How many major league baseball games all time?
Prof Wonmug
ever including regular season and post season?
Richard R. Hershberger
> Does anyone know how many major league baseball games have been played
> ever including regular season and post season?
I don't, but it would be a straightforward, if tedious, exercise to
calculate it from the information at www.baseball-reference.com or
retrosheet.org You will have to make a philosophical decision about
what counts as a major league and how to treat tie games, but these
are not killer questions.
Richard R. Hershberger
Gerry Myerson
<ede28381-b5aa-4278...@s29g2000yqd.googlegroups.com>,
Maybe not so tedious at those sites. E.g., with a few clicks
(and no calculations on my part) I got bbref to tell me that
since 1901 (and including games played to date in 2010)
there have been 351956 regular season team games played,
now divide by 2 to get 175978 actual games.
--
Gerry Myerson (ge...@maths.mq.edi.ai) (i -> u for email)
Prof Wonmug
I asked because after the perfect game by the A's, someone said that
it was only the 19th perfect game ever. I got to wondering if perfect
games occur about as often as would be expected by chance given the
average odds of a player getting on base and spoiling the perfect
game.
I just found something on Wikipedia that claims that a perfect game
occurs about every 11,000 games.
http://en.wikipedia.org/wiki/Perfect_game
The actual text is:
"Over the past 134 years of Major League Baseball history, there have
been only 19 official perfect games by the current definition. In sum,
a perfect game occurs once in about every 11,000 major league
contests.[4]"
Clicking on note [4], we are taken to
http://en.wikipedia.org/wiki/Perfect_game#cite_note-3
which says,
"Holtzman (2003), writing in June 2003, before the three most recent
perfect games, references Buckley (2002), although there are at least
two arithmetic errors. It is unclear where the dividing line is
between Buckley's facts and Holtzman's conclusions, but regardless of
that, the numbers do not work out. The total number of games sits at
388,382 as of July 24, 2009,[1] which squares with an estimate of
about 360,000 in 2002. Each game is a paired contest, so the total
number of games actually played is half that number, or about 180,000
as of 2002. It appears that the author corrected that one figure but
failed to correct the arithmetic otherwise. 180,000 divided by 16 is
more like 11,000 than 22,000. He also got the percent wrong. 1 divided
by 22,000 is .0000454, or .00005 rounded. However, expressed as
percent ("per hundred"), it's .005, not .00005. Correcting the error
otherwise, 1 in 11,000 is more like .009 percent. The full quote in
the cited article is: "According to James Buckley, Jr., perfect games
occur once every seven to eight seasons. Buckley's Perfect, published
last year, is an analysis of the 16 perfectos and also includes
perfect games broken up with two outs in the ninth inning. Buckley
estimates that since the birth of the National League in 1876, there
have been about 180,000 games. A perfecto surfaces once in
approximately 22,000 games or .00045 percent. Don Larsen of the 1956
Yankees authored the only perfect World Series game.""
Excel calculates that 1 perfect game every 11,000 games works out to
an on-base average of 29.1534% (=1-(D17^(1/27))), which seems about
right.
If there is one every 11,000, then it would take about 209,000 games
to get 19 perfect games, which is close to your estimate of 175,978,
which did not include the post-season.
So, it looks like perfect gamnes occur about as often as one would
expect by pure chance.
I then had Excel calculate the odds of a perfect game and the mean
games between perfect games given various on-base percentages.
% On % Perf Game MG2PG
50.00% 0.00000075% 134,217,728
45.00% 0.00000977% 10,237,818
40.00% 0.00010235% 977,049
35.00% 0.00088851% 112,548
30.00% 0.00657124% 15,218
29.75% 0.00723521% 13,821
29.50% 0.00796354% 12,557
29.25% 0.00876222% 11,413
29.15% 0.00910282% 10,986
29.00% 0.00963774% 10,376
28.00% 0.01405974% 7,113
27.00% 0.02040409% 4,901
26.00% 0.02946161% 3,394
25.00% 0.04233057% 2,362
24.00% 0.06052949% 1,652
23.00% 0.08614887% 1,161
22.00% 0.12205463% 819
21.00% 0.17215969% 581
20.00% 0.24178516% 414
19.00% 0.33813919% 296
18.00% 0.47094915% 212
17.00% 0.65329374% 153
16.00% 0.90269435% 111
15.00% 1.24254176% 80
If a pitcher (plus his defense) can keep the on-base percentage down
to 27%, he should get a perfect game every 5,000 games on average. To
get it down to around 300 games, the number of games a good pitcher
mught be able to pitch, he'd need to keep the on-base average down to
under 20%. But then he'd probably be winning 30+ games / year. ;-)
Dale Hicks
says...
> On Fri, 14 May 2010 09:02:37 +1000, Gerry Myerson
> <ge...@maths.mq.edi.ai.i2u4email> wrote:
> >
> >Maybe not so tedious at those sites. E.g., with a few clicks
> >(and no calculations on my part) I got bbref to tell me that
> >since 1901 (and including games played to date in 2010)
> >there have been 351956 regular season team games played,
> >now divide by 2 to get 175978 actual games.
>
> I asked because after the perfect game by the A's, someone said that
> it was only the 19th perfect game ever. I got to wondering if perfect
> games occur about as often as would be expected by chance given the
> average odds of a player getting on base and spoiling the perfect
> game.
>
> an on-base average of 29.1534% (=1-(D17^(1/27))), which seems about
> right.
>
> If there is one every 11,000, then it would take about 209,000 games
> to get 19 perfect games, which is close to your estimate of 175,978,
> which did not include the post-season.
>
> So, it looks like perfect gamnes occur about as often as one would
> expect by pure chance.
Shouldn't it be 17 games per 351956 (games pitched)? There's a
pitcher on both sides in every game. I think we can probably ignore
the double-perfect game for the purposes of the thought experiment.
Two perfect games were in 1880, before Gerry's 1900 cutoff.
That's 1/20703, for an average OBP of 0.307 for things to happen
exactly as the odds state.
Using the data from
http://www.baseball-reference.com/leagues/MLB/bat.shtml
It seems that the historical OBP since 1900 averages out to .329. So
there should have been 1 per 47697 games (seven instead of
seventeen).
> If a pitcher (plus his defense) can keep the on-base percentage down
> to 27%, he should get a perfect game every 5,000 games on average. To
> get it down to around 300 games, the number of games a good pitcher
> mught be able to pitch, he'd need to keep the on-base average down to
> under 20%. But then he'd probably be winning 30+ games / year. ;-)
Pedro's career OBPA was .276. He won 20 games only twice. And
started only 409 games, so we shouldn't have expected a perfecto, I
guess.
--
Cranial Crusader dgh 1138 at bell south point net