Comparison of GeoGuessr World Maps

	B	C	D
1	Problem
2	Popular GeoGuessr world maps are too similar to one another in how locations are distributed and there is murky rationale for those allocation decisions. I enjoy playing these maps and respect the decisions made by these prestigious mapmakers, but I think the allocation of locations should be done differently. The primary reasons I see behind their decisions were twofold: preferences for both "countryness" and rural locations. My analysis comparing the allocations by country are detailed on the Comparison tab.
3	Solution
4	I generated an objective method to weight world maps, by population and estimated Google Street View coverage. A Populated World is my forthcoming re-playable world map designed to represent where people live. Each of the total 50,000 locations will represent approximately 43,000 people. My map will be egalitarian in that respect, if the Google car drives past your house, you have an equal chance of being represented with a map location, no matter where in the world you live. Each of the hand-picked locations will be optimized for no moving, panning, or zooming (NMPZ).
5	Methodology
6	Weighting this map involves two simple steps:
7	1) Gathering objective population statistics by country and region.
8	2) Estimating the percentage of the population covered by Google Street View in each country.
9	The net result is that of the world's 7.8 billion people, I estimate that 2.2 billion are covered by Google Street View. Urban locations represent approximately 74% of the covered population and rural locations represent 26%. I will further breakdown the top-level country locations by region and city based on the statistics available for each country. The number of allocated map locations are calculated on the A Populated World tab.
10	Responses to Potential Objections
11	#	Potential Objection	Response
12	1	Every mapmaker has the license to make their own decisions, maybe they were optimizing for the enjoyment of seeing novel and beautiful locations and not sheer accuracy of representation.	Fair enough! Again, I like playing these maps and respect the mapmakers tremendously for providing their expertise. My primary criticism is that these maps are all weighted so similarly and their biases are all in the same direction, with very marginal tweaks around the edges. Instead, I think we can reimagine the weightings for the big world maps to come.
13	2	Ugh, [Indonesia / US / Brazil / Japan] locations all look the same to me, why would I want to see more of those stupid rounds where I get bad scores?	That's an incredibly ignorant take. These large, populous countries are incredibly diverse and beautiful. If you want to get a great score in GeoGuessr, just go grind the Famous Places map repeatedly. If you want to learn about how people on this planet live, perhaps you should be more open to exploring big countries in a way that is representative of the entire world's population.
14	3	You're an American, so of course you just want more United States rounds.	Nah, I'm not upweighting the United States beyond where it should be on objective measures. I'm enthusiastic about exploring more Brazil and Japan as well, two countries that I enjoy seeing and have visited many times. I really want to learn more about Indonesia and Bangladesh, places that I've never been. I'm not cheerleading for any country or countries, in fact, I think that home-country boosterism is antithetical to GeoGuessr's tagline: "Let's explore the world!"
15	4	Small countries are important and deserve extra representation.	I'll admit that when I first started playing GeoGuessr, I was in awe of how people would get the tiny country of Lesotho correct. I had no idea what the beautiful country looked like and it's been an amazing journey learning about it. Now it has gotten to the point where I recognize Lesotho instantly, I've seen this road dozens of times, and I get a big score with little effort, so the novelty has worn off. Just because a small collection of people are citizens of a central government, are they somehow more important than anyone else?
16	5	Rural rounds are inherently better because civilization is bad.	Hey, that's a preference. If you want to worship the different shrubbery around the world, I get it. I'm more interested in people and how they live their lives. In some countries, a majority of the population live in rural areas, so that's where a majority of the locations for that country will be, so I'm not anti-rural.
17	6	It's not the country's fault if stupid Google Street View doesn't fully cover it.	Totally agree, but if you artificially upweight locations beyond the actual coverage available, you're just going to get more repetitive locations in the same general areas. Look, we all want every road in the world to have top quality coverage, but that's just not the reality of the images we have in GeoGuessr at this point in time. One day.
18	#	Potential Objection	Response
19	7	Someone could cheat by Googling signs in urban areas.	Googling is often considered cheating and cheating is bad, but trying to design a map to be cheat-proof is less effective than actually enforcing the norm of playing fair.
20	8	I disagree with your Google Street View % Coverage estimate for XYZ country, so there.	This is not a science, I made the best estimates I could and cross-checked with some others in the community. Remember, it is meant to represent the % of people whose houses the Google car drives past, not the % of road surface area mapped. These estimates are only in 5% increments, since I don't think I can get more precise than that. If a country has 2.5% or less official coverage, I'm rounding that down to 0%. I don't give any country higher than 95% coverage. I am open to reallocating these figures if you present good reasons. I will also be refreshing periodically as Google rolls out new coverage for previously unmapped roads.
21	9	Aren't you just somehow trying to capitalize on all the excitement behind A Community World and starting controversy just to get attention?	Nope. I fully support, endorse, and am helping in a very small way with a A Community World. I think project is admirable mainly because MatePotato organized this global community's unique knowledge and talents. The main advantage of this new map will be its local knowledge of locations, where regional experts are showing the best representations of their countries and regions. The weightings for this map have been finalized and I am not here to try and influence any decisions for this map, but I thought it would be good to raise this issue for future maps. A Community World's creation did, however, prompt me to rehash my long standing objections to country-weightings in popular GeoGuessr world maps and thought that I should fully present my analysis, arguments, and proposed solution here.
22	10	Country streaks on your map would be boring because 51% of the time the answer would be the US, Indonesia, Brazil, Japan, Mexico, Russia, or Philippines.	Hey, you did some math, nice! You're probably right, but country streaks seem to be most interesting on maps that are optimized for "countryness", which isn't the point of this map. Lots of GeoGuessr game modes are instead about getting the closest distance, not reflexively plonking within the appropriate political boundaries for that color bollard that you see on the side of the road.
23	11	Maybe the other mapmakers were just optimizing for land area instead of population, did you ever think about that? Idiot.	SPOILER ALERT They weren't. On the whole, countries with large land areas are underrepresented and countries with small land areas are overrepresented. Go check out the most Overrepresented and Underrepresented countries on the Top tab.
24	12	Europe is just better than North America, so that's why.	Ok, tone down that blatant continentism. It's actually not even that simple, Canada is overrepresented in these maps and countries like France, UK, and Italy are underrepresented.
25	13	If I ever played your stupid map and saw Japanese script, I would just plonk Tokyo and win.	You could do that, but the Tokyo metro area is large geographically and you would still be wrong about 70% of the time because not everyone in Japan lives near Tokyo, so you might not win?
26	14	Who the hell do you think you are?	toro3317, hi.
27	#	Potential Objection	Response
28	15	Ok... so just go play the default GeoGuessr World map then, idiot.	I don't know how the default World map is weighted, my instinct is that it is closer to my proposal, but probably falls somewhere in between. The default World map is not optimized for NMPZ, the starting view is always pointed North and I enjoy playing the NMPZ game mode. You also get trekkers and indoor locations, which my map will avoid. Plus, I enjoy the process of hand-picked mapmaking and will be making portions of this map on my Twitch stream.
29	16	I'm triggered by the sight of poverty on the streets of Bangladesh and outraged at the idea of how environmentally destructive American suburbs are, I never want to see them. I thought GeoGuessr was a safe space for me.	Aww... you poor baby. Seriously though, aren't you at all curious how other people on this planet live?
30	17	Why don't you just conform to this allocation norm? The existing maps are perfect, leave them alone!!!	Ok, you can keep playing the same maps. I thought some variety might spice things up a bit.
31	18	I don't trust the population estimates produced by those Spooks at the CIA or anything the US Government says.	Yeesh, well... these estimates are produced by career professionals who take care to get the figures as accurate as possible. Sorry that your anti-American paranoia got in the way of embracing facts and figures.
32	19	I see that you classified XYZ as a [country / territory], but I think it instead should be a [country / territory], I'm pissed!!!	I'm not the arbiter of such things. I have my opinions, but please just go yell at the clouds.
33	20	Nerd.	Guilty as charged, but you've also made it this far, so...
34	21	I still hate your ideas and your face, but when will I be able play this garbage map you're making?	Sometime in the first half of 2022 I hope! It's going to be a long process, but I hope to finish A Populated United States before the end of 2021. I would appreciate any feedback or help on any locations!
35	Conclusion
36	I hope you at least find the compilation of statistics all in one place useful. Thanks for checking this sheet out.
37	Come yell at / cancel me here:
38	https://www.twitch.tv/toro3317
39	https://twitter.com/TTVtoro3317
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100