J. Zimmerman over at fangraphs has started (yet another) a pretty interesting project... assigning a percentage chance that a given player will see DL time in the following year based on statistical factors including their age, games started in the past three years, DL trips in the past, etc.Starting Pitcher DL Projections (Part 2)
Obviously we're up near the limit of the value statistical analysis here, where things either become negligible or "duh" very quickly, but I found it interesting to see the criteria that he picked for this analysis.
Per his current model (which is likely to be adjusted as he takes more factors into account):
Pettite - 56.5%
DiceK - 55.1%
Hudson - 55.0%
Westbrook - 55.0%
Carpenter - 54.7%
Ubaldo - 30.0%
Porcello - 29.3%
Danks - 29.2%
Cahill - 29.0%
Kershaw - 27.3%
I'm very interested to see if there's some way he can't apply some sort of a multiplying factor to this based on pitch speed, throwing motion, etc.
Obviously if the difference between the Top 5 and Bottom 5 is only 25% we're not looking at a very valuable fantasy tool here (especially considering we're not even discussing how long
these DL stints will be), but I did find it interesting.
Using some kind of combination of this score (for more established pitchers) with the Verducci effect (for younger/developing pitchers) I'm sure I'll find at least a handful of pitchers to avoid for 2011 that I wouldn't have expected to see on my no-go list.