner who does things a specific way, only, may not know how to judge someone who does things completely differently, but someone who’s viewed many speedruns may be able to help. There isn’t a huge distinction between this response & “unaffiliated members with moderation interests”, but rather puts the focus on the stream-viewers who know a given game very well by viewing, rather than being specifically interested in moderating.

The results are not very surprising - active runners are considered the most qualified, with a minor amount of drop-off to those who are community researchers. Lukewarm response to those who are avid run watchers, and unaffiliated members are the lowest. But what is still telling is that a fair amount of respondents do think that those other than runners/routers are at least viable to be considered for moderation. Still, the discussion points will help give context as to some of the issues with moderation.

Write-in & Analysis

There are many, many ideas at play for this question. The overarching concepts will include who’s fit to moderate, who’s actually reliable and willing to moderate, and who should not be part of the equation whatsoever.

A select handful of respondents feel strongly that only someone who actually performs runs of a game is qualified to moderate:

One must have gone through the process of running a game to understand what’s important to moderate. Any other level of interest is secondary. Overall, this is not a very common view. And some respondents think differently, that runners, as those who may or may not (case specific, of course) solely learn and perform a speedrun of a game, may not be wholly qualified to moderate:

However the most common write-in response, in line with the data above, followed the first two selections - runners & researchers. Many responses are what you’d expect - community members who know the game well. Many of these responses reflect that anyone else don’t have the same wherewithal to be able to actually judge a run:

Some respondents discuss how moderation matters in size of the leaderboard, where cheating (a concept brought up in the next set of questions) may come into play:

Part of a moderator's major job is to be able to discern fake runs from real ones. Does it happen that often? Likely not. But especially for top times, it is definitely important. More on cheating later in this analysis.

Sometimes, games are so small that those who ordinarily would say the first two categories only, might make an exception to include an outsider:

Having someone independently verify a game may be better than having no one (or the runner themselves) verify their run.

Arc brings up a church-vs.-state argument for SRC’s design:

“A problem here, I think, is that SRC only has one privilaged runner role - moderator - for two roles - verifying runs and editing the leaderboard rules/layout. I think this should be separated, as there's often people that can be trusted with verifying runs but not with maintainng the leaderboard itself. I think anyone that was actively involved with a game can be its moderator, as long as they aren't shown to be ignoring the games' own community's decisions.”

Part of this bigger discussion is that moderators are often times the decision makers as well, as they control the major gate to having a speedrun legitimized. Part of the danger is the power of an outsider to cause problems, and the game specific community may have to struggle to fight back. What Arc is highlighting is that it’s one thing to allow someone to verify runs (if they’re interested and knowledgeable), but it’s a whole other issue to let someone both verify runs and control the means of how runs are organized (timing methods, categories, console segregation, etc.)

A few respondents bring up points about how a moderator is important not only for decision making, but as a community representative. :

Moderators typically serve as a major liaison for newcomers between an outsider and an insider to the community. A situation where moderators are fumbling for answers, relying upon others, or simply giving misinformation is a potentially harmful one for community members, both old and new.

The respondents who include the first two (runners & researchers), but then also specifically include avid runner/watchers, reflect the following sentiments:

Many of these responses have the same general idea as the above strict runners & researchers, but trade specific speedrun knowledge for availability of moderators and willingness to moderate.

A large part of this moderation conversation comes from motivation to moderate. Who’s motivated at the end of the day to help out with moderation? Sometimes community members genuinely love their games and their communities, and don’t find it a burden whatsoever to dive in. But others do find it cumbersome, and who would be a qualified moderator might just not bother moderating effectively due to lack of interest. And that is a common sentiment among respondents who feel that either avid run watchers or unaffiliated moderator-interested members may have a stake in moderation:

Getting into responses with the unaffiliated members with interest, some responses highlight the possible need for an objective third-party look at a speedrun (if a leaderboard was big enough to warrant so):

One last comment - anonymous says:

“Also: Moderators shouldn't just be those that first submit the game to SRC”

The fact that this is a thing that can happen at all is alarming, and is further discussed in the next question’s responses.


Moderation is a tricky problem. It is the crossroads of at least the following (not exhaustive) list:

One reading this might think, similar to many leaderboard problems, that it seems trivial- it’s just someone clicking Yes/No to a speedrun video to see if its legit, right? However, as the analyst it’s my opinion the effect of a leaderboard’s moderation staff trickles down heavily upon the community. They may serve as the gatekeepers to outside members who might want to join a community, and legitimize a community’s speedrun activity. They often have a say in community issues such as category definitions, and often their opinions are considered as a strong representation of their own community to other speedrun community members. There are of course exceptions to all of this, but in my opinion the moderation role is a powerful one, and is worth considering its treatment and power.

The takeaway here is for community members to recognize the importance of the role and consider how in any given community, how a moderator’s role influences how outside individuals join a community and how a community is maintained with respect to the verification of runs. Clearly, evident from above, respondents do not see eye to eye on who’s qualified to moderate, but the overall sentiment certainly does (at the very least) ask a moderator to be familiar and knowledgeable enough to adequately verify a run.

F4. What options as a community member would you like to see available in cases where you feel the moderator staff for a game’s leaderboard is inadequate and/or certain moderators need to be removed? (optional)


This question allows respondents to discuss any relevant frustrations or cases with respect to moderation problems. The prior question may seem hard to put into context- some of the below discussion points may help pull it together. Speedrun community members clearly have some opinions on whether or not certain members should be responsible for moderating, but do speedrun community members even have the ability to posit a remedy to something they (or their community) don’t agree with?

Write-in & Analysis

This analysis will take a few responses and talk about the ideas at play. There are many ideas that may not necessarily cohesively belong together for this question, but certainly some important considerations.

One common sentiment is to have a community come together and figure it out:

But the issue may not so much be whether or not the community can decide something, it may be more along the lines of whether or not they can reasonably turn things around on a leaderboard like SRC, or when there may very well be a large divide among their own community.

Common suggested remedies (some of which may be in use on some leaderboards) are:

One of the more commonly cited issues with this question and the previous question is what happens to a leaderboard when new games are released. Historically there’s been activity of squatting a new board as soon as a game is released.

Some reflect on what the current landscape looks like, and what would be helpful for the future:


The major overall sentiment is that it’s a case-by-case basis, where communities should get together and discuss when problems arise, then bring it up to SRC staff (for SRC problems).

But wouldn’t it be nice if institutionally this sort of thing was protected against, rather than constantly have to be reactive to things like a rogue moderator, or how when a new game is created on SRC anyone who first submits can just claim authoritative overlord status? Preventative measures may be worth thinking about related to this question. It’s very easy to write off many of these issues as case by case, but if something is causing a community a problem in the first place, can’t the process be designed better in the first place?

It’s also worth considering how SRC was designed in the first place in context of these sort of issues. Remember that SRC wasn’t designed to have a specific say in every community matter, they simply serve as a common-grounds aggregator of community pages, and the staff’s job is to keep the site up & running. But what happens when these issues start cropping up everywhere? The burden (and to some degree the decision making power) then lies on them, the staff,  to mediate the issues. Do they have the wherewithal (the manpower) or the desire to keep up with that demand going forward? It is hard to tell.

G. Cheating

G1. How much cheating (in percentage) do you think occurs in the greater speedrun community? (Enter 0 - 100)

Premise / Analysis

Before we jump into some deeper discussion, let’s talk at a bird’s eye view about cheating in speedrunning.

Cheating has no definition, but there a few common avenues for which a speedrunner can cheat their runs:

The above are some examples, certainly others may exist. The idea is that anything a player does to give themselves an unfair advantage over someone else is considered cheating. Sometimes the ‘unfair’ distinction is hard to define or even identify (like a player using a different controller than everyone else), but cases like splicing or editing a ROM file are typically seen as inexcusable and blatant attempts at cheating.

There have been a few examples of notorious cheating in the speedrun community, which typically open the greater community’s eyes that this is something that is real and can affect the community. It’s the topic typically ignored or swept under the rug until it rears its ugly head, then a given community may undergo a reality check. Unfortunately there aren’t many resources in terms of historical records of cheated runs, but it is certainly a fascinating and alarming phenomenon.

The next set of questions were designed to poll respondents on how much cheating they think occurs, then how they think cheaters should be reprimanded, and then perhaps some pertinent examples.

The question here itself is straightforward. Although this question was intended to gauge respondent’s views on an overall level of falsified runs in the speedrun ‘market’, not only was the form designed poorly for data validation (meaning non-percentage answers could be put in), many of the responses are ludicrous (e.g. claiming anything over 10% of all runs are cheated is somewhat absurd, in my opinion), and therefore the overall statistic is meaningless. Further, it’s tough to give a baseline at all for what community members should reasonably expect (who knows, really?). Therefore the data section of this report will be ignored, and any insights will be included in the next question about reprimanding cheaters. Most write-ins here simply say “idk, hard to tell”, which is not unexpected. It is suggested if the reader would like a snapshot of what respondents think in terms of their listed percentages, the data source can provide.

G2. What do you think is an appropriate approach to found cases of cheating?


This question is a much more stringent ask on how cheating should be handled. Yes, we’re all mostly aware it happens, in varying degrees that is at times hard to tell. But how should we treat these cases? The question gauges the respondent’s sensitivity to these matters in terms of strictness and forgiveness.

Something to keep in mind is that this is all on the internet, and even if you ban someone under one name, they may come back as another. However in my experience as the analyst, this is somewhat rare, and those who are ousted by the community/leave on their own will on bad terms are typically gone forever (or do a really good job hiding their original identity).

In reviewing the responses, and conceptually, most cheating cases are case-by-case - there’s specific facts to each case that contribute to what would be appropriate treatment. The point of this question is to feel out how strict/lenient community members are given the current speedrun environment.


The split here is certainly weighed against those who think offenders should be banned forever (~12%). The majority, roughly 48%, believe that anyone caught with cheating activity should have their slates wiped clean for past activity and eventually given another chance. And the most lenient case (40%) throws out the falsified run and effectively posits monitoring the runner’s activity more closely into the future.

Respondents seem to shy away from the most serious case of ‘ban forever’, but this option was specifically part of the question because, as evident here, some respondents do strongly feel that way. But together almost 60% of respondents do feel that cheaters should have their activity fully removed, with varying levels of “second chance” leniency.

Write-in & Analysis

We will look at each category and assess their points and differences.

“First offense forgiven” responses:

Many responses with this answer tend to reflect upon the hope that the community doesn’t jump to impulse decisions to ban community members unfairly. Perhaps the runner needs a reality check, or needs to be sternly told that both what they did was wrong and that their activity is being tracked for the future.

Respondents in this section highlight that sometimes runners may not know they’re cheating. In cases like splicing or using a blatant advantage, perhaps not, but sometimes runners might not be aware of something they’re doing may be considered an advantage to others (for example, using a keyboard script in a PC game where most of the community disallows):

Many respondents here say that second offense is an immediate perma-ban, when the offender specifically was told not to repeat their breach:

One respondent discusses motivation and why cheaters would cheat in the first place, and thus influences their opinion:

Some respondents reflect, similar to the idea in the premise section, that it’s unrealistic to actually “ban” someone because they can come back in some other form:

Other general notes in this section point out how all legit times shouldn’t be removed if proven so, and future runs should definitely have closer scrutiny.

RamblingJosh wraps up this section with a good overall sentiment:

“It's uncommon enough that I think it should be handled on a case by case basis. Everyone has a story, everyone has their reasons. If someone is cheating in something as benign as speedrunning, it's probably important to listen to their story.”

Cheating cases are indeed fairly rare. The responses for this question with this answer err on the side of understanding situations (not that the others don’t, just as a point of emphasis in the responses and the ideology).

“Times removed, banned for a period of time” responses:

Naturally these  responses are stricter than the previous section in concept, but many of the ideas are the same. The below will point out responses that are specific to a stricter mindset and a mandatory ban period.

Responses here highlight the non-mistake quality of a runner falsifying their run:

And some responses posit how important the entire speedrun process is on being honest about what you do:

Certain respondents reflect on how a harsher punishment will likely influence a stronger realization from the runner that their actions are wrong:

Many comments talk about younger runners being more susceptible to this kind of cheating activity, in which most responses garner a sense of teaching and awareness through punishment to help younger runners learn from their mistakes. Some respondents reflect:

A very common sentiment among these two categories is to invalidate all proven cheated runs immediately, but not touch the others (assuming they had undergone the same level of scrutiny). Depending on interpretation of the next section, ‘banned forever’ may or may not discard older runs on a leaderboard that appear legitimate, while the runner is ‘banned forever.’

“Banned forever” responses

Finally, the most strict response category. Respondents who chose this likely have a very strong opinion about cheating activity.

Respondents do make some distinctions here between what the ‘ban’ would entail, where there may be a difference between being banned strictly from submitting times to any leaderboards vs. being banned ‘from the community’ (which is certainly harder to enforce).

Many respondents have a very strict response to this. They won’t be listed here (some of them use choice expletives) but my guess as the analyst is that these cases for ‘ban forever’ are intended only for cases where the evidence clearly points to a blatant case of cheating (or, the cheater confesses). Grey area or ‘accidentally cheated because of a mistake’ are very much not the same idea, and would likely revert to treatment of one of the above options.

Respondents, similar to the above section, further emphasize trust as a mainstay of the speedrun effort:


Cheating is a sour but realistic part of the speedrun community’s musings that need to be addressed head-on. There’s certainly some varying opinions, most of which are on the fringe cases of less severe cheating and how strict the initial punishment should be (mostly for newcomers & younger speedrun community members). For the more severe cases, there are a range of opinions on how strict to handle the case. The motivations for strictness are somewhat difficult to tell, but the common sentiment is to protect what is already an arbitrary hobby at times- letting runners get away with unfair advantage may tarnish the reputation of speedrunning overall.

The take-aways here really do lie in the specific reflections of respondents- there’s a lot of varying concepts at play. There are problems with a blanket approach, but there’s also value in setting a precedent for community members to adhere to (for example, no tolerance for second time offenders across the board will deter anyone who cheated the first time from ever messing around again).

G3. What would you like to see as either safeguards or enforcement options against cheating? (optional)


Self-explanatory, this question polls if anyone has any insights as to what they think would be helpful to curb cheating activity, either in advance or as a scouting measure.

One of the subtexts of this question is how much effort the speedrun community should put into the safeguard aspect of monitoring runs. Theoretically, moderators, who both sit on leaderboards’ moderation staff and are knowledgeable enough to render an opinion, approve runs that pass a reasonable test (timer start/stopped correctly, gameplay looks valid, etc.) Though this does happen in practice, a common sentiment among responses is that moderators often don’t do an extensive enough job- some lazily accept runs that shouldn’t (timing method is wrong, emulator was used when disallowed, etc.). What are the ideas and tools that would be helpful for both runners and moderators to generally know in advance of the verification process to limit how much cheating occurs?

The subtext here is who should be responsible for the effort, and if its even worth doing so. Why would anyone want to moderate in the first place, especially if one’s primary interest in speedrunning? The idea that a gathering of community members (individuals who want to speedrun) takes on administrative/auxiliary responsibilities to bind together their hobby is what makes a community, so community members do it either out of respect for the community idea and/or decision making power. The key is keeping both runners and moderators interested and engaged, otherwise runners may not understand what moderators are policing for, and moderators may fall off policing anything in the first place- but certainly that can be difficult as community members move on from speedrunning projects (or perhaps never really cared to moderate in the first place, in which case they probably weren’t a good moderator choice to begin with).

And another concept at play is whether or not the safeguarding is actually worth the effort. Cheating occurs so infrequently, right? What’s the point of all this effort? The problem with big cases of cheating being exposed is that it comes and goes in waves- community members talk about a topic for a little while, then move on and forget about it. The unfortunate truth is that one never really knows how much cheating occurs unless they’ve specifically checked every video specifically and carefully, which of course is unrealistic. But perhaps the community could take steps to instill a more robust set of principles to disallow the activity in the first place.

Write-in & Analysis

Responses generally cover a wide array of ideas for safeguarding and monitoring run verification, which will be presented here in bullet point form, then discussed if additional color will be helpful:

Many respondents comment about how moderators should actually moderate:

As discussed briefly above, sometimes moderators don’t actually do their job- moderator may be more of a title than a functional position.

Many respondents comment that extra safeguards are not worthwhile for various reasons:

The sentiment here isn’t that the community shouldn’t be monitoring cheating activity, but moreso that a top-down community approach might not yield any specific benefit. After all, cheating is a very specific case-by-case basis, and perhaps only the most competitive times actually matter.

Some respondents talk about how tools to auto-detect splicing or other cheating activity may be part of the future:

To elaborate on Chlorate’s point, a few Super Mario 64 runs were analyzed by spectrogram (visual representation of sound waves) and found to be cheated given their audio activity. Perhaps as community members pay attention to these kind of issues, more tools may be developed to automate the process.

EvilAsh25 has some sentiments about the moderation effort:

“In the end though it becomes more and more work for the moderators of the game, so they will have to find a balance between rules, ease of use of emulators, and the amount of work that has to go into it.”

The balance aspect is very important- again, moderating and administrative effort is not the actual act of speedrunning. Some community members enjoy moderating, but some either find it a burden or don’t bother being thorough. Developing the silver lining between realistic and ideal is important.

One last sentiment:

Thanks dram.


A common theme among many of the responses is how much of this effort hinges on the efforts of the moderators. Communities should truly be tasked with finding the right moderators- community members who are specifically set up for the task. Not just members who have been around a long time, or have a good time- specifically members who have the patience, interest and knowledge/skillset to actually moderate.

And another point, simply- who cares? Who cares about cheated times? Is it really that big of a deal if some runs out there are cheated? I think the answer is a resounding ‘a lot of community members care.’ Currently, although there’s certainly money in Twitch streaming, there is not much money being thrown around for personal performance of speedruns. Motivation to improve times comes primarily from community recognition and self-satisfaction. The speedrun community hinges on community members being honest about their time in order for competition to be meaningful, and as the analyst I believe it to be the case that cases of cheating are eye-opening to many community members when they are exposed. So all of the above ideas that have been discussed about safeguarding may be cumbersome and time consuming, but realistically if no structure was in place to vet fake activity, then competition would be meaningless and uncredible.

Another brief point- many runners might speedrun for personal satisfaction rather than a super-competitive time, but naturally most runners compare their times to the fastest times to estimate how good their run is, and if the fastest activity is falsified, then the personal endeavor of what’s a “good time” may become hazy as well.

The take-away from this question and the previous cheating questions is that it’s serious- the verification process is the backbone of what makes the speedrun process meaningful in the first place. It might not be glamorous, but it’s important and communities should assess their moderation approach and think about the consequences of not doing so.

G4. Do you have any specific experiences dealing with a case of cheating that you would like to share? (optional)


Respondents who have shared a story, however elaborate, typically have an insight as to what happened, which may be helpful to briefly explore.

Although the question asked for specifics, this analysis will not highlight specific cheating cases. The question was clearly asking for these cases, but naturally there are points of contention in these cases and it would be unfair to pose one side of any argument in this report. Still, understanding the complications that come from problematic cases are helpful for consideration. If you’re interested in hearing about these cases, please refer to the data set for question G4’s responses.

Write-in & Analysis

The following list are a few examples of how cheating activity was exposed in submitted responses:

Many of these are similar to concepts already discussed in the above questions.


I encourage the reader of this analysis to review the write-in responses, many of them are interesting. But as stated briefly, all cases should be viewed with the lens that there are (at least) two sides to an argument.

Through these responses, many community members do take cheating seriously, and feel strongly about cheating cases within their community (when it hits close to home). Often, those who cheat and who are exposed will leave the community (or be very sly about reappearing under a different name). Communities can learn from past mistakes to not repeat them.

H. Community Events

H1. Tournaments have blossomed since ~2015 into a staple of community representation. What is your general impression of tournaments? Please write-in regarding specific tournament examples if possible, including tournament formats, timelines, preparation demand and general broadcast quality.


This question takes an example of a community activity that has blossomed in the last 2 years and polls respondents’ opinions on its importance and role in speedrunning.

Prior to tournaments, a few other types of competitive landscapes existed. There were the leaderboards, which was strictly a runner’s personal best representing their times, pitted against other runners’ times. There were SRL races, which were sometimes broadcast to larger audiences (e.g., speedrun showcases (such as GDQ Hotfix or Race for the Record). Similarly there were also relay races, where multiple games in a series (or sometimes not in a series and loosely related) would be raced in relay form. But runners didn’t typically have a longstanding way to continually play the same game and pit their skills amongst one another over a longer period of time (RIP SRL Season 2). Races and relays were fun, but very much one-off events that didn’t build up into something bigger. Eventually, tournaments would be the answer to that.

Tournaments began their rise in popularity somewhere around 2015* in the speedrun community as a mainstream online event. Tournaments come in some different forms but are formatted what one would expect - multiple rounds of runners doing individual races with each other, climbing a bracket and eventually yields a winner. Modern tournaments came at a time when Twitch was growing, and began having a level of production value that was not typically seen before in speedrunning. Live races would be rebroadcast to a larger audience, and typically dedicated commentary would be commonplace, for which the commentary covered both individual races as they occurred and the overall tournament landscape for any given game’s tournament.

Because tournaments have become such a major recurring event, they are both the subject of much effort from both active runners/participants and auxiliary/administrative efforts (tournament organizers, community members who rebroadcast the races from individual runners’ feeds, commentary, etc.). They too are subject to scrutiny from some community members, some of which think a different format would be more beneficial/better, and some who don’t think they’re a worthy investment at all.

*A very notable exception is SRL’s Mystery Tournament, which dates back all the way to 2012. Mystery Tournament is a series of blind races where participants play games they’ve never played before, and race to a fairly short goal for each game. Although these tournaments follow typically a similar format to modern tournaments, they were the most notable type of quasi-‘speedrun’ tournaments before regular competitive speedrun tournaments became mainstream.


Before reviewing the data: In retrospect, this answer should have included an answer for “Do not care.” Many write-in responses reflect this, and the results of this dataset would most definitely change if the option were available. Still, the choices here should reflect a general opinion on tournaments - the respondents who commented “Do not care” were in each of the three categories (fairly evenly spread among blue/red, but mostly yellow responses had this notion), meaning they did still lean one way or the other when forced to choose.

Overall respondents believe tournaments to be a worthwhile endeavor (>90%). But about 1/3 of the respondents do believe that tournaments are flawed in some way or another.

Less than 10% believe tournaments are not valuable at all. Likely, respondents believe some of the negatives listed in the above (red) answers accumulate to enough drawbacks to not make tournaments worthwhile.

Write-in & Analysis

Three opinions are fairly divided, each will be reviewed independently:

Strong support for tournaments:

Many respondents simply like the idea, and find that tournaments are another medium for competition (and don’t find that tournaments take-away from other aspects of speedrunning). There are many responses that say “Great!” or “I love them!”, as indicated by the poll data numbers.

Some respondents discuss how they are good for exposure to unrelated parties, meaning outside individuals stumble into a tournament stream and see a level of home-style organization in a speedrun medium perhaps they haven’t seen yet:

Respondents here also discuss the motivational benefits. Some runners (or would-be runners) may be undecided about learning a particular game/category, and having a big tournament with many active runners may help them get invested & start participating:

Some respondents are very excited over the growth of tournaments, and propose that they should be a bigger, more mainstream event (perhaps with more incentives):

Garrison points out how tournaments might be a good indicator about how to assess the ‘cream of the crop’ of a game, rather than strictly judge by fastest time:

“As games get more and more optimized, records will be far less frequent and tournaments are a great way to see who the best overall player actually is than someone who fluked a god run.”

As we discussed earlier, recordings have the unfortunate reality of being cheated by being edited or being a representation of an unfair advantage. However, tournaments are a fantastic way to legitimize a runner’s speedrunning skill, typically seen live. With many active viewers, many of which are knowledgeable about a game, the likelihood of someone attempting cheating (and successfully doing so) live is likely quite low. Others comment on this idea:

Support for tournaments with criticism:

Tournaments are subject to criticisms of varying degrees. As discussed below, some criticisms are simply in organization and form, while others are more about where speedrunning is going.

Some respondents talk about how tournaments may set up newer runners for failure by design, as more experienced runners will (typically) knock newer players out fast (very much game dependent):

Respondents use a term “oversaturated” to describe how many tournaments have been occurring in the speedrun world:

By far the most common complaint among respondents is scheduling and time. Many respondents feel that tournaments tend to drag on too long, and scheduling of matches is often a chaotic and sporadic operation. Much of the burden to maintain a tournament falls upon the organizer(s), which is usually not many community members, and the burden can be too much to monitor every participants’ activity to schedule races. Still, respondents discuss how the end result as a viewer and/or participant can be undesirable:

Respondents suggest alternate tournament logistics, including weekend-only tournaments, different tournament formats (comparing double elimination to Swiss or round-robin) or only broadcasting to a certain number of matches of a tournament to viewers.

Some respondents talk about incentive - some runners may not need the exposure (or necessarily want it), or may not feel its generally worth their time. Mrzwanzig’s view below is indicative of this:

“I think tournaments are a fun way to motivate people to run their game more, but the stakes in them never feel very high. Because the goal in speedrunning normally is to gradually lower your best time, a competition of who can get a faster no reset run on a particular day is somewhat removed from that, and can get tiresome in long tournaments.“

The point raised here is that community members within speedrunning view motivation and goals very differently. One runner may be very interested in performing live to an audience to showcase their ability to play, whereas others may be much more interested in lowering their personal best times. Many of the responses in this section hint at where the motivation and value for these types of tournaments lie.

Incentive can imply that perhaps larger goals should be at stake. What about monetary goals for tournament performance? As of writing this analysis, likely very few (none that the analyst knows of off hand) have had significant monetary payout for performance.

Mobiusman brings up a strong opinion:

“Not exactly a drawback but I don't think they're a meaningful representation of the runners. Speedrunning fundamentally isn't a 1v1 competition.”

Whether or not you as the reader agree with the opinion is one thing, but the point is certainly raised regarding how speedrunning is ultimately a solo effort (the actual performance of the speedrun)*. Other competitive gaming mediums (like fighting games) are direct interaction among players, but ultimately speedrunning is the player vs. the game, and that’s it. Tournaments simply showcase a speedrunner’s ability to do that consistently - interaction with the other player is essentially non-existent (for example, during A Link to the Past randomizer tournaments, racers are strictly prohibited from viewing each other’s progress). Whether or not a race is close in any given tournament is almost up to chance based on how prepared the runners are and their likelihood to perform compared to their opponent. So what value is there in a tournament?- this response posits.

* Certainly exceptions exist for co-op games, but most speedruns are 1-player

Other various points brought up in responses:


Many responses here are actually not anti-tournament, and more “I don’t care / I don’t watch them.”

The responses here are fairly limited and not particularly helpful for painting an overall picture. There are some sentiments about tournament commitment, and the draw away from what some consider actual speedrunning (pursuit of personal best times). More meaningful discussion points were available in the responses above in the previous section.


There are no major take-aways for this section - tournaments are a community event that some members enjoy to put together, and have varying degrees of success and involvement. The varying opinions with this questions highlight various sentiments about effort, worthiness, importance and enjoyment in participating/viewing. Because tournament organization is so decentralized (it is up to each specific community to put together), tournament organizers should be tasked with understanding commonplace problems with past tournaments and doing their best to organize to their best ability for their participants and viewers.

H2. Are there any other notable competitions or events you think are worth highlighting, for which you wish to see more of? (optional)

Write-in & Analysis:

This section will briefly highlight some various competitions, events or ideas that respondents would like to see more frequently. To be clear, a later section will cover ‘other speedrun content’ - this section is dedicated to tournament and competitive style events:

There are many types of events outside of the large GDQ events that occur in the speedrun community, many of which have varying levels of audience, participation and ‘community importance.’ The above is a quick recollection of some common recurring ideas and events.

I. Community Content

I1. How highly do you value ancillary speedrun content, including podcasts, talk shows, community PBs review, compilation videos and other miscellaneous community driven projects?


All of the discussion in this survey so far has been about speedrun-specific topics. However, as a community speedrunning is bound together by much more than just individual times on a leaderboard or even community events - there exist additional content-creation mediums that tie it all together.

This question is designed to see how community members think about these productions - if they are worthwhile and helpful to the community.

This survey and analysis is an example of this type of content, but is non-recurring and doesn’t fit most molds. Most podcasts/talk shows and similar content have longer-term lifespan and likely cater to specific crowds.

P.S.: Respondents really do not like the word ‘ancillary’.


On a scale of 1 to 7, respondents reflect on how highly they value such content. A fairly even spread, no major take-aways.

Write-in & Analysis

This section will explore various concepts as they arise, some of which may be more for proponents vs. opponents of the ancillary content concept.

Overall, a common sentiment is one where the respondent appreciates the idea, but don’t find them very valuable personally:

Some respondents show impatience/dissatisfaction with some types of videos aimed at garnering hits (YouTube as an example), which are not ancillary content in the sense of trying to foster community involvement, but are moreso about interactions for popularity. Similarly, content can be focused around the ‘popular’ runners rather than a wholesome view overall:

Some respondents talk less so about specific ancillary speedrun content, but moreso about the concept of commentary. Commentary is a huge aspect of modern marathons (especially GDQs)

The quality of these types of projects are a common sticking point for many respondents. For example, SummoningSalt’s videos are highly regarded as quality speedrun content, while other less genuine videos are typically ill-received from community remembers. The common sentiment is ‘good when done well, bad/pointless when not’.

An anonymous respondent gives the following sentiment:

“I think this type of content is important in that it gives off a big impression to non-speedrunner. I feel like while there is some positive content out there, most of it is centred on things such as cringe, and generally paints speedrunning in a bad light. I don't think this type of content needs to be gotten rid, nor will it, but I feel like there could be more positive content, that highlights the good parts.”

There are respondents that comment throughout this whole survey the idea of how the image of the speedrunning community is reflected and perceived by outsiders. Some community members care, some do not. It is reality, though, that newcomers will find content in various methods, whether its a Twitch stream of someone simply grinding out personal best attempts, or a compilation video on YouTube. Some argue that it’s important to maintain and monitor that image, and some, again, do not care at all.

Bony points out that given how much the speedrun community has grown, that having a centralized “speedrun content provider” like in the past is overly challenging:

“Having a single, prominent show (like TSSB and TFS, to a lesser extent, used to be) would be excellent, although it's almost impossible to do currently with how much activity there is. I don't think either of the aforementioned shows were flawless in their structures for the general community's benefits and interests, but they certainly had excellent ideas and acted as great means of keeping the community updated. As far as I know, many other speedrun shows exist currently, yet none of them seem prominent enough to be especially beneficial in general. Though I'm unsure if having a general speedrunning show is even feasible now, having more theme-based shows and projects can only help matters”

Some of the specific examples here will be elaborated on in the next question. The sentiment is that speedrunning is such a large community compared to ~2013-2015 (when those two above projects mentioned by the above respondent, TSSB/TFS, roughly began or were in their prominent visibility stages), that trying to piece it all together in one recurring podcast/production in modern times would be very difficult.

Respondents who generally favor the content like the accessibility of current events and information from this type of content, both for community members and outsiders who might be able to learn about speedrunning tacitly:

Respondents who answer 6 or 7 have a wide array of feedback, but generally say that the type of content is very valuable and wish more specially curated content existed. Some comment on the documentation of the history of speedrunning, and how this content is a good way to measure over time how things have changed in speedrunning or what used to be commonplace ideas/sentiments among runners & viewers.

Otherwise, another common sentiment, similar to the above points about how “they’re not for me”, even many respondents who favor the content like them, but understand they’re not for everyone.

Oddtom wraps up this section with a positive spin on this type of content:

“Speedrunning is about the community, and I don't think the general public understands that or understands why we value it. In my experience, people look at speedrunning and say "why would someone waste his time doing that?", but don't realize that what we really get out of it is not the gaming, but the community. It's the same reason you go to your book club or your writing meetup group or whatever. We're a group of like-minded people who are happy to find other like-minded people. We just also happen to like doing games quickly.”


Similar to other questions, respondents are very divided on this topic. Some do not value them whatsoever, and have an opinion that a ‘greater speedrun community’ doesn’t exist at all, and that speedrunning is purely an individual effort (maybe with your immediate game community & its players). Others value well curated and organized content very very highly, and wish to see more of it.

The take-away is that regardless of whether or not one community members likes it as a concept or in practice, ancillary content does exist and does impact the way that both community members (among communities and sometimes even within their own) and outsiders view speedrunning. Having an opinion on the type of content out there, and voicing one’s constructive feedback or concerns/criticisms will be how speedrunning is perceived and thrives in mediums outside of individual speedrun streams.

I2. Are there any instances of specific community content creation, whether it is an active ongoing project or a discontinued project, that you find/found notable and wish to see more of? (optional)

Write-in & Analysis

This section will highlight many of the answers provided by respondents for this question.

The first two are probably the most notorious within the history of speedrunning, and are highlighted by many respondents:

Respondents reflect very highly about the above two programs. Even if some respondents did not regularly watch the content, they are commonly cited as well-made productions.

Other responses for active/discontinued projects that respondents felt were worth mentioning include:

No major take-aways here, mostly a reference section for the types of content discussed in the prior question.

J. Other

J1. Besides timing, are there any instances you can think of where a community handles a certain area of approaching speedrunning their game/series incorrectly or inaccurately (including moderation, outreach, category decisions, etc.)?


This is a very open-ended question- the purpose is to understand and briefly explore instances where there are both common and specific community points of contention. There are many community decisions that sometimes runners don’t agree with- this survey analysis discussed timing methods earlier, but there are certainly more (which are discussed below).

There are varying opinions on whether or not someone external from an community should have an opinion (well, more than just an opinion, more like a opinion taken seriously by that community as a point of consideration).


The data section here is not particularly helpful in gauging overall satisfaction- each runner’s opinions and each community is so specific and unique, the data set is most likely affected by the sampling of those who took the survey.

But at least with that lens, respondents do show signs of unease within either their own community or another community. “Yes, my own community” does make sense being the lowest category- usually it is extended circumstances where respondents, as part of a given community, feel like they cannot influence their own community to the point of seeing eye-to-eye with the way they would prefer an issue to be handled (i.e., it may be hard to admit one can’t change things out of one’s control even for one’s own community). For commentary on other communities, likely respondents (and community members) in general wouldn’t feel too obliged to comment on other communities’ musings and decision making unless they felt it was fundamentally wrong, then they’d speak up (evident through responses herein).

Write-in & Analysis

Most respondents who wrote-in their answers have very specific cases. First, an overall bullet point list of some of the overarching issues referenced here:

Let’s discuss two popular examples that illustrate what these kind of problems are. We’ll discuss Super Metroid and A Link to the Past in very broad terms, as they are popular games with category definition complications and history. I’ve asked community members to help me paint the appropriate picture for each, as I (as the analyst of this report) am not a community member to either. Communities, games and their histories are often sensitive subjects, and the below truly aims to be a brief neutral overview of facts rather than any sort of bias or protest into their decision making, which will help illustrate why some of these issues in speedrunning are both important to community members and how they are complicated in nature.

First, let’s discuss Super Metroid (SM) and its category naming conventions. A game with an incredible amount of history and many players, easily one of the most popular speedruns. Without getting too far into the details, the category “any%”, which has the classic definition of roughly ‘beating the game as fast as possible with no restrictions’, is commonly accepted universally among most speedruns. Though there were other categories for SM, “any%” was (and still is as of this report) the most competitive, prestigious category. Dating back through history, Super Metroid continually has had multiple major findings that massively influenced the speedrun routes for various categories, to the point where the initial understanding of the category “any%” have been blurred. True “any%” (again, beating the game as fast as possible with no restrictions) eventually became an extremely glitchy and short category*. Meanwhile, the old “any%” continually lost its easily understood worldwide meaning. (What’s referred to as the old “any%” here implies that these game breaking glitches that significantly changed the game & speedrun were not allowed to be used).

* For the reader’s knowledge, the player can go out of bounds, trigger the end of the game and roll credits in somewhere around a quarter of the time of a regular run, at the highest competitive levels

What followed was a period of time where the old “any%” category was not changed - meaning it lost its meaning as a true fastest completion, but rather was still the commonly known category “any%” before the glitches came into play. And true “any%” was called “any% glitched” - which was technically a redundant category naming (any% means no restrictions, so why should the tagline “glitched” be necessary?) This proceeded to cause some long established disarray among both inside and outside community members.

Eventually, for the leaderboards, the term “No Major Glitches” (a somewhat common term in modern speedrunning for this type of problem among many different games, for which its commonly understood that A Link to the Past initially coined the term (more below about ALttP)) was used to describe what was the old “any%” - in the advent of the major glitches found, the community eventually decided for “any% NMG” treatment and the old “any% glitched” went back to being called “any%”. Again this is for the leaderboards’ reference - to date, most SM community members still refer to the old method (“any%” is the classic route, “any% glitched” is true “any%”)

A quick chart in case the reader gets lost in the details:

Common terminology among SM runners

Common conceptual modern speedrun definition

Current SM leaderboard definition


any% NMG

any% NMG

any% glitched


any% major glitches

The second column may seem a bit confusing, but just keep in mind that as stated before, “any%” is commonly understood as ‘beating the game as fast as possible with no restrictions’ among most speedruns.

However, this highlights another conceptual problem - what constitutes a “major glitch”? Current “any% NMG” does indeed have what some may consider a combination of bugs and glitches. Nothing “major” though - but what does that mean? According to Straevaras (a community member who helped me understand some of these issues), there are a few factors that describe a major glitch, most of which revolve around anything game-breaking (e.g., the space-time beam glitch, which allows you to corrupt game memory and do a handful of glitchy things, including change your immediate surroundings) and some game progression involving out of bounds exploration (where the player can go in and out of arbitrary barriers that are clearly not intended to be surpassed, like walls and doors.) However, there still exist some glitches that are either undecided or commonly disputed as being a major glitch. Straevaras gave a specific example:

“There are other glitches that are very ambiguous, and the community can't come to a consensus on whether or not they should be in the category or not.  The one specifically that comes to mind is an ammo underflow glitch that sniq found about a year ago or so now, where you perform a Crystal Flash on the same frame Draygon grabs you, and because the CF counter is also used by the Draygon fight, causes your Missile/Super Missile ammo to underflow and effectively give you unlimited ammo (think MP underflow for Tellah from FF4).”

So clearly, even with an overall attempt to keep with modern speedrunning definitions, a community still struggles to figure out what the best category definitions are among themselves. The SM community is truly unique - it contains a huge number of runners ranging from experience, age and community involvement time (i.e., how long they’ve been around SM speedrunning). They’ve come to a place where generally runners agree on certain topics- but there certainly exist fringe cases that are continually up for debate.

Let’s also briefly talk about A Link to the Past (ALttP), which has similarly had category disputes. ALttP has a very complicated past, most of which, as the analyst of this report and as a speedrunner who is not an ALttP runner in any context, I am not privy to. However, I asked EvilAsh25 & ChristosOwen to help me understand the naming convention history for its most popular category. A very (emphasis on very) brief overview is the following:

Out of this, a few assumptions and questions come to play, similar to Super Metroid above. What constitutes a “major glitch” in concept? Are there glitches used within “any% NMG” now that are very questionable if they constitute a major glitch or not? And even though “no s+q” may have had more meaning historically, why is it still a standard for runs of this category now? As of this report, for ALttP most categories are split between Major Glitches and Minor Glitches, where the primary indicator of the separation for each glitch is its application rather than its mechanics (i.e., does this glitch completely break the game with respect to the speedrun or not, regardless of how it works). This allows them to return to normal naming standards such as “any%” and “100%”, then define the rules under the appropriate Major/Minor Glitches rather than have long & complicated names.

A handful of respondents in this section talk about these types of problems related to categorization. Why is there such a long and complicated history behind the naming of categories for a game like ALttP? And what effects do all the historical implications & segregation of categories have on the community and its current players? Unfortunately, the answer to these problems is incredibly complicated and sits within the community themselves. To date, regular debates occur within ALttP’s community about naming conventions, categorization of glitches, rulesets, and more. The history of these decisions sits within a large number of debates and conversations to try and figure these problems out, and attempting to discuss them here in detail would result in a report probably as long as this analysis.

Let’s bring this all back in. Why does something like the category definition of “any%”, “any% NMG”, and “any% glitched” matter? Why does the effort of trying to figure out what constitutes a “major glitch” matter? Why can’t runners just play the game as they want? There are so many factors why this is important, but as the analyst of the report, I believe the main reason is that these communities (and many others) want to preserve competitive standards. Meaning, when a runner turns on the console to speedrun a game, they know specifically what they’re getting into, they might know the other runners’ times doing the same thing, they know their own personal goals to achieve in comparison to what they’ve done before, and they knew if they completed a run how they performed relative to a previous run. At its core, speedrunning is competitive, whether it's against yourself or against others’ times. I believe these long winded debates that may seem to get lost in the very specific details at times are, at least in their most genuine and non-self serving motivations, to preserve and encourage a proper competitive landscape.


This section was to explore, as a general sense, areas in speedrunning that are up for debate. Besides the two major examples explored above, there truly is one overall take-away from this section- and that is that decisions in speedrunning are complicated. They’re complicated for many reasons, including personalities of community members, outside vs. inside community member opinions, time investments into the hobby and practice, community attempts to be consistent and relevant, the list goes on. If nothing else, the reader of this analysis should understand that speedrunning is at its core a simple concept (beat the game fast) but in the thralls of a competitive landscape, it can be an incredibly complex beast to understand and manage well.

J2. How important do you think modern day maintenance & upkeep of Speed Demos Archive’s (SDA’s) non-forum (front page & video verification) efforts are for the overall speedrun community?


Speed Demos Archive (SDA) is a website centered around speedrunning. The website maintains an archive of polished (good quality) speedrun videos, which they actively update roughly once a week via their front page ‘blog’. SDA developed a stringent verification process for any runs submitted, meaning the content (videos) they hosted on their site had been vetted by other community members. The website also holds a forum, which houses sections for both these submission & verification processes, but also contains many speedrun planning & discussion topics (open to everyone).

In order to illustrate SDA’s prevalence in speedrunning, it is important to briefly understand its history. In the history of speedrunning in the west, SDA has been around “forever”. According to their FAQ, the site was found in 1998 for Quake ‘demos’, then expanded in 2003/2004 to more/all games. Speedrunning was an incredibly niche hobby back in 2004- this was so long before the era of modern speedrunning with Twitch, YouTube and other major platforms. Internet infrastructure was a fraction of what it is in modern times, and reliably uploading and hosting speedrun content was not easily accessible by a speedrunner. Further, speedrunning simply wasn’t popular- a very small handful of community members grew the practice by submitting runs and communicating on the forum.

However, the practice (speedrunning) definitely grew over time. Through the early thralls of streaming services (e.g., ustream, justin -> Twitch) and YouTube, SDA’s forums were one of the only main aggregators of universal speedrun content, especially the forums. The forums allowed basic communication to exist among speedrun community members who might otherwise be isolated from each other. Early GDQ (Games Done Quick) planning and discussion all happened on SDA’s forums for multiple years (before maintaining its own separate website). Anyone could visit either a previous topic for a game that already had discussion, or create a new one. And similarly before the advent of personal videos being hosted on Twitch/YouTube (and LONG before any speedrun leaderboards), SDA’s verified videos through its robust vetting process served as the most up-to-date and reliable measure of the fastest/cleanest run of a game.* Another boon to SDA’s importance was its Knowledge Base, effectively a flexible wiki that held speedrun information hubs for those who were willing to set up the page. To date, many community members reflect positively on this arm (even in modern times).

* It is important to note that SDA has never claimed to be a ‘world record’ video site/host. Their approach has always been to verify runs that are submitted with the fastest time and update their site accordingly. Throughout older speedrun history, it was not uncommon for the fastest recording of a game to be a SDA video. In modern times, it is fairly unlikely for any given game that the SDA video will be the fastest recording of a game (more on that below)

As Twitch, YouTube, IRC, Discord, SRC and GDQs evolved in their own ways over time, SDA’s critically unique one-stop-for-all-your-needs function rapidly became outdated. The verification process that was the gatekeeper of a run being hosted (where it was fairly highly regarded to have a run on SDA) is now commonly considered an undesired endeavor in both process and reward (briefly, most modern speedrunners don’t care about the video quality/vetting standards SDA imposed on everyone). The SDA forums (which are still used in some capacity) have been rapidly outclassed by other more modern communication methods- SpeedRunsLive’s IRC server with individual channels (early 201x), Discord (~2014 onwards), SRC’s forums that tie directly to a leaderboard page for any game, and general modern social media (Twitch & Twitter interactions). Further, SRC’s leaderboards (and other communities’ leaderboards) functions now gave a true leaderboard overview, rather than only hosting the fastest run (SDA’s method with their pages - other runs accessible through archival). All in, many of SDA’s unique characteristics have been outclassed otherwise.

Naturally, SDA is considered to be the birthplace of many aspects of modern speedrunning. Many speedrunners who have been running for years had their start somewhere on SDA. And many speedrunners appreciate what SDA has done.

This question is to poll on whether or not SDA is relevant anymore in its non-forum efforts. Like stated earlier, SDA is still updating their front page every week, and community members do indeed submit runs and have them verified in a very similar way to the ‘old days’ of SDA. The question is whether or not these kinds of efforts are worthwhile in contrast to the modern change of speedrunning. The question was narrowed on purpose - rather than trying to evaluate what’s relevant to some community members, the goal was to review if this somewhat arduous process of verifying videos by older, time-tested standards is important and relevant to anyone, in addition to any other efforts maintained by SDA outside the forums.


Respondents were asked, on a scale of 1 to 7, how important are non-forum (front page & video verification) efforts.

To be clear, this question is not about the forums. Community members do use SDA’s forums in some capacity (though much less compared to the past). But this question is about everything else SDA does in modern times.

Write-in & Analysis

The responses will go from general unawareness/disinterest in SDA’s modern activities, to stronger support for their modern activities.

Many respondents were unaware SDA had modern activity or unaware anyone checked them:

While others strongly (with/without reasonings) believe they’re mostly irrelevant:

Some respondents reflect on how the novelty of archival & proper video quality (some of SDA’s main benefits) are lost in modern times:

Bismuth shares an overall sentiment to SDA’s modern efforts:

“It's basically obsolete at this point. It used to be incredible in the early 2000s, but since the beginning of the live-streaming era (2009, 2010), it has been rapidly decreasing in value. Nowadays, it doesn't really serve any purpose, which is actually great, because it means that the community has evolved immensely.”

Some respondents (through their scale rating) don’t think SDA’s efforts are relevant any longer, but do look fondly back upon the archival aspects:

Many respondents generally reflect on the outdatedness of the archival and verification processes. As highlighted in the Premise, there are simply other means for both archival (i.e., modern YouTube and Twitch have all the personal and community archival power that a community needs, provided the videos are well maintained) and verification (modern verification and moderation is faster and less intricate than its SDA predecessor, which took between weeks and months to have one video verified).

Some respondents do like SDA’s efforts, but think they should be more modern and make a serious attempt to update their processes:

Even the strongest votes (6-7 on the question score) still mostly admit that the process is outdated, but that the archival and history is incredibly important:

Archival is a fairly large discussion point among respondents, and for good measure. SDA was a central archival point for speedrunning for a long time because it was one of the only mediums (and certainly the most prevalent) for speedrun video content. Now, however, with massive desegregation of communities, an overall effort to document speedrun video history is mostly not achieved. It appears many of the respondents here are aware of this when asked the question, and highlight SDA’s importance in doing so. But realistically from ~2012 onwards, archival has been nowhere near the magnitude of an overall focus and importance of speedrunning as it had been before.


SDA is clearly important to speedrunning, for many reasons discussed above. Even if newer modern runners don’t know about SDA or don’t care, the fact remains that SDA was the birthplace of many speedrun practices and communities. Even if things have changed since the SDA days, SDA even being a ‘thing’ to branch off of is a major contributor to how speedrunning is so (comparatively) popular and how speedrunning was able to assess old practices & develop modern standards.

As the analyst, I believe (like many things) it’s important to consider the roots of a community, but also be realistic about modern practicality and to asses whether or not an effort (like SDA’s front page) is worthwhile. The take-aways through the responses are that the arduous efforts that SDA undergoes to upkeep its front page are mostly lost upon respondents, but also that the idea and process of archival is highly appreciated by the speedrun community while simultaneously being something that has been mostly lost in the last few years. It seems that a focus to maintain helpful and respected ideas such as archival and the SDA Knowledge Base would be better received (evidenced by respondents’ feedback) than some of the other efforts, such as front page verification.

J3. How important do you think Games Done Quick (GDQ) is for speedrunning as a whole (as a hobby, as an artform, as a science)?


Games Done Quick (GDQ) hardly needs an introduction for most modern speedrun community members. But as a brief explanation, GDQ is a semi-annual charity event, which benefits various charities by running a weeklong+ stream of speedruns. GDQ had informal roots as a small gathering of a handful of individuals getting together in person and putting on a charity event, and year after year has grown into the massive project it is today. It’s expanse and influence is wide and vast over speedrunning, and regardless if any community member agrees or disagrees with GDQ, its mission and its values, it is a major player in speedrunning.

GDQ and pretty much everything relating to it (administration, outreach, organization, speedrun influence, on and on) are continual prevalent topics of debate among community members. The goal of this survey is to tap into community reflections and adaptations about the question asked, but simultaneously be objective in its outview. Simply put, the following discussion is not intended to be in favor or against GDQ by design, and is only tapping into respondents’ answers for relevant highlights & standpoints.

This question is specifically asking how GDQ impacts speedrunning. It may seem like a cyclical question (speedrunners influence GDQ, which in turn influences speedrunning), but the idea is that GDQ serves a platter of speedruns through a marathon to a much greater audience outside of the core speedrun community. Both short- and long-term effects on the speedrun community have occurred due to the GDQ events, including growth (new community members joining and becoming speedrunners) and hobby awareness and interest (outsiders understanding what speedrunning is).

One note - some respondents who are involved with GDQ asked to have their results removed for the analysis to prevent bias, which was accommodated.


The results are unsurprising. Respondents in the 1-3 range may have a harsher view on how a live marathon affects the specific act of speedrunning, whereas higher ranges 4-7 have a generally more appreciative view of how the hobby is affected overall by its largest publicity & audience event. But clearly ~90% of respondents are in the 4-7 range, with 7 being the most chosen option.

Write-in & Analysis

The reflections here will begin in the 1-3 range, then move to the 4-7 range. Remember that this question isn’t asking if the respondent likes GDQ or not, it’s asking if they believe GDQ is important to speedrunning. Responses that discuss their purely personal views on GDQ are omitted from review.

Some respondents discuss how a major effect, perhaps both good and bad, is to bring in an audience.

Certainly, many in the community consider growth a good thing. But one should consider the type of viewers and community members who are brought in. The idea is roughly that GDQs will grow in viewership and interest, and the speedrun community should have an after-effect of viewership and interest via increased viewership (for broadcasters, more followers, more average viewers, etc.) and community participation (community members becoming interested in speedrunning). Likely the latter (community members who join speedrunning) is a much smaller contribution to the community in terms of #s than the former (viewers).

To bring this point further, there are a fair amount of ‘good old days’ sentiments about GDQ, from a time when it was, as recalled by responses, a smaller by-speedrunners-for-speedrunners kind of event. Naturally the platform grew over time, and some respondents feel like those times have changed (which they have). Whether or not these changes are good or bad for speedrunning is a personal interpretation - however, the fact is that changes have occurred and will continue to do so as GDQ grows.

Some sentiments reflect on how the marathon is a charity event first and foremost, which influences the scheduling and programming of the events:

A common idea is that other content (like the TASBot example above), safety strategies & the ‘general audience’ preparation of the runs may turn off individuals who are very invested to speedrunning, who are more interested in the pursuit of fastest times. Having a GDQ run is typically an ‘honor’ (whatever you want to call it- it’s typically viewed as a positive/cool thing to have) for a runner/community, but naturally some speedrun community members do not enjoy viewing the content as much as others.

To bring this back to the point of the question, the overall landscape is that many speedrunners submit to GDQ, and a large amount of time and focus goes into this type of GDQ style run preparation for those who attend, which may bring attention away from what some consider the core act of speedrunning, going as fast as possible. It is a difficult happening to quantify, and every speedrunner & their community is unique, but it’s certainly true that as GDQ grows and more community members become involved, the greater speedrun community is spending more and more time adhering to GDQ style runs. Even other online or in-person marathons naturally require the runners to have safer strategies & risk mitigation in order to reasonably complete a game under estimate.

Many respondents highlight some of the speedrun community benefits of the event, which include speedrunners being able to meet up in person bi-annually, among others:

These are generally holistic responses that highlight the idea that it’s nice to have a medium where community members can interact about things they have a common interest in. Things like meeting up in person and hanging out with your speedfriends may not seem like its directly important for ‘the core act of speedrunning’, but certainly there are long term community benefits of garnering relationships and interest in the hobby that cater to the importance to speedrunning as a whole.

Some respondents discuss how if GDQ were to stop, others may take up its mantle, perhaps in different size and form:

There certainly are many other marathons that raise money for charity that either occur randomly or regularly (once a year or more). None are vast in its size like GDQ, but they have varying amounts of viewership and community benefits.

Many respondents simply reflect that giving to charity is a great and noble endeavor. Although the act of charity has no impact on speedrunning (going as fast as possible), it of course is a beneficial byproduct that viewers are interested and willing to donate by viewing speedruns performed live.

Many respondents also reflect upon the benefits of exposure and accessibility of speedrunning.

However, as noted in the ‘take up the mantle’ sentiment before, some respondents reflect that the accretive benefits of GDQ exposure has had its run. Some community members may feel that the heyday of GDQ introducing speedrunning to a greater audience has passed, and that exposure exists in many other forms in the current Twitch/online environment.

Measuring growth and success is fairly easy with respect to monitoring GDQ’s event growth (in terms of viewership and donations), but is quite difficult to quantifiably assess the growth for the hobby as an after-effect. Most community members know that viewership and interest generally increases, but it’s hard to define in terms of how many more outside viewers and supporters are actually watching & interested. The points above are saying that in the last few years GDQ has tremendously driven awareness and interest, and although they still accomplish similar ideals, perhaps their growth to the speedrun community (rather than growth through event viewership & donation) has reached a concept of diminishing returns or something similar.

Finally, to illustrate the positive messaging from many of the 6-7 respondents, here are some general ideas about GDQ and its importance to speedrunning:

And many more.


This question likely has the most accurate and relevant statistics in the data section and had the most amount of write-in responses. Community members feel very strongly about GDQ, because it is so mainstream and such a forefront of the speedrun community. This idea of a ‘greater speedrun community’ may be difficult for community members to grasp at times, but if you think about GDQ and what it does, it is the ultimate culmination of a greater speedrun community- speedrunners who may exist in their own specific communities come together and contribute their time & skills to a bigger effort.

This question asked whether or not GDQ is important for speedrunning. The answer is a fairly clear yes- it is a major player in the speedrun field and has invariably caused changes to how speedrunning has evolved over time. Outsiders have continually gained knowledge and interest about the hobby, and have contributed in various ways, ranging from donating to charity, joining the community as a speedrunner who became interested after watching GDQ, or becoming a regular viewer in broadcaster’s streams.

Many of the discussion topics were about growth, exposure, and game scheduling. Naturally, topics related to the core act of speedrunning such as developing specific speedrun strategies & competitive histories (among other niche and specific topics) are not the forefront of discussion. It’s important to remember that, directionally, speedrunners comprise the marathon with their content that aligns with a marathon’s vision and goals, then the marathon is executed to the public. That’s what the general public sees, and is how they interpret the hobby, which is what major marathons will show to them. There’s certainly a lot more to speedrunning than a marathon, but it would be wise to think about how audiences are presented with the content through these mainstream avenues, and how that affects a community member or a community.

GDQ is a major player and is responsible for an incredible amount of growth in the community. But it is important to assess that GDQ through its staff’s visions and messaging, along with every specific GDQ-runner and contributor to the event, shapes the way that the hobby is broadcasted to a very large audience. Everything from quality content, charitable efforts, speedrun topics & inside jokes, are all reflective of how the community is perceived. The take-away is that respondents here and many more community members feel very strongly about how GDQ achieves this, and it would be wise, in my opinion as the analyst, for the community to better understand & communicate what they want out of such a large event and how it's handled.

J4. Do you think modern Twitch (within the last 2-3 years, including the advent of partners, subscriptions, and affiliates) has impacted the core act of speedrunning (the specific act of an individual or community pushing a game to be completed as fast as possible)?


Whereas the previous question was geared towards whether or not a major event was important to speedrunning as a whole, this question somewhat similarly and somewhat differently asks whether or not Twitch has had an impact on the specific core act of speedrunning. This question gets at the heart of ‘focus’ in speedrunning - are speedrunners as a whole interested more in streaming success over speedrun success? Are there cases of games and communities that have changed due to the Twitch landscape over the years? Do speedrunners specifically choose and tailor their speedruns & games of choice to cater to their audience (with hopes to grow)? And does it really matter- meaning that there are speedrunners who will always push their fastest time regardless of their audience and growth potential? And how does money, a more and more involved element of Twitch and its landscape, play into all of this as a motivator? Responses to this question will review these ideas.

One thing to keep in mind, again as we’ve discussed briefly before here, is that speedrunning (in that, literally the requirements to speedrun) is not tied to streaming or broadcasting in any way. As long as a runner has a run and proof of the run, that’s a speedrun. Generally streaming that activity shouldn’t influence the run itself, and is not required (more on this later).  There may be some exceptions like live races, but streaming is purely a value-added benefit in terms of social awareness and showcasing. It’s important to distinguish this while one thinks about what Twitch is and does for speedrunning.


The trend for this question’s data set is eerily similar to question J3 - the general trend is increasingly positive with a slight exception in rank 2. Roughly 85% of respondents rank 4 or higher for this question. So, again, although Twitch and live streaming is not necessary for speedrunning, it has a massive impact on the landscape of speedrunning.

Note: It appears some respondents may not have understood the ranking system. If one were to rank this question as a 1, the respondent should have identified Twitch as not a major factor in speedrunning, whereas a 7 would indicate a strong correlation. Though this is mostly evident in responses, some seem to have gotten the idea wrong. Still, the overall trend is certainly preserved.

Write-in & Analysis

Responses here will be viewed similarly to question J3- responses in the general rank 1-3 range will be reviewed first, then move to the 4-7 range.

A few respondents here discuss how there are large streamers who get into speedrunning only as a way to change up their content to their audiences, and have no impact on speedrunning.

To give a little more context, typically the type of speedrunning referenced above is learning from others’ runs and trying it out yourself, but not actually diving into ‘pushing a game’. It’s not wrong to say it's speedrunning - some runners are good at execution, and others are good at strategy investigation and routing (and the best are good at both). But typically the focus for larger streams that the above sentiments are referring to are just learning runs for their audiences or learning runs as a semi-casual pure enjoyment aspect of their stream.

A few sentiments simply say there’s no correlation - speedrunners who have either small or large audiences are specifically unique in every case, and Twitch & its bundle of services and content don’t influence the streamer-  the streamer takes those options and chooses what to do with it, in context of speedrunning and having an audience:

Some respondents reflect on the whole no-reset run (finishing a run no matter what happens, which is not always common) mentality when a large host or raid* has come through, where the streamer may otherwise not care/bother with finishing a ‘dead run’. Similar responses comment on how the ‘grinding’ streams, which are streams filled with resets to crank out a very fast time that accepts little mistakes, may have dissipated over time:

* Hosts/raids are when one streamer (usually at the end of their broadcast) directs their page and viewers to another streamer, thereby giving the new streamer their viewers. It’s typically seen as a nice gesture, especially when a large streams hosts/raids a smaller one.

There are a fair amount of middle ground arguments in the responses here:

Some of these responses, and other similar ones, get into how some streamers have invariably split off from speedrunning over time. However the viewpoint here isn’t necessarily good or bad, it’s just commenting that it has happened, and that the core act of speedrunning is really unaffected and it's moreso about who chooses to stick around versus who moves on.

Some respondents discuss how the benefits of Twitch’s platform, with viewership and monetary incentives, are a boon to speedrunners - the motivation exists for them to continue their speedrunning and streaming activities simultaneously:

Whereas some other respondents do negatively reflect on how money and fame have affected speedrunning. This is really reflecting on the idea that someone speedruns specifically for money, rather than out of interest of the hobby. If someone speedruns and makes a lot of money, that may (or may not) be different than someone who gets into the hobby to make money:

The two above paragraphs are really two sides of the same coin, and many community members are split on this topic. Some say Twitch’s advent has caused streamers to ‘go all in’ on speedrunning provided it works for them, which benefits speedrunning. Others say money, viewership and fame through Twitch’s growing platform has caused otherwise decent speedrunners to change focus and interest away from the hobby. It’s certainly a case-by-case basis but community members feel differently about the overall landscape.

One respondent discusses briefly how the streaming aspect may be a distraction:

The point is certainly valid - remember the idea that Twitch is a value-added prospect for speedrunning. Broadcasting your content may make a runner more engaged to churn out runs, but it might come at the cost of concentration and performance (keyword might, some/many are unaffected).

The balance between grinding out runs (doing many attempts to get a personal best) versus finding strategies and investigation (routing, using spreadsheets, using TAS tools, hacking to find an answer, etc.), as some respondents highlight, is certainly affected by modern Twitch:

Some discuss the benefits of the streaming platform with respect to VODs, highlights, clips and general strat sharing among community members:

Some rank 7 responses, which point out some various topics about Twitch’s importance:

Finally, Hetfield90 summarizes aptly a major (perhaps the largest) benefit of Twitch to speedrunning, albeit an indirect one to the core act of speedrunning:

“Community is everything to most people. I assume most speedrunners wouldn't be speedrunning if they were the only person they knew doing it.”


The main motivation for this question was the subject of how community members reflect upon all the changes in modern streaming and how their once-very-niche hobby has now blossomed into a much larger medium, most of which is attributable to Twitch.

As the analyst of this report, I’ve experienced the general community backlash from the idea of money being thrown around Twitch as a bad thing during the advent of Twitch partnerships (~2014 onwards). It’s my opinion this sentiment has waned over time, and community members are much more accepting of the fact that Twitch and money doesn’t necessarily dictate the hobby. It’s also my experience to have witnessed many speedrunners who were staunchly against money in the hobby who now make a full-time living streaming some variety of content involving speedrunning.

But still, there are so many different sentiments from this section. There are definitely community members who are saddened or disappointed by the prospect of live streaming taking away from what used to be (in their eyes) talented speedrunners who’ve been swayed away. And there are others who see the live streaming aspect of broadcasting and promoting speedrunning as a massive benefit to the community, and has long term benefits for motivation and interest in pushing games.

One take-away is that it’s too difficult to summarize these ideas into one or two cohesive points, and that this issue is a forefront of how each individual community operates and how speedrunning evolves. Even if ~85% of responses have rank 4-7 stating that Twitch has significantly impacted the act of speedrunning, there are so many different sentiments on whether or not that’s a good or bad thing for the hobby with varying degrees of severity. But that’s okay- it’s just representative of how it’s a complex of an issue it is.  

And the other take-away is to remember that Twitch and live streaming is independent of actually performing a speedrun, amidst all these conversations and discussion topics. How a broadcaster, how a community, how a live event portrays (and perhaps benefits from) speedrunning via live streaming is a value-added/detrimental (depending on your view) separate step. It has a massive impact and is clearly important to the hobby, but going fast is still going fast, and that’s up to the runner and their motivations to achieve.

J5. After all of the above questions, is there anything else you think the speedrun community could be better at doing, or can learn from other communities? (optional)


Clearly the most open-ended question. We’ll summarize some major common themes, and take a look at a view various sentiments to wrap up the survey.

Write-in & Analysis

Common sentiments/requests among responses:

And some overall positive sentimental responses to close out the survey:


“I honestly think we're on the right track. We should do more of some of the things we're already doing (sharing information as a community, advertising events publicly, removing that initial barrier of entry and getting new runners access to information), but this is all stuff that's currently improving, so I think we're moving in the right direction.”


“Actually, not particularly; in fact, I'd like to point out the opposite - that, as far as I've seen, the speedrunning community at large is a very friendly place in general, and that I've seen few communities as welcoming and hospitable as people who try to get through video games as fast as possible.”


“Enjoy what we do and improve yourself. Speedrunning is a great hobby to learn about patience, perseverance, mental toughness, dealing with loss, learning how to win, making friends, competition. Let's try to preserve the good things that we have. Help out the individual. They may be the next great speedrunner, but may need a little encouragement. We can be open to new ideas also. I think there is an "old boi" tag on some of us, that we don't want to change, and it was better back in my day. While that may be true, new ideas are also good and we should always try to have more.”

III. Closing Thoughts

I hope you enjoyed reading this survey analysis. Truthfully, I was expecting this project to be done in a weekend or two from the outset, and when I received almost 800 responses with very, very detailed write-in responses, I felt obligated to put something more meaningful and cohesive together. This effort took me multiple months.

There are so many ideas and concepts at play for our community. Truly, most problems and issues come down to a case-by-case basis. But that’s not to say there isn’t value in taking a step back and assessing how we all approach issues & problems as a whole. A community is just that- a collection of individuals who are drawn to a medium through interest. We enjoy speedrunning, and we spend a lot of time talking to each other about it, so it seems wise to understand each other a bit better and take some time to reflect.

I’d like to once again thank everyone who helped with this survey, either by helping me with the survey format or by filling out the survey.

Thanks for reading!

- cleartonic