Hamate fractures and power in the PITCHf/x era

Hamate bone fractures are one of the more common upper extremity injuries seen in baseball, not only due to the abundance of occasional wayward pitches or foul tips striking a glancing blow to the area, but also due to where hitters position their hands on the bat, which can introduce repetitive, high compression forces on the bone during a swing, but in particular, during a checked swing. Along with the forces and motions mentioned, the hook-like protrusion of the bone—also called the hamulus or hook of hamate— and certain vascular features make the hamate (lettered 'H' in the image below) predisposed to injury.

Image courtesy of South Side Sox.

Once diagnosed, treatment of the fracture can be approached either with immobilization or surgery. Due to a number of variables, which include misdiagnosis and poor compliance with immobilization, surgery to remove the fractured bone fragment or the entire hamate is more frequently recommended. Recovery from surgery is relatively short, requiring only four to six weeks for a player to return, with pain and sensitivity of the scar being a big hurdle to how long recovery and rehabilitation takes.

While the road to returning back to action after a hamate fracture is not as fraught with hurdles as some other injuries, it does come with baggage, as the question of whether the oft-seen lack of power found postoperatively is permanent is posed.

Is the power outage real? Is it permanent? Can we see changes to a hitter's power in their PITCHf/x data?

Let's try to answer some of these questions, but first let's discuss the materials and methods of this exercise. First, data was collected from 2007 through this season on hitters who were reported to have sustained a hamate fracture at the MLB level from a number of sources, with this article and internet searches being the main tools of data collection. Fractures and surgery dates were confirmed via Baseball Prospectus player cards and additional internet searches. The mechanism of injury for each player was also confirmed when possible, again using BP player cards and internet searches.This gives us the following players, season of injury, and mechanism of injury:

Name	Season	Mechanism of Injury
Carlos Gomez	2007	Swing
Dustin Pedroia	2007	Unknown**
Willy Aybar	2007	Swing
Erick Aybar	2007	Unknown**
Ryan Zimmerman	2007	Swing
Joe Mather	2008	Swing
Tony Pena, Jr	2009	Swing
Everth Cabrera*	2009	Swing
Chris Dickerson	2010	Swing
Matt Antonelli	2010	Unknown**
Troy Tulowitzki	2010	HBP
Tony Gwynn, Jr	2010	Swing
Danny Espinosa	2010	Unknown**
Domonic Brown	2011	Swing
Michael Brantley	2011	Unknown**
Pablo Sandoval	2011	Swing
Nick Markakis	2012	Slide, Swing
Pablo Sandoval	2012	Swing
Gordon Beckham	2013	Swing
Mike Zunino	2013	Foul tip
Wilson Ramos	2014	Foul tip

*Cabrera suffered additional fracture in opposite hand in 2011 while in minors.

**For Pedroia and Brantley, mechanism unknown because player didn't know when he was originally hurt. Mechanisms for Espinosa and Aybar could not be confirmed.

The main mechanism of hamate injury is attributed to swinging a bat in our sample of 21 fractures across 20 players, at 62 percent; Pablo Sandoval shares the distinction with Everth Cabrera of suffering hamate fractures to both hands, with Cabrera suffering his breaks in the major and minor leagues. The reason for this dubious honor arises from both players being switch hitters and sustaining fractures from both sides of the plate. Three other switch hitters show up in our sample, with both Aybar brothers and Danny Espinosa all suffering fractures, but thankfully, only once.

Moving forward with performance data, for each of the above players, game logs were collected for each player and categorized into pre-surgery and post-surgery performances and the statistic isolated power (ISO) is the measure of power, calculated as follows:

ISO = ((2B) + (2*3B) + (3*HR)) / AB

Players who accumulated at least 550 plate appearances before and after their surgeries were then selected, due to sample size considerations. Switch hitters were also removed from the analysis and will be discussed separately at a later time due to concerns regarding sample sizes from each side of the plate and any potential confounding arising from having two separate swings to maintain. With these additional criteria satisfied, here are the remaining players and their ISO numbers—after a significant number of plate appearances after their surgery, did they suffer a huge drop in power?

	ISO, Pre	ISO, Post	PA, Pre/Post
Brantley	0.092	0.116	942 / 1330
Gwynn	0.069	0.077	1042 / 668
Markakis	0.157	0.097	4310 / 1057
Pedroia	0.120	0.151	679 / 3989
Tulowitzki	0.188	0.246	2104 / 1692
Zimmerman	0.185	0.191	1466 / 3514
Average	0.135	0.146	--
SD	0.049	0.064	--

Aside from Nick Markakis, it appears our sample did not suffer a power drop on average, post-hamate surgery. With this caveat, it should also be noted that of our group, Markakis is the oldest, breaking his hamate in his age-28 season. If fact, their ISO increased roughly eight percent. However, this result did not attain statistical significance (p=0.26); considering the small sample size and the resultant effect size calculation (Cohen's d of -0.280), the use of the t-test is probably not recommended and additional data points should be gathered before leaning on any comments about significance of these differences.

Let's take a look at how each of our six hitters' ISO evolved over time as they continued to come back from the injury. Here, I have plotted their pre-surgical average ISO's as dotted lines for reference, for a better appreciation of when each player's power stabilizes and returns (or moves past) their pre-injury values:

Overall, we see a good amount of individuality to each player's return from hamate surgery, but a very rough estimation puts these six at around a full season's worth of games to be played before power numbers begin to stabilize after surgery, which isn't that surprising a result. However, the return of an increase in power post fracture compared to pre-injury value is intriguing.

Let's now move on to some PITCHf/x data for our data set and see if we can note any differences in that data. On the left is pre-injury ISO, with post-surgery heatmaps on the right.

Brantley:

Gwynn:

Markakis:

Pedroia:

Tulowitzki:

Zimmerman:

It appears to be a mixed bag as far as how each player's ISO hot spots evolve, with some obvious, across-the-board declines, such as Markakis, and some more subtle changes in spite of an overall increase in power after hamate surgery. For example, post-surgery Troy Tulowitzki is still hitting the ball as hard as he ever has, but with time, he seems to be a little more selective with what he swings at.

Overall, it appears that with time, player's who succumb to a hamate injury will find themselves back to previous power numbers upon their return to play, though it will probably take roughly a full season of at bats to do so. Of course a number of other factors could predispose a player to not bounce back quite as quickly or as robustly—age of injury, previous injury history, and even the time it took for correct diagnosis and treatment approach all play potential roles—but compared to other upper extremity injuries, the outlook for a full return to health and hitting power with the hamate fracture is generally promising.

***

Data courtesy of FanGraphs and Brooks Baseball unless otherwise noted.

Stuart Wallace is an associate managing editor and writer at Beyond The Box Score. You can follow him on Twitter at @TClippardsSpecs.