Statistics seem to be more prevalent in baseball than in other sports. Sabermetrics or even the multitude of traditional statistics are more widely used in baseball than in soccer, football, or basketball. Where else do you see statisticians drooling over a player’s wins above replacement or measuring a player’s singular impact on the game the way fielding independent pitching does. I sought out a statistician’s perspective as to why statistics are so important in baseball. Dr. Allan Rossman is the head of the statistics department at California Polytechnic State University San Luis Obispo and I got the opportunity to pick his statistical mind and ask him some questions about the prevalence of statistics in baseball.
Q: How are statistics important in analyzing performance?
A: From a fan’s perspective it’s interesting and fun to compare players across era’s. Of course nowadays there’s more complicated statistics such as park adjustments and era adjustments. There are many adjustments which go beyond batting average and earned run average. I think from the business perspective it’s much more important and serious. That’s sort of what moneyball was about. Finding players who had skills that were undervalued in the marketplace. So for instance on base percentage (OBP) wasn’t all that well-known or all that thought of 10 or 15 years ago. The challenge now is what are the new ways to measure performance that other teams don’t know about so that team’s can acquire players at an affordable rate.
Q: If you were to go about creating a model for a player. In very general terms how do you make a prediction
the hard part is defining what is the variable you want to measure. how are you going to measure the performance of the player. whether it’s with an old-fashioned metric like batting average or with a newer metric like wins above replacement. That’s the really hard part. To some extent that’s not a statistical issue. The actual statistical modeling I think is the easy part. Where you look at other factors such as how he did in college or the minors and predict how he will do at the major league level. The hard part it seems is how you are going to measure value.
Q: Is there a sub-field emerging within statistics where statisticians do not just delve into mathematics but also into athletics and player performance?
A: I don’t know if I would call it a sub-field but there are more and more statisticians interested in that. There are even journals being developed. There’s a journal called the journal for quantitative analysis in sports. There are yearly conferences in sports which examine quantitative analysis of sports.
Q: In terms of fantasy sports, do you feel that you have a statistical advantage over other players of this statistics based game.
A: Not particularly, the only place where I might have an advantage over a less statistically minded person is that I don’t believe in things like hot hand or momentum. I think I might be more cognizant of the fact that somebody who has a really good april won’t have as good of a May or June simply by something called the regression effect. It’s possible that my statistical thinking helps me there.
Q: Do you think statistics will shift towards more comprehensive metrics such as batting average on balls in play (BABIP) or wins above replacement (WAR)?
A: I think so. Those metrics give a better approach.
Q: What can you tell us about the regression effect?
A: Regression effect was first encountered in measuring the heights of men and their sons. Very tall men tend to have sons that are very tall but shorter than their dad. On the other end of the spectrum, Very short men tended to have sons that were slightly taller than they were. Thus, a player with a hot April may not continue on that same spectrum many months down the road. And vice-versa a player with a very poor month will likely not continue on that same poor trend in the coming months.
How are statistics affecting youth leagues?
After conducting this interview, I decided to get a player/coaches perspective. I spoke with a few other players including Cole Walsh. Cole played four years of baseball at Palos Verdes High School and is currently employed at Spring Training youth baseball camps. I asked Cole what he thought the impact of statistics are on youth baseball. “I think the very serious players begin playing for club teams at a young age and that’s where you see statistics emerging as a way for the best teams to separate themselves. Club teams are beginning to scout other teams ahead of upcoming tournaments. I haven’t seen any major statistical analysis yet simply because it requires a lot of effort to accumulate a large amount of data required for statistics such as batting average or other averages which pro players are associated with. However, I think that eventually even at the youth level people are going to start trying to put together something along those lines,” Cole said. What he says is spot on. It’s easy for pro and minor league players to accumulate stats because they have so many people watching them, but even at the high school level there are plenty of statistics available. Maxpreps.com is a website where all high school athletes statistics can be made available and has become a valuable tool for high school athletes looking to play at the next level. It is only a matter of time before we see a trickle down effect into youth baseball with statistical data.