Great stuff from Bill James
(no surprise there) at MLB.com. He recounts how his affinity for baseball statistics kept him from becoming an economics professor, and how difficult it was for the truths he uncovered to gain acceptance.
“When you first discover something like this (that lefty pitchers are harder to steal on than righties) and you print it, you first think, or at least I did, that the whole world is going to be aware of this now and that people will stop saying that left-handed pitchers are easier to run on than right-handers,” said James.
“You quickly discover that nobody is paying that much attention, and that you can demonstrate that proposition A is clearly false and people will continue to assert proposition A for the next 100 years anyway. So repeating that experience a few hundred times, by the early 1980s, I had come to think of that as just the way the world was and perhaps was not the first person to realize that you actually do change opinions, it just takes a long time.”
Well, I did make professor status, and I share James’ bemusement at how my work sits ignored on library shelves. While I’m sure that my research will never change the world in the way James’ has, you never know!
James also says this, when asked about what statistics he uses to evaluate players:
“Well, I think the more critical question is what do you look at second. I think the things I look at first are the same things everybody else does. Won-loss record and ERA for a pitcher and home runs, RBIs and batting average for a batter,” said James. “Those are the first things you see and the first things you look at. The real question is what do you look at second.”
“The second thing I look at for pitchers is strikeout-to-walk ratios,” said James. “It’s the most individual of the things a pitcher does.”
I think its time I introduced some of you to fellow economist Gerald Scully, who discovered the importance of the strikeout-to-walk ratio in a 1974 paper published in the American Economic Review. This is one of the great papers in economics of the past 30 years. Scully used bivariate regressions (it was early days in computing) to measure player productivity. He found that the strikeout-to-walk ratio was the best pitching metric for predicting wins, and hence used it without any fuss. The object of the exercise was to estimate the difference between player pay and player value under the reserve clause system. He found that pay was about 1/8 of value. What was great about the paper was that it predicted what would happen when out of the blue, two years later, the arbitrator’s decision in the Messersmith-McNally case opened the floodgates to free agency. Salaries of free agents quickly rose 700%, implying that Gerald Scully’s estimates were in the ballpark.
It is rare to see an economics paper with such a startling proposition as Scully’s – players are underpaid by that much? – confirmed so quickly and conclusively.