Wednesday, March 18, 2009

Wednesday Math, Vol. 64: Statistical correlation and Earl Bleeping Weaver

Baseball season is almost upon us, so this week's math post will combine math and baseball. Some people have said that baseball is merely an excuse to collect data and anecdotes. I think there's a little more to the game than that, but I do love the data and the anecdotes, and will be relying on both in this post.

The Society of American Baseball Research, or SABR for short, loves to collect and sift through the data produced by the national pastime. Many years ago, I was reading Bill James, probably the most famous member of the society, and he said that the formula above had a very strong correlation. If you take the runs scored by a team and square them and do the same with the runs allowed, runs scored squared divided by the sum of the squares of runs scored and runs allowed is usually very close to the winning percentage of the team for a season. The reason for general correlation is obvious, since you win games by scoring runs and lose games by giving runs up, but why you should have to square the totals before adding them is not as clear. Could the correlation be better if we raised the number to the power of 1.8 or 2.1? Nobody knows for sure.

The same methods for predicting winning percentages using points scored and points allowed in other sports are not as good at predicting. In football, there are too few games. In basketball, they score too many points per game. Maybe another power might work better in basketball.

While the formula is a good enough predictor, Bill James noted that one manager consistently, year in and year out, outperformed the predictions, and that was Earl Bleeping Weaver. (His nickname was not Bleeping, but my mom reads the blog, so I expect my readers will figure out the correct nickname without too much prodding. As Earl Bleeping Weaver would say, get a bleeping clue.) Whatever record the formula would predict, the Baltimore Orioles under the guidance of Earl Bleeping Weaver would win five or eight or ten games more than the prediction. No other manager outperformed the prediction as well or as often as Earl Bleeping Weaver.

I do not know if Weaver ever knew about this statistic. I'm sure if he did, Earl Bleeping Weaver would have had something to say about it, and it likely would have included profanity. Some people include profanity in their conversation for emphasis. In the case of Earl Bleeping Weaver, it appears to have been more like a respiratory function, nearly essential to his survival.

Whatever secrets he knew about baseball, Weaver kept his cards close to his bleeping vest. "Success in baseball is pitching, defense and bleeping three run homers." he was quoted as saying. We mere mortals are left to ponder the depth of Weaver's bleeping wisdom from such small pearls as these.

Not everyone appreciated Weaver's bleeping colorful language, not that Weaver gave a bloop what other people thought. When he was managing, born again Christianity was making its way into baseball clubhouses, with the devout holding prayer meetings and proselytizing where they could. One of the Christians quietly confronted the manager about his bleeping use of profanity one day.

Devout Christian Ballplayer: Skip, don't you want me to walk in the ways of the Lord?
Earl Bleeping Weaver: I want you to walk with the bleeping bases loaded!

Here endeth the bleeping lesson.