Thursday, May 25, 2006

Sample Size and how the media uses them

There has been a lot of talk and print about Alex Rodriguez's disappointing numbers in the clutch this year. The New York newspapers are blasting him, the talk show hosts are poking fun at him. They are comparing ARod's numbers with Jeter's numbers in the clutch. What I find hilarious and stupid at the same time - is that they are looking at a sample size of 26 and 20 at-bats respectively! This is in a sport where one gets 500 to 600 at-bats per season. 20 at-bats is not very indicative of actual talent, be it in the 1st inning or in clutch situations.

Smart bettors find these type of situations and bet against the media. They know that the average reader is following the media and drinking the kool-aid with them. So it is possible that the line is shaded in that direction. If that is the case, then it makes for a solid bet against the media and their followers. Unfortunatley in the A-Rod clutch hitting situation - there really isn't a good way to bet on or against A-Rod & the media. But keep stuff like this in mind. The media will no doubt use small sample sizes as an indication of true talent when there are much bigger sample sizes to use (such as 26 at-bats vs 3 years of 500 at-bats each)...and when they do it in a way where it is bettable, then that's where one can pounce on them. There are situations where the sample sizes will never be large - and in those cases, it is not as clear. I'm talking about the situations when there is a much bigger population to draw conclusions from, but people are focusing on very narrow situations and trying to put a lable on someone based on that.

No comments: