from Amazon
Tuesday, December 06, 2011
Increase in pitcher workload
Jason gives us some great data.
Our wonderful stats team pulled a report showing instances of a pitcher throwing at least 750 more pitches than the season before and found 1112 instances of pitchers that qualified since 1988, which was the first season that full pitch count data was available on the pull. As a whole, there was an average of 865 pitches thrown in the base season with 2205 thrown in the following season—an increase of 61 percent. The breakdown of the metrics for each season are represented in the table below.
And he shows that the FIP for the group in year 1 goes from 3.53 to 4.19 in year 2.
Except… there’s no control group. If you had a group of pitchers that had a 3.53 FIP in one year, you’d expect to see a higher FIP in the next year. That’s because a low FIP would imply more good luck than bad luck. And since luck is not persistent, then some of that good luck will go away.
Let’s try to figure out what we’d expect using regression. Jason noted 865 pitches, which would imply about 234 plate appearances. For a FIP regression, we’d probably want to add about 300 PA. So, we’d regress about 55%. The league FIP is probably around 4.4 for this time period. So, regressing 3.53 55% toward 4.4 gives us an estimate of 4.00 true talent FIP.
Jason is showing 4.19, so it’s still above what we’d expect.
So, there might be somethign to it, but we’d really need to look at it more carefully. I also didn’t see any minimum threshold settings either. So a pitcher that threw 20 innings one year and 120 innings another year would seem to qualify for the study.
I also didn’t see any control for reliever-starter, where you’d naturally get a jump, on the change of role alone.
Anyway, good start, but we need more.
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date