“” Caught Looking: Batter Eye Movement, Arm and Elbow Injuries - Driveline Baseball

Caught Looking: Batter Eye Movement, Arm and Elbow Injuries

| Research
Reading Time: 8 minutes

Here are two studies the R&D department at Driveline read this week:


We’ve written a lot about the value of workload monitoring for pitching with more detail than throw or pitch count alone—including throw intent, throw distance, warm-up throws, throws in the field, and taking into account the accumulation of volume over some amount of time, among other things.

A lot of that conversation has been around the potential for tools like MotusTHROW to improve that process, but none of that data has directly highlighted the importance of incorporating more workload context than this paper. 

Shitara et al. recently published a prospective study of daily maximum intent throws and their relationship with elbow and shoulder injury incidence.

90 pitchers were separated into groups based on how many average daily throws they executed (lower volume group had 38 pitchers, higher volume group had 52 pitchers); the cutoff to separate the groups was determined using a receiver operating characteristics analysis (basically, they used a quantitative approach for the cutoff rather than just choosing a random number).

The authors found that:

  • There were 30 injuries in total
  • 17 of those were in the lower volume group (44% of the group), 13 in the higher volume group (25% of the group)
  • Median time-to-injury was 61 days for the low volume group and 106.5 days for the high volume group

Let me preface this discussion because there many factors going into injury.

Looking at the relationship between injury and one single measurement is a significant limitation in itself (and kind of the point of this summary, honestly). There is a lot of potential bias in this categorization of pitchers, including the fact that those who threw more pitches were likely more conditioned for throwing more pitches, which undoubtedly impacts injury risk. It is not to say that the authors of this piece thought this was the best measurement of workload, but this really highlights the use of more involved workload measurements.

It is clear with these results that counting pitches is limited when it comes to avoiding injury. Some things that could potentially improve upon this measure of workload:

  • Counting all throws instead of just maximum intent pitches, including warm-up throws, throws from the field, drill work, etc. 
  • Considering the accumulated long-term (chronic) workload for each player, individually
  • Use a measurement that doesn’t require the athletes to remember how many throws they completed each day

Lastly, this discussion is in the frame of avoiding injury, but think about the implications this can have for performance. If we can promote better recovery and put athletes in situations where they are more ready to perform as measured by their accumulated workload, they will absolutely perform better. 

This is not all for a sales pitch for MotusTHROW, but MotusTHROW happens to cover all of these bases. I hope that this interesting study and other workload management studies can continue making it clear that quantitative and contextual workload monitoring is a tool that can be used by the entire baseball industry to improve performance and mitigate injury risk. 


This study is a bit denser and uses more of a basic science approach to describing how we should approach youth overhand throwing training. It should be noted that this is very theoretical work. Part of this analysis gets pretty ‘noisy’ as far as some of the data modeling goes, but the idea is very interesting. 

Tsutsui et al. at Waseda University in Japan looked at the inertial properties (mass and size) of the forearm and how it develops over a young man’s growth compared to shoulder musculature. This is important because the large amount of stress placed on the elbow and the shoulder in the throw arise from accelerating the forearm, hand, and ball toward the throwing target.

If shoulder musculature development lags behind an increase in inertia (the size and weight of the object the shoulder needs to move) of the forearm, then that discrepancy in time could mean that the shoulder is unfit to withstand that increase in load until it catches up to the development of the forearm/hand mass.

This experiment found that with some modeling, the peak inertia value of the forearm and hand mass peaked about 4 months earlier than the peak shoulder musculature, which could theoretically mean that there is an average of about four months between the ages of 12 and 13 years old when these athletes are more at risk for a shoulder injury.

This timeline is created from a small group of 8 to 14-year-olds, so it cannot be generalized.

This analysis is limited—as far as application goes—because we cannot be sure that this difference in peak forearm inertia and peak shoulder musculature actually leads to an imbalance of loading the tissues in the arm.

We also don’t know that lean body mass in the shoulder explains one’s ability to create or withstand the forces during the throw. It is known that the inertia of the forearm and hand mass alone does not explain the load that is placed on the shoulder—how fast it is being moved and in what direction is important. Lastly, this sample only included athletes from the ages of 8 to 14 years old—which cannot explain the relationship between forearm/hand inertia development and shoulder musculature in older, more developed athletes.

That being said, this is an interesting way to look at the development of young athletes and highlights how we might want to pay more attention to the load they place on their bodies throughout these development years. Maybe when athletes are in this rapid-development phase of their life, we can emphasize workload monitoring and not overload their conditioning level.

Not to continue beating a dead horse about workload monitoring/management, but it’s clear it has a place in the future of baseball.


Vision research is back.

At The Ohio State University, Fogt et al. looked at eye and head movements of two different hitters using a pupil tracking system, an internal measurement unit (IMU), and an electromagnetic motion capture device. This article is also more on the research (less applied) side of things, but it adds to the evidence suggesting that the ball is tracked by hitters to some extent, at least by skilled hitters.In this study, they used the rotational movements of the eyes and head to compare the hitters’ gaze position to the ball’s position with respect to the hitter. They found that both hitters (former D3 college baseball players) followed similar head movement patterns throughout the pitch.

Both hitters:

  • Moved their head slightly upward for a moment
  • Then moved their head downward for about a half-second (450ms, 512ms)
  • Had their gaze position under the ball for about 450ms when the error (distance between their gaze position and the ball) grew significantly before the location of where bat-ball contact would be made

However, one hitter moved their eyes downward for the first 450ms while the other did not move their eyes much at all. Although as with much of the vision research in baseball thus far, it wasn’t collected during a live setting but rather against a pitching machine throwing tennis balls.

More research needs to be done comparing elite hitters to non-elite hitters to identify elite vision patterns, but even more research still needs to be done on how to train it.

The main takeaway from this paper is that head movement is important in tracking the baseball. There may be a differentiation in how different hitters use their eyes in combination with their head movement to pick up the most information about the pitch to make quality contact.

This week’s edition was heavy on the theory and research side of R&D, but this information can still be used long-term to improve how we approach training for baseball players. If you are looking for more research and development content, hop on over to the R&D Podcast on YouTube, where we discuss player development topics and tools that we are developing in-house. You can also catch us on all podcast listening platforms!

By Kyle Lindley (@kylelindley_)

Comment section

Add a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Your Cart
    Your cart is emptyReturn to Shop
      Calculate Shipping