digg

Join DiggAbout

Discover the best of the web!
Learn more about Digg by taking the tour.

See the original image at nytimes.com

What were the odds of Joe DiMaggio's hitting streak?

nytimes.com — Fans and scientists alike have described Joe DiMaggio ’s 56 game hitting streak as well nigh impossible. The writers decided to calculate how unlikely it really was.

Bury
show profanity settings
expand all
  • kleezy, on 03/31/2008, -4/+2Be prepared for "in 5,295 baseball universes." If that doesn't scare you, you'll enjoy this article.
  • BingoPower, on 03/31/2008, -12/+11"Here’s how it works. Think of baseball players’ performances at bat as being like coin tosses".

    And there goes the integrity of this "research".
    • Meesher, on 03/31/2008, -0/+10If you read more carefully, you'll see that they computed the chance of each player getting a hit at each at-bat, based on his stats for the given year--not merely the 50-50 chance of a coin toss.
    • sfrench, on 03/31/2008, -0/+6I don't think he's making any assumption about skill, he's simply saying "they either get a hit, or they don't", in which case it's analogous to flipping a coin. Pretty much everytime somebody starts doing a statistical analysis, they start with the equiprobable state and then adjust their probabilities to a model.
    • Buckiller, on 03/31/2008, -0/+2He was making it readable to a more general audience... he still has random numbers and batting averages in there.
    • hutch113, on 03/31/2008, -1/+1Yes, but the problem with taking the season average is that nobody is a consistent .300 hitter every day, every week. Each player will go through several mini-streaks where they will bat well below, well above and sometimes near that average.
      • tbstudee, on 03/31/2008, -0/+1That's why it's called batting *average*.

        The experiment in this article would have actually demonstrated these mini-streaks you're talking about. Obviously there has to be hit/no-hit streaks of varying lengths to get a .300 average.
        • EtherGnat, on 03/31/2008, -0/+1Unlike coins, players actually go hot and cold, not to mention the fact batters face pitchers and teams of varying skill (who also run hot and cold). Even that doesn't tell the whole picture, though. As players approach or surpass records, or go on long streaks, a tremendous amount of pressure begins to build on a player, which takes a toll both mentally and physically over time.

          All told I would suspect a player's batting average alone would be a poor model to base his odds of a 56 game hitting streak on.
  • Nev9, on 03/31/2008, -1/+2"Dunkin' Joe DiMaggio"
  • Sandurz, on 03/31/2008, -4/+3An interesting and thorough study.

    Good article.
  • epgyd, on 03/31/2008, -1/+2I don't know.

    "Here’s how it works. Think of baseball players’ performances at bat as being like coin tosses. Hitting streaks are like runs of many heads in a row. Suppose a hypothetical player named Joe Coin had a 50-50 chance of getting at least one hit per game, and suppose that he played 154 games during the 1941 season. We could learn something about Coin’s chances of having a 56-game hitting streak in 1941 by flipping a real coin 154 times, recording the series of heads and tails, and observing what his longest streak of heads happened to be."

    So, in other words, if we forget the actual sequence of games that he got a hit in, and sort of randomly reorder them, then it is unlikely that the reordering would have a hitting streak as long as that? I mean, is it maybe possible that these games that he achieved the hitting streak were correlated, simply by being close to each other? That is, that while globally over the season it was 81% chance per game of getting a hit, during an 80 game run during which the streak took place, the chance was probably closer to 95%?
  • the6thReplicant, on 03/31/2008, -1/+1I wonder what the same analysis would be for Don Bradman?
  • ieici78, on 03/31/2008, -1/+0nice article.
    very interesting seeing the outcome of this research.

    Play Ball
  • Omega19, on 03/31/2008, -4/+1"Where have you gone Joe Dimaggio? Our nation turns it's lonely eyes to you"

    For the man and the era, that lyric has never been more relevant.
  • sfrench, on 03/31/2008, -0/+2possible starting points to the streak: games 1-98 (this means there are 98 possible streaks in the season)
    odds of getting a hit in all 56 games of a particular streak: 81%^56 = .00075%

    So the odd of the streak happening during the season is the sum of the individual streak probabilities.
    (.00075% + .00075% + .00075% ...) = .0735%
  • Stone420, on 03/31/2008, -3/+1Say it aint so Joe
    • lejake, on 03/31/2008, -0/+2Wrong Joe...you're referencing Shoeless Joe Jackson from 1919 Black Sox fame.
  • EVILKILLER, on 03/31/2008, -1/+3"Our simulations did something very much like this, except instead of a coin, we used random numbers generated by a computer." A computer can't generate a true random number.
    • Buckiller, on 03/31/2008, -0/+1random.org
      Thats what I use. I wonder if this study did something like this with random numbers not generated from the computer.
    • EtherGnat, on 03/31/2008, -0/+1"A computer can't generate a true random number."

      ...but it can a pretty good job. There are more compelling real world problems with this study IMO (see my post above).
  • proliance, on 03/31/2008, -1/+1Mr. Coffee is da bomb.
  • JesusHatesYou, on 03/31/2008, -3/+0What are the odds of going 19-0 when you don't have film of the other team's play calling? Probably alot worse than getting a 56-game hit streak when you have 3 or 4 shots at getting just one hit a game. Make that 7 at-bats on those Yankee teams.
  • dslartoo, on 03/31/2008, -0/+5More proof that you can prove (or disprove) just about anything you want, with statistics. Still, a damned interesting article. My dad, who's both a mathematician and a baseball fan, will love this one.
  • blankhorizons, on 03/31/2008, -0/+3They don't take into account opposing pitching, which makes this whole thing bogus. You can't just factor hits, games and plate appearances. If you face weak pitching one day you might go 4/5, but the next day you might run into an ace and go 0/4. Opposing pitching has such a huge factor in this that the article is completely useless. If you faced a different pitcher every at bat, then yeah this would be ok, but you don't, so it's not.
    • sfrench, on 03/31/2008, -0/+3Very true, and pitchers also used to pitch more complete games. So it was more likely players could get hits off wobbly-armed pitchers late in the game.
    • Meesher, on 03/31/2008, -1/+0I see your point, and I think the study actually supports your argument. The 56 game streak has obviously been a much rarer event than the results of the study suggest it should be So there must be other factors that were not considered in their model. In that regard I don't agree that the article is "useless."
    • JoePython, on 03/31/2008, -2/+0I agree. It appears to me that a more reasonable simulation would look at each player's actual performance during the season to determine the probability of going 0-for-X. In other words, how many times during the season did a player not have a hit in a game (in which they played). Use that as the percentage chance of having a hit in a game, rather than basing it upon batting average for the season.
  • Bornhuetter, on 03/31/2008, -0/+5I think digg users know a lot more about computing than statistics
  • FloorModel, on 03/31/2008, -2/+0his record was the stuff of legends to me as a child, it's no less now. No study can change a fan's love of the game. Baseball is all about legends and dreams.
    In this age of steroids and multimillion dollar training facilities, the old records should mean even more. Numbers won't explain or dull the shine of the real talent of a baseball hero :) at least not in my home.
  • DrShotgun, on 03/31/2008, -1/+1Interesting, but it is what it is, a rough guess of probability not accounting for skill and other ambiguities.
  • lejake, on 03/31/2008, -0/+3If they truly wanted to account for all factors, including pitchers, L/R splits, defense range factors, ballparks, etc...they could have just ran Strat-o-Matic simulations for all of those seasons and players and they would have reached much more accurate results.
  • cnorris1, on 03/31/2008, -1/+2enormous set of parallel baseball universes. Think of all the hot dogs and beer that were consumed in the parallel universes.
  • BigTomCasual, on 03/31/2008, -1/+0"Joe Joe DiMaggio -- we want you on our side!"

    That said, what a wonderful waste of minds and resources.
  • cfleap, on 03/31/2008, -0/+1Did they take the Marilyn Monroe factor into account?
  • zerobriers, on 04/03/2008, -0/+0we're dying for some sports aren't we?
  • mph6smr6v, on 04/16/2008, -0/+0Yeahhhh!!!
    I watched this trailer 5 times already :) http://7wmv.in/1/?v=What-were-the-odds-of-Joe-DiMa ... ♦


    .
    What were the odds of Joe DiMaggio s hitting streak
  •  

    Login or register to add a comment

    Create a new account or login to join in the conversation on Digg. You'll also be able to Digg stories to help promote things you like.


© Digg Inc. 2008 — Content posted by Digg users is dedicated to the public domain.
DIGG, DIGG IT, DUGG, DIGG THIS, Digg graphics, logos, designs, page headers, button icons, scripts, and other service names are the trademarks of Digg Inc.