Baseball Musings
Baseball Musings
May 30, 2004
Psychological Pitchers

Back in February, I noted an article about the Orioles using psychological tests to help determine who would make good pitchers, and what in what roles those pitchers would be most comfortable. Almost two months into the season, I thought it would be interesting to go back and revisit that story. Unfortunately, the Baltimore Sun link has expired, and you have to pay to get to the story. Nonetheless, so far, I'd say this method isn't working. The Orioles are last in AL in ERA, and are trailed only by the Colorado Rockies. Now, it could be that this method is great, and the Orioles haven't had time to implement it fully. But from the original story, they have used it some over the years. In fact they used it to make Arthur Rhodes a set up man.

Well, Rhodes was moved into the closer role with the A's this year. He's doing okay. He's saved 8 of 11, which is okay. His ERA is about the same as last year. He's striking out more batters, but he's also walking more and allowing more HR. But overall, he hasn't been a disaster as a closer.

So far, the scorecard has the Orioles psychology 0-2. When you introduce a new method, it's not good for it to be a miserable failure in it's first season. It tends to make others think it doesn't work.

In their 1991 Major League Handbook, STATS, Inc. first offered projections for the next season based on Bill James' work, including projections for minor league players who might make the big team in 1991. When Peter Gammons reviewed the book, he noted that STATS had predicted that Jeff Bagwell would win the NL batting title in 1991. Bagwell didn't win the batting title, but he had a great rookie season. As Bill James told me later, if Bagwell hadn't played well in 1991, no one would have believed in the validity of the system.

So my guess is a lot of baseball people are going to be pooh-poohing the Orioles use of these tests. That's too bad, because I believe it's going to take a while to see if these really work.


Posted by David Pinto at 02:52 PM | Management | TrackBack (0)
Comments

Of course. The problem could be with the method (a faulty test) rather than the system itself.

And there are bound to be a lot more variables than just psychological makeup whether pitchers perform well.

Posted by: Jurgen at May 30, 2004 03:08 PM

Just to clarify, the first STATS red book didn't actually predict that Bagwell would win the batting title. It included projections for most MLB hitters and then included major league equivalencies for the prior season for 25 or 30 prominent minor leagues including Bagwell. Gammons treated the two different sets of numbers as the same thing, noticed that Bagwell had the highest average in his league and then publicized it.

Posted by: Michael at June 1, 2004 09:15 AM

Psychological tests are probably great scouting tools but I think they also have to be used in lockstep with more traditional physical testing, can the guy throw the ball over the plate, etc.

Posted by: steve at June 1, 2004 11:25 AM