By Eno Sarris
Two out of every five pitchers who pitched in the bigs last year will be headed to the disabled list this year. Look around your favorite rotation and try to decide who will go down. You've got your ideas.
Research has provided some guide posts in predicting pitcher injury, but those posts are as shaky as the arms they were built upon. Maybe a trip through a specific rotation would help us mine the research out there in a practical way. Maybe the Texas Rangers can help.
One particular starter on the Rangers is a virtual touchstone for research on pitcher injuries. We all love Yu Darvish, and he can make magic with his arm, but it's reasonable to worry about Darvish and his health. Why? Let's count the reasons.
The strongest piece of research shows that past disabled list stints predict future ones. Russell Carleton at Baseball Prospectus showed that you're ten times as likely to show an elbow injury this coming season if you had one last year, for example. Darvish went on the disabled list last year for what was called an upper back strain, but the trapezius muscle is basically part of your shoulder apparatus. Uh-oh.
This sort of finding, that past health results predict future ones, descends all the way to the smallest parts of the body. R.J. Anderson took at look at blisters, the flap of skin that can fell a giant, and found a similar result: past blisters, and perhaps sheer number of pitches, can help project a pitcher's future blisters. Darvish had blister problems last year and only eight pitchers threw more pitches last year than the Rangers' ace. Yikes.
If you ask Darvish, he has a couple different sliders. His classification systems only show one slider, but he throws it a ton. By PITCHf/x, no pitcher threw more sliders than the 1290 Darvish threw last year. Might not seem like such a big deal that almost a third of his pitches were sliders -- after all, the whiff rate on that pitch is high, and it's at least a big part of why he has such excellent strikeout numbers -- but there is research that suggests the slider is a bit tough on the arm. Jeff Zimmmerman found that high-slider pitchers went to the disabled list 46 percent of the time, a jump from the 39 percent the general population averages.
To make matters worse, the second-worst pitch for your arm might be the curveball. Darvish throws two curveballs, and those pitches, along with a four-seam and a cutter, make up most of his arsenal no matter what classification you use. Let's not forget to mention that a few organizations have banned the cutter -- or at least the 'baby slider' version of the cutter that usually goes three-plus mph slower than your four-seamer and features slider-type bite. Supposedly the baby slider kills your fastball velocity over time. That hasn't been shown in the numbers, but it is worth noting that Darvish has a cutter, it's his third-most-used pitch, and it goes three mph slower than his four-seamer. Maybe that's the other slider he was talking about. Maybe we should be worried.
We're not done yet. You see, Darvish doesn't feature the best control: Only six starters had a worse zone percentage than Darvish last year, and even after improving his control, he showed a worse-than-average walk rate. You know what's coming by now. Jeff Zimmerman found that pitchers that threw more strikes, hit the zone more often and walked fewer batters all had better health outcomes than pitchers with worse control. Billy Beane once said he thought guys with good control had good mechanics and therefore stayed healthier longer, and there's some evidence he was right.
There are a few last harbingers of doom for Darvish. Josh Kalk's "Injury Zone" research was picked up by Jeff Zimmerman, and it showed that late-game inconsistencies in release point, combined with velocity loss and a drop in zone percentage, could be predictive of injury in a specific pitcher. Darvish was fairly inconsistent all year with his release point (see his release points compared to Cliff Lee's below), and dropped in zone percentage and velocity at the end of the year. But he never had all three markers at the same time, maybe.
Now it's time to pull back on the throttle.
Yu Darvish is not necessarily going to go under the knife this year. And his health may have something to do with a completely different way to talk about best practices for pitchers: biomechanics. Kyle Boddy has built a lab for his Driveline Baseball academy, full of weighted balls, motion-capture cameras, and the type of muscle sensors you see pictured here.
At first, Boddy says they built the lab "in the hopes of quantifying the pitching delivery and predicting injuries to the arm." And though they learned something about injuries and can perhaps project injury rates for certain populations, he wanted to be clear: "We are a long way from predicting pitcher injuries."
That doesn't mean that they haven't learned that certain movements put more stress on the arm then others, and can lead to injuries. They type of pronation you typically do when you throw a changeup, for example, has been "linked to fewer elbow injuries and increased performance" in their lab. But that doesn't mean that the specific muscles the pitcher might want to use for that action are being activated. In other words, for each pitcher, there is probably a healthier way to throw their changeup, even if it's a pitch that might promote health.
And so the goals at Driveline Baseball have changed, and in that change we can find perhaps the future of dealing with pitcher injury. Instead of dealing with them en masse, or establishing "best practices" and then trying to mold pitchers into those buckets, their organization tries to, in a word, "tweak" the pitcher's own natural mechanics in order to put less stress on certain parts of their bodies. The emphasis is on using video and practice with weighted balls to slightly change small parts of their deliveries to make them more effective and healthy.
Back to Darvish. He does something in this area very well. Many pitchers try to move their front shoulder in tandem with their back shoulder -- imagine a straight line from shoulder to shoulder -- in order to get their throwing arm out from behind them. According to Boddy, pulling the the glove side and clearing it away from the body before the pitching arm comes through allows the pitching arm to "better apply force in a straight line." David Robertson might be best at this, but as you can see in this GIF comparing Robertson to Darvish, they are both good at clearing away the front arm.
So Darvish may have an advantage in his mechanics. And using things like zone percentage and walk rate as a proxy for mechanics is a problem, especially for a pitcher that's new to the league. Darvish had to deal with a new ball in his first season -- and a new strike zone. Both may have contributed to his command issues. There's evidence he was squeezed in the strike zone a bit his first year, for example.
Maybe his mechanics are okay! He's young! Since he's young, his DL projection -- without any inputs based on his pitching mix and control -- is only 37 percent. Maybe his elbow will survive the sliders.
From the perspective of the Rangers, it might not matter much. They know they're likely to lose 600 pitcher days to the disabled list on average -- if they're unlucky (like the Blue Jays) as much as 1200 days -- and that they have to plan for that sort of eventuality. That's just another reason why teams are reluctant to trade young pitching. The Rangers aren't just planning on Darvish, Matt Harrison, Martin Perez and Alexi Ogando to contribute in the starting rotation, and hoping for Derek Holland's quick return. They're signing Tommy Hanson, and keeping Nick Tepesch around. They're re-upping Colby Lewis just in case. They're keeping an eye on prospect Luke Jackson. And they're feeling lucky that that most of their guys are projected within shouting distance of being league average, and they're feeling good about their offense.
Since all the Rangers can really do is prepare, and teach their pitchers best practices, it makes sense to develop this sort of depth. And invest in good training practices and facilities.
What are you going to do? If you somehow developed a team full of change-up throwing control artists with perfect mechanics… one of them still might fall down the stairs and hurt his knee.
* * *
Eno Sarris writes about baseball at FanGraphs www.fangraphs.com most of the time. He also started BeerGraphs www.beergraphs.com for the beer nerds out there. He doesn't always play daily fantasy, but when he does, he plays it at DraftStreet.com.