Pitch Counts

by

Kenneth Matinale

June 8, 2008

Joba, Joba, Joba.  The Yankees have finally moved Joba Chamberlain from the bullpen into the starting rotation.  A big part of the decision depended on pitch count.  Any time pitch count is mentioned it is for a particular game.  However, the Yankees should consider Joba's pitch count for the season in deciding how to use him.

Pitch Count Estimator

tangotiger.net: Pitch Count Estimator

Tom Tango developed a pitch count estimator for seasons when pitches were not counted.  See the above link.  While there is no date, I think he did this in 2000.  His example of Brad Radke and Randy Johnson concludes that Johnson threw about four pitches per batter faced and Radke threw about 3.5.  Johnson threw 14% more pitches per batter faced.  That's why the Yankees need to track Joba's cumulative pitch count, not merely track his innings.

How many pitches do starting pitchers throw in a season?  How has that total varied over time?

Tom Tango has two versions:

bPCE - basic pitch count estimator: (3.3 x BIP + 4.8 x K + 5.5 x BB)

xPCE - more exact pitch count estimator

The more complex version as written by Tom Tango:

Number of pitches per batter faced = Pitches per BIP x BIP rate

+ Pitches per BB x BB rate

+ Pitches per K x K rate

Well, that's true if we don't consider HBP, Sacs, and IBB. For ease, I (Tom Tango) am removing these for the balance of this discussion.

To estimate pitches per BIP, we use the following

Pitches per BIP = 2.5 x [(1 - BIP rate) ^ 0.08] + 1.

That ^ means "to the power of". When the BIP rate is 1, this results

in 1 pitch. When the BIP rate is 0, this results in 3.5 pitches. The

other two equations are:

Pitches per BB = 1.5 x [(1 - BIP rate) ^ 0.10] + 4.

Pitches per K = 1.9 x [(1 - BIP rate) ^ 0.07] + 3.

The power of this equation is that it can theoretically be applied to

any environment (MLB 1911, college 1992, high school 2001, etc), as

long as that environment uses the current 4-ball, 3-strike,

2-strike-foul rule.

I have used these equations  to produce spreadsheets.  The data was derived from the Lahman database using Microsoft Access.  It is based on AL & NL pitcher season from 1903 through 2007.

__________________________________________________

Individual Seasons

For pitcher seasons with at least 200 innings here is summary data:

200 Innings Year Totals

As you can see in 2006 and 2007 the most pitches thrown by an individual were about 3,700, the lowest totals ever for a full season.  In 1903, 1904, 1907 and 1908 the maximum was well over 6,000 pitches.  The first time the max dipped below 5,000 may (remember, these are estimates) have been 1924, about when home runs were increasing a lot.  From 1947 through 1968 the max was probably below 5,000 each season except 1953 and 1965.  Expansion in 1961 and 1962 increased the number of team games per season from 154 to 162, about 5%.  From 1969 through 1979 the max was probably over 5,000 each season except 1975 and 1976.  From 1989 through 2005 the max was usually about 4,100 to 4,200.

Were pitchers in the dead ball era iron men?  I doubt it.  See the career section below.  The pitch count estimator may be less accurate for the years before home runs became a dominant force: 1903-1919.  I also have a hunch that pitchers simply did not throw as hard back then.  After all there is a player who makes about 150 throws each game and does not get a sore arm: the catcher.  That's because he does not throw breaking pitches and he does not throw too hard.

Here are a couple of articles on the subject:

hardball times: what pitch counts hath wrought.

hardball times: the long and the short of plate appearances.

The data for pitches estimated per season:

PCE (Pitch Count Estimator) by Year Totals

Some terms and equations:

Balls In Play - BIP: [BattedOuts]+[sumofH]

basic pitch count estimator (BPCE) (3.3 x BIP + 4.8 x K + 5.5 x BB)

BatterOuts: ([SumofAB]-[SumofH])+[SumofSH]+[SumofSF]

RatenonBatterOuts: [IPouts-BatterOuts]/[SumOfIPouts]

BattedOuts: [sumofIPouts]-[sumofSO]

Faced: [sumofH]+[sumofBB]+[sumofIPouts]

BIPrate: [BIP]/[Faced]

PitchesPerBatter: [bPCE]/[Faced]

PitchesPerBatterFaced: [xPCE]/[Faced]

The most significant columns are in bold.

I had wondered if pitchers one hundred years ago threw significantly fewer pitches per batter than pitchers today. The difference using xPCE is about 3.3 versus 3.6.  We should also consider pitches per out.  Pitchers are paid to retire batters, not to face many batters.  From 1903 through 1920 pitches per out were about 4.7 to 4.9.  After 1920 they never dipped below 5.  The highest number was in 2000, about 5.49 pitches per out.

______________________________________________

World Series

I learned that there are World Series pitch counts in each box score but NOT totaled for the WS back to 1974. I found this by trial and error. Did not see it stated.

World Series game data.

This spreadsheet can be used for individual games:

Based on an indication that there are actual pitch counts I entered the WS data for 1911 and 1912 leaving the actual column blank.  The first row has Larsen's perfect game in the 1956 WS, which I understand took 97 pitches. The basic pitch count estimator (bPCE) calculated 99.6 and the more complex (xPCE) 98.

I uploaded an Excel spreadsheet. Data was  entered into columns A through H. Formulas are in the columns to the right starting with I (BattedOuts).

______________________________________________

Individual Season Highs

Season highs 1903-2007

Of 4,652 pitcher seasons from 1903 through 2007, 66 had at least 5,000 pitches thrown by individuals; three had at least 6,000 pitches: Jack Chesbro (1904) 6,238; Ed Walsh (1908) 6,218 and Joe McGinnity (1903) 6,170.  For seasons 1903-1909 Chesbro had 27,845 pitches; Walsh 1904-1917 41,786; McGinnity 1903-1908 27,845.  Nolan Ryan had the most career pitches, about 90,000.

Of those 66 seasons of 5,000 pitches, 47 occurred before 1920, i.e., before players started hitting a lot of home runs.  Twelve after 1946, all in the 1970s; of those 12, five were by knuckleballers: Phil Niekro (3), Wilbur Wood (2); Nolan Ryan had three in five years; Mickey Lolich, Steve Carlton, Bill Singer and Gaylord Perry had one each; four did it in 1973: Wood, Niekro, Ryan and Singer.

pitches per batter faced:

fewest: Oscar Jones (1904) - 3.32; about 5,340 pitches

most: Nolan Ryan (1977) - 4.04 (21.7% more); about 5,241 pitches; only season over 4 per batter

Ryan had the three highest, Feller the next two.

pitches per out:

fewest: Pete Alexander (1916) - 4.45; about 5,189 pitches

most: Nolan Ryan (1977) - 5.84 (31.2% more); 5,241 about  pitches

Ryan had three of the top five.

46 of the 66 seasons had 20 wins.  Vic Willis (1904) had the fewest wins and the most losses: 12-2922 of 66 seasons had 20 losses.  Most wins: Jack Chesbro (1904) 41-12, Ed Walsh (1908) 40-15.

There were 963 seasons with at least 4,000 pitches thrown by an indiviual pitcher.

most: Bob Feller (1938) - 5.95 (40.7% more); about 4,957 pitches

______________________________________________________

Individual Career Highs

Tab "xPCE top 100" of this spreadsheet contains the top one hundred pitchers in number of pitches thrown for careers that started in 1903 or later.  This contains a bias against dead ball era pitchers, including Christy Mathewson who played from 1900 to 1916 and had an estimate of about 67,000 pitches.  Christy Mathewson is the only Hall of Famer pitcher who would have been ranked among the top twenty or so in pitches thrown.  While the numbers become increasingly unreliable the further back one applies the equation, ten Hall of Fame pitchers whose careers started before 1903 are shown in a separate tab in the spreadsheet.  Cy Young is estimated to have thrown 102,000 pitches.

Of the top ten only two are old timers, Johnson and Alexander, despite the fact that individual season highs are dominated by dead ball pitchers.  Nolan Ryan is the all time leader in the number of pitches thrown, almost 90,000.

Of the 151 pitchers who threw at least 40,000 pitches (see tabs in Career highs 1903-2007 with data on all 151):

138 had height listed in the database:

tallest: Randy Johnson - 82"

shortest: Dolf Luque - 67"; Eddie Cicotte and Joe Bush 69"

115 were at least 72".

heaviest: Rick Reuschel - 235

lightest: Tommy Bridges - 155

46 were at least 200 pounds.  123 were at least 180.  142 were at least 170.

65 had a first season before 1946.  82 before 1959.

12 were active in 2007.  77 had a final season after 1970.  97 had a final season after 1950.

pitches per batter faced:

most: Nolan Ryan - 3.925 (18.5% more); about 90,000 pitches

Of the 36 fewest, only Lew Burdette (1950-1967) started pitching in 1950 or later; he threw the eighth fewest - 3.35.  Robin Roberts (1948-1966) is the only other pitcher of those 36 to start pitching after 1935; he is number 35 - 3.44.  Burdette's xPCE for his three complete game victories against the Yankees in the 1957 World Series (Oct. 3, 7, 10,) were: 132, 114, 115.

Of the 29 pitchers who threw at least 60,000 pitches:

fewest: Ted Lyons - 3.33; about 60,297 pitches

most: Nolan Ryan - 3.925 (17.9% more); about 90,000 pitches

pitches per out:

most: Bobby Witt - 5.75 (28% more); about 42,525 pitches

Next most:

Sam McDowell - 5.578 (24% more); about 41,710 pitches

Nolan Ryan - 5.558 (23.5% more); about 90,000 pitches

Lew Burdette had the seventh fewest: 4.74.  Robin Roberts had the fifteenth fewest: 4.79.  Of the 21 fewest they were the only two who started pitching after 1933.

Of the 29 pitchers who threw at least 60,000 pitches:

fewest: Pete Alexander - 4.66; about 72,588 pitches

most: Nolan Ryan - 5.558 (19.2% more); about 90,000 pitches

Four pitchers were active in 2007:

Tom Glavine (21) - 5.14 (10.3% more); about 67,064 pitches

Roger Clemens (25) - 5.24 (12.4% more); about 77,295 pitches

Randy Johnson (28) - 5.43 (16.5% more); about 62,829 pitches

__________________________________________________________

Individual Career and Season Data

Individual Career and Season Data

The data in this file is for the 151 pitchers with at least 40,000 pitches in their careers for seasons 1903-2007.  It contains a combination of career and season data.

Career data is for seasons 1903-2007 regardless of the number of innings thrown in any particular season.  Season data contains minimum and maximums for seasons in which a pitcher threw at least 200 innings.

The sort field in each tab is in bold; here are the tabs:

xPCE - sorted descending on career pitches; Nolan Ryan 89,802;

xPCEseasonMax - sorted descending on maximum pitches in a season; Ed Walsh 6,218;

PitchesPerOutCareer - sorted descending on Pitches Per Out in a career; Bobby Witt 5.75; minimum: Babe Adams.4.5; Witt is 28% higher than Adams; Witt is least efficient, Adams most efficient;

PitchesPerOutSeasonMin - sorted ascending on minimum Pitches Per Out in a season; Babe Adams 4.26;

PitchesPerOutSeasonMax - sorted descending on maximum Pitches Per Out in a season; Randy Johnson 6.07 (42% higher than Adams).

The final percentage difference is in stark contrast to that shown by Tom Tango in his example in the first section of this document, Pitch Count Estimator.  He drew a sharp contrast between Randy Johnson and Brad Radke, showing that Johnson threw 14% more pitches per batter faced.  I think that pitches per out is a much more important way to compare.  The difference between Randy Johnson who was still active in 2007 and Babe Adams who pitched a hundred years ago is startling.  To retire a batter Randy Johnson throws FORTY-TWO percent more pitches.  This could be a significant factor in the difference in innings thrown now versus a hundred years ago in the dead ball era.

__________________________________________________________

xPCE for MLB 1903-2007 plus chart of pitches per out.

***   The End   ***