Skip to main content
ABC News
The Mystery Sabermetrics Still Can’t Solve

We did everything we could.

That was the sentiment coming from the Miami Marlins in the wake of this week’s devastating news that Jose Fernandez, the team’s brilliant 21-year-old ace, would likely be out for the remainder of the year with a torn elbow ligament requiring Tommy John surgery.

“We’ve protected him,” Marlins manager Mike Redmond said Monday. “We’ve been consistent in how we’ve used him, with his workload. We’ve given him extra days. [The injury is] a question I don’t think anybody has the answer to.”

Redmond is right. Fernandez hasn’t been abused in the traditional sabermetric sense of the term: He’s seldom been allowed to throw more than 100 pitches in his starts and has only crossed the 110-pitch threshold once. (He’s also never started on fewer than four days of rest.) Miami pulled the plug on Fernandez as soon as data on his fastball speed removed all doubt that something was wrong.

For all the progress made by science and mathematics in countless areas of baseball, the prediction and prevention of injuries — particularly those to pitchers like Fernandez — remain a frustrating mystery. In a game where everything is dissected with painstaking rigor, not even sabermetricians have been able to make much headway in reducing the rate at which pitchers get hurt. They’ve been at it for more than a decade, and they’re as stymied as the rest.

Fernandez is just the latest ace felled by an arm injury, leading Sports Illustrated’s Tom Verducci, for one, to say there’s an “epidemic” of lame arms sweeping the game. In addition to the Marlins’ phenom, former All-Star hurlers Matt Moore, Patrick Corbin and Josh Johnson were sidelined earlier this season with similar ailments, adding to a growing list of victims that’s included big-name pitchers such as Stephen Strasburg, Matt Harvey, Kris Medlen, Brandon Beachy, Scott Baker, A.J. Griffin and Jarrod Parker (among others) in recent seasons.

Bleacher Report injury expert Will Carroll reported last summer that a staggering third of all current major league pitchers have undergone Tommy John surgery at some point in their careers, and the procedure has been performed more than ever before in recent years.1 And the spate of pitching injuries this year is causing the sports-media equivalent of a moral panic. It’s only mid-May, yet professional baseball has already seen Tommy John surgery claim 31 pitchers,2 a pace that threatens the record of 65, set in 2012. The most common month of the season for the surgery — June — is yet to come.3

As Bill James, the godfather of sabermetrics, has repeatedly noted, starting pitcher injuries haven’t really increased in recent seasons, despite our perceptions of the contrary.4 But they aren’t decreasing, either.

Sabermetrics has been able to solve a lot of baseball’s mysteries, but it can’t quite crack the pitcher injury question. As recently as a decade ago, the prevailing hope was that hurlers would see a significant reduction in injuries as more teams implemented data-driven policies for monitoring their starting pitchers. Now that suggestion sounds like a hopelessly naive relic of a different time.

The most definitive early sabermetric attempt to solve the puzzle of injured pitchers was conducted at Baseball Prospectus in 1998 by Rany Jazayerli. His Pitcher Abuse Points (PAP) framework, which was later expanded upon by BP’s Keith Woolner, centered on a simple hypothesis: “Throwing is not dangerous to a pitcher’s arm. Throwing while tired is dangerous to a pitcher’s arm.” To quantify this effect, Jazayerli and Woolner set up a scale to separate ordinary starts5 from high pitch-count outings that put tremendous strain on the arm, with a stress factor that compounds as more pitches are thrown.

Whether MLB front offices were directly influenced by Jazayerli’s work or not, the underlying concept behind PAP has had a profound impact on the way teams treat their pitching staffs. For example, the 130-pitch outing — once a staple for major league starters — has all but disappeared from the game over the last decade-plus. Clearly, the behavior of teams has changed when it comes to their prized investments on the mound, but it’s not clear this change in attitude has translated to greater pitcher durability.

Relatively recently, both Jazayerli and James offered some evidence that promising young pitchers are indeed burning out less frequently in the era of meticulous pitch-count management. But studies revisiting the predictive value of PAP itself, with the benefit of a decade’s worth of new data, have been inconclusive. In “Extra Innings: More Baseball Between the Numbers”, Corey Dawkins reported a failure to find any significant correlation between PAP and either short- or long-term future pitching injuries, while BP’s Russell Carleton could only find compelling support for the notion that eliminating 130-pitch starts reduces the probability of future injuries.

Of course, pitch counts are a blunt instrument at best. They represented the most granular data available to Jazayerli and his Prospectus comrades in the late 1990s and early 2000s, but the installation of Sportvision’s PITCHf/x cameras in every MLB stadium during the mid-to-late 2000s promised much greater things. PITCHf/x can measure an array of details about any given pitch, including its velocity, movement, spin and release point. Armed with better tools, a new generation of analysts were eager to find correlations between these state-of-the-art data points and pitcher injuries.

The newer, more data-rich models tended to settle on two factors as important predictors of a future injury: sudden, unexplained declines in a pitcher’s fastball velocity and increased variation in his release point. Not coincidentally, both elements aligned with existing biomechanical theories about the effects of fatigue and what puts pitchers at risk, and fit in with prior research like PAP. (As the hypothesis goes, tired pitchers — like those who have accumulated more abuse points — find it harder to maintain consistency in their release points.)

But there’s hardly a consensus among analysts that PITCHf/x is providing useful data.6 Again writing in “Extra Innings,” Dawkins expresses a major concern about any findings that rely on PITCHf/x’s release-point data because there’s a large degree of measurement error inherent to the system. Attempts have been made to adjust for these kinds of inaccuracies, but only so much can be done about the fact that PITCHf/x is unable to precisely track balls before they cross a threshold 50 feet away from home plate. And after analyzing the data himself, Dawkins was unable to replicate a statistically significant relationship between velocity or release-point data and future injury risk.

As BP’s Carleton found in a separate analysis, perhaps the most significant predictor of a pitcher getting injured in the future is simply whether he was injured in the past. On its face, that’s not especially useful,7 but it plays into another old hypothesis that might be worthy of new life: the “injury nexus,” formulated by Carroll and FiveThirtyEight editor-in-chief Nate Silver way back in 2003. The idea of the injury nexus is that pitchers are especially vulnerable to catastrophic damage during a very specific, formative time in their careers, before they hit the age of 24. One big medical theory as to why: Pitchers’ bodies aren’t completely mature until roughly that age, putting them at greater physical risk of injury.

But there’s also the possibility that survivorship bias is at work.8 Pitchers who have bad mechanics (or any other flaw that would put them at greater risk of seeing their careers end early via injury) are automatically weeded out of baseball at a young age, leaving behind only the group of pitchers who made it through that initial checkpoint.

If a pitcher like Fernandez was going to get hurt, it would be at this stage of his career, when the largest possible pool of pitchers is facing its first test of injury resistance. Keep pitching without incident, though, and Fernandez’s odds of future ailments would be significantly reduced, simply because the biggest test — whether he can handle a major league workload — would already have been passed.

Of course, that’s just one theory. But when it comes to predicting pitcher injuries, theories are still all anybody has, despite mountains of research and nearly two decades of work from baseball’s brightest minds. And while they test those, Jose Fernandez will be recuperating somewhere, just as Harvey, Strasburg, Medlen and others before him.


  1. It should be noted that it’s unclear how much of the upward trend is because of increased injuries and how much is due to a combination of better diagnostic tools and the procedure becoming more widespread.

  2. According to data compiled by Jon Roegele and Jeff and Darrell Zimmerman.

  3. January is the most common overall month for TJ surgery among pitchers.

  4. Back in March, James wrote:

    “The durability of starting pitchers today is essentially in line with historic norms, but has trended downward slightly with the end of the steroid era. The durability of starting pitchers right now is essentially the same as it was in 1967, less than it was in the 1970s, and very slightly less than it was in the heart of the steroid era. The durability of starting pitchers now is distinctly greater than it was in the 1940s and 1950s.”

  5. That is to say, those under 100 pitches.

  6. From a community knowledge perspective, it doesn’t help matters that some of the best practitioners of PITCHf/x analytics have since been swallowed up by major league front offices, their research never to be seen by the public again.

  7. Except to prove the old axiom true: “There are two types of pitchers — those who are injured now, and those will be injured soon.”

  8. As was acknowledged by the authors at the time.

Neil Paine was the acting sports editor at FiveThirtyEight.