The quality of pitching in Major League Baseball has never been better. From the major leagues down to the lowest levels of Rookie ball in the minors, pitchers are developing velocity at unprecedented levels and strikeout rates continue to reach new heights.
Where teams haven't been as successful is keeping those pitchers healthy. Major league aces, top prospects, first-round draft picks, international players and even high school underclassmen have had their careers derailed by Tommy John surgeries.
For a while, much of the focus on keeping pitchers healthy centered around workload. Pitch counts are an easy target and part of the puzzle, offering objective, numerical data that's easy to track and a variable teams can manage. Teams have poured time and money into mechanical analysis—particularly a pitcher's arm action—to try to unlock any patterns of what leads to a durable pitcher. All sorts of companies have popped up with different wearable gadgets for sale focusing, naturally, on the elbow itself.
Dr. Phil Wagner of California-based Sparta Science, who spoke at the MIT Sloan Sports Analytics Conference in March, wants to look at the problem from a different angle—one that might seem counterintuitive.
"What if," Wagner said, "you could predict an injury from a vertical jump?"
Wagner isn't talking about just knowing whether an athlete is at risk for pulling a hamstring, straining a groin or even an injury higher up like an oblique strain. He believes the data obtained from a vertical jump can detect when a pitcher is at higher risk for a UCL tear.
"People are looking up the chain, which is valuable, but it doesn't start there," Wagner said. "If (the lower body) is off, what you do up the chain doesn't matter. You can do all the rotator cuff and UCL strengthening in the world, but if you're not setting up right with your lower body, there's not going to be a lot of longevity."
Major league teams are buying in to Wagner's approach. Sparta Science already has worked with the Rockies the past three and a half years and just started its first season with the Padres, joining the NFL's Atlanta Falcons and Jacksonville Jaguars and NBA's Cleveland Cavaliers and Golden State Warriors as Sparta Science clients. They also worked individually with Padres righthander Tyson Ross, a pitcher whose durability has concerned scouts since the Athletics drafted him in 2008 but who just rattled off two straight seasons of 30-plus starts with combined 3.03 ERA.
Sparta gathers its data from an athlete jumping off a force plate, a portable 40-pound device that the Rockies and Padres can bring to all of their minor league affiliates and use for amateur players they're looking at in the draft and in Latin America.
"Why are we jumping? The answer is reliability," Wagner said. "You need an assessment that has good data. We looked at a bunch of other tests—sprinting, agility, throwing, hitting, jumping off one leg. The reality is those tests aren't as reliable. What we have to look at is what variables are reliable, specifically in the jump."
The vertical jump off the force plate spits out data into what Sparta calls a movement signature that spotlights three quantifiable variables: load, explode and drive. An athlete's load—how well he goes from standing into the squatting position—tells how quickly he can develop force. NFL linemen and major league catchers often score well in this category. The explode variable—how quickly he comes out of the bottom position—explains how well an athlete transfers force. Basketball and soccer players tend to fare well here.
The third variable is drive, which shows how long an athlete can produce force. Sparta found that high rotational athletes such as quarterbacks, golfers and pitchers tend to score well in drive. But if a pitcher has a high drive while his load variable lags, Wagner said, it can be a dangerous warning signal. "As that drive goes up, without load, it's predictive of a severe elbow injury," Wagner said.
Why would there be a statistical association between elbow injuries and a simple movement like a vertical jump?
"We know for sure that it's a highly predictive model for UCLs," Wagner said. "What we don't know is why. That's always hard to prove. From a research standpoint, that's part of the discussion, is why things happen. It's a conjecture of hypothesis. The theory is that guys who throw, when they don't load, they just kind of drift and their arm angle changes and drops. So rather than using their strength in their legs—that load—they use momentum."
In that context, the idea that a simple vertical jump could potentially forecast when a pitcher is at elevated risk of an elbow injury doesn't seem quite as far-fetched. It's completely in line with what many pitching coaches believe about pitchers putting more stress on their arms if they rely too much on their upper body instead of getting more leg drive into their mechanics. "That loading portion is similar to that pitcher loading that back leg," Wagner said, "and the drive is similar to his finish over the front side."
It also aligns with how teams handle their most vulnerable pitchers—the weak, skinny 16-year-olds they sign out of Latin America. Teams handle those gangly pitchers with extreme caution, focusing on adding muscle and strengthening their legs to build their durability while keeping a tight cap on their workloads.
For the Rockies and Padres, they have all of an athlete's Sparta's information centralized into a dashboard that they can manage. The dashboard includes other data, including a player's previous injuries and pitch counts thrown in the last 30 days that light up red if it's more than two standard deviations above his normal pitch count and other factors like nutrition, hydration and sleep that highlight when an athlete is at risk.
By the time a pitching coach or a quantitative analyst in the front office armed with Pitch F/x or StatCast data notices that a pitcher's lower release point might indicate an underlying health issue, it might be too late, with the damage already done to the elbow. But if a team sees that a pitcher's load variable is too low, the club can prescribe workouts tailored to enhance that value and keep him off the mound until that risk factor is alleviated, especially with a prospect where the team doesn't need to have them contributing immediately.
Knowing when a pitcher is at risk and being able to alleviate that risk could potentially keep him away from the disabled list and the surgeon's table. Or maybe the pitcher will blow out his arm anyway. But if a preventative, data-driven model can help keep a pitcher healthy an extra few years, the value to the player and to the team is obvious.