2012 Starting Pitcher DL Projections
A pitcher’s fantasy value is more than ERA, WHIP or K/9. Even if they have great values in rate stats, they are useless if they aren’t pitching. Today I am going to give the chances for certain starting pitchers to end up on the disabled list in 2012.
Last year, I determined a formula to get the percentage chance a pitcher will end up on the DL. Historically, any starter that pitched 120 innings in the previous season has a 41% chance of ending up on the DL. So of a team’s 5 starters, 2 will spend time on the DL on average.
The analysis takes 3 factors into consideration: the pitcher’s age: how many starts they had over the past 3 years and over the past 3 season, did the pitcher spend any time on the DL during that season. Age is a factor because the older the pitcher the more likely they are to break down. The game starts looks to see if the pitcher has a history of making a full season worth of starts. Also, it helps take into account missed starts that did not require a trip to the DL. Tracking DL trips helps to determine if the pitcher has history of injuries. Basically here is how much each of the 3 factors change the % chance a pitcher will end up on the DL:
One year older = +1%
33 more game started = -3%
1 year of Injuries = +8%
At the end of the 2011, I went back and look at the projections and they held up decently.
I have finally gotten around to calculating the 2012 chances. Here are the 5 most and least likely pitchers to end up on the DL in 2012 (complete list on Google Docs):
| Name | DL Trips | GS | Age | DL Chance % |
| Bartolo Colon | 2 | 38 | 38 | 61.2% |
| Aaron Harang | 3 | 74 | 33 | 56.9% |
| Bruce Chen | 2 | 57 | 34 | 54.4% |
| Brandon McCarthy | 3 | 42 | 28 | 53.8% |
| Ted Lilly | 2 | 90 | 35 | 52.4% |
| Madison Bumgarner | 0 | 52 | 22 | 29.8% |
| Felix Hernandez | 0 | 101 | 25 | 28.9% |
| Rick Porcello | 0 | 89 | 23 | 27.8% |
| Clayton Kershaw | 0 | 95 | 23 | 27.3% |
| Trevor Cahill | 0 | 96 | 23 | 27.2% |
A difference of ~35 percentage points exists from the least likely pitcher to go on the DL to the most likely. With many fantasy season ruined because of too many pitchers on the DL, this data will hopefully allow an owner to know the injury risks when they draft a pitcher.
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
What about the things I see in other injury projections like slider % and innings increase?
Slider % is an issue. I will look more into it in a few days.
The innings increase/Verducci effect has be proven wrong 3 different times. B-Pro just looked again a month or two ago.
Thanks. Yeah, I’ve seen some pretty weak results on Verducci for the last couple years. Are there any teams that have realized it’s not a big deal and unleashed their young SPs more? I guess it would be hard to separate the bull-headed “old school” teams from the ones that actually studied the issue analytically.
Halladay is around 100 and Tommy Hanson is around 35. just doesnt pass the smell test. If anyone had to bet their life on only one staying healthy this season, there isn’t a person in the world who would pick Hanson, no? Maybe I’m reading these wrong.
i agree with your larger point but i don’t know what “halladay is around 100” means, he’s at 45 if you’re talking about the % chance of DL trip this is spitting out
you’re reading it wrong: Hanson is 35.8% and Halladay is 44.9% but your point is still valid. Then again, looking at formulas is probably always going to be fruitless if you pick out the freak of natures like Halladay.
I do agree that other factors should be considered like slider %. I do not think Pineda and Bumgarner should be two of the lowest risk pitchers out there for this reason. Pineda especially because if this included minor league DL trips he’d be seen as much more risky.
My thoughts exactly. When Bumgarner makes the top 5 expected healthiest SPs with a 32% slider usage rate, I’m inclined to think something’s missing from the model. Pretty neat otherwise, though.
Bumgarner calls it a cutter. I’m inclined to believe him.
Sorry my 100 for halladay was the rankings.
I put a lot of stock in a pitcher’s ability to maintain a heavy workload over the course of a 3 season period, without a significant increase (eg three straight seasons of 200 innings pitched). I also think it depends on the types of injuries one experience. A pitcher who was on the DL all of last season due to having Tommy John surgery is treated differently than Josh Johnson, I think. I think innings pitched makes more sense than using games started.
When I originally looked at the data, GS was better variable than IP. If I remember right TBF was better than IP.
It makes sense that TBF is better than IP. Workload should be measured in pitches, not outs. Bad pitchers face more baters in an inning, and often throw them more pitches, than good ones.
Have you tried restricting the sample to players who landed on the DL and then running the regression using the log of ‘days on DL’ as the y-variable? I assume just using ‘days on DL’ wouldn’t give meaningful results.
I would love to see some research done into DL rates for relievers vs starters and if anything they’re doing in particular (higher avg velocity, over use of hard thrown breaking pitches, more frequent usage without rest, greater pitch counts for starters, etc) could cause the difference. It seems to me that RP’s get hurt (but subsequently tend to bounce back) more frequently than starters but I don’t even know where to begin looking up starter vs reliever injury statistics.
No statistical amount of analysis is going to lead me to conclude that Halladay has a good chance of ending up on the DL.
Why some great pitchers like Halladay can log all of those innings year in and year out probably has more to do with his genes and his mechanics than any thing else. With these guys, you can chuck probability theories and data out of the window.
And I think the F/X Pitch data that is analyzed by a computer does need some tweaking before we can trust it’s reliability. The Bumgarner case is one. But I also see that Zach Britton–per F/X texasleaguers.com–didn’t throw any sinkers at all last year. His bread-and-butter pitch is the sinker, so something is wrong in how the data is analyzed.
Uh, “…this data…”? Come on, man.
Any thought to applying extra risk to younger pitchers who have not passed through the injury nexus? Like Bumgarner? General thoughts on the concept?
Where is Clay Buchholz, maybe I am just missing him???
According to this research, Brad Penny has a 46.3% chance of ending up on the MLB disabled list. Think I’ll take the under.
I appreciate the effort, but shouldn’t we all already know that older pitchers and pitchers with an injury history are more likely to get injured this year? Seems like common sense.
By the way, we can probably add Josh Johnson to the most likely list, probably with a likelihood of 80%. Would anyone bet on him starting 30 games?