MASH Report: Hitter Injuries & Exit Velocity

The following is an experiment. I wanted to see if hitter exit velocity (EV) drops and by how much when a batter players through an injury. I am not sure before starting it where it would head and after a small stab at the data, I seemed to end up with more questions than answers.

The following may sound like a scientific study and in many ways, it is one. I am trying to take a theory and find a solution. Let me start with the theory.

First, injured players perform worse while playing through injury. I have tackled this idea as I looked back at the players’ stats the season before, of, and after the injury.

Talking to several hitters, they say they know when something is bugging them. They just can’t get hit the ball as hard as when they are healthy. Instead of trying to figure out if a player’s slump is from bad luck or injury, we have StatCast’s exit velocity to help us see if they are hitting the ball as hard.

So to see the results, I examined at hitters who were known to be playing through an injury of some kind. I limited it to 10 players right now because the data collection is time intensive at this point. Additionally, I quickly found some changes I would like to use in the future.

For each player I collected:

• The approximate date of injury.
• The average exit velocity for 2015 and 2016. One huge limitation now and for a few years to come is the lack of data. I am living on the tip of the spear looking at this data and I expect to get cut a few times.
• Additionally, I had a theory that a hurt player will see their exit velocity jump around more in their injured seasons than healthy. With this in mind, I collected the player’s standard deviation in their exit velocity.
• Finally, I plotted each player’s average exit velocity using the batted ball in question and the nine before and after. Using just the daily values made finding a trend too hard.

Here is the data on the players I collected.

 

Nick Markakis (article link)
Neck (surgery after 2014 season)
2015 Average EV: 87.3 mph
2015 Standard Deviation EV: 13.7 mph
2016 Average EV: 91.4 mph
2016 Standard Deviation EV: 11.8 mph

I will just start with the exception to the rest of the players, a player hurt and then healthy. Markakis declared he was finally healthy after his 2014 neck surgery. A steady improvement can be seen during the 2015 season. His 2016 season is one the most stable seasons I will examine today.

 

Andrew McCutchen (article link)
Thumb/Heel (early-June/early-July)
2015 Average EV: 90.1
2015 Standard Deviation EV: 13.1
2016 Average EV: 89.3
2016 Standard Deviation EV: 14.3

He was thought to be possibly hurt in 2015, so his yearly values then may be off, but his production in early July tanked which is in line with his injuries.

 

Adrian Gonzalez (article link)
Back (spring training to early May)
2015 Average EV: 88.2 mph
2015 Standard Deviation EV: 12.8
2016 Average EV: 87.8
2016 Standard Deviation EV: 13.3

Gonzalez definitely had an exit velocity drop in early May which then quickly came up which is in line with the article. Interesting note, Gonzalez’s exit velocity after the shot mentioned in the article was 88.2 mph which is exactly the same as his 2015 season value.

 

Freddie Freeman (article link)
Cracked Rib (early season) and Finger (August)
2015 Average EV: 89.5 mph
2015 Standard Deviation EV: 12.8
2016 Average EV: 90.1
2016 Standard Deviation EV: 14.0

The time frames are a little vague, but looks like the cracked rib was during the first couple months and the finger just started to really affect him in June.

 

Jason Heyward (linked article)
Wrist (early May), Hip (late May), and Side (late June)
2015 Average EV: 89.6
2015 Standard Deviation EV: 13.8
2016 Average EV: 86.2
2016 Standard Deviation EV: 15.3

If there is a poster child for what I would expect from a hitter playing through an injury, it is Heyward. His overall exit velocity is down from 2015. His exit velocity is all over the place in his injured season. When the injuries happened, there were at the same time he saw velocity drops.

 

Yoenis Cespedes (linked article)
Quadriceps (July 8th)
Late April injury
2015 Average EV: 91.1 mph
2015 Standard Deviation EV: 15.0
2016 Average EV: 91.4 mph
2016 Standard Deviation EV: 14.7

As nice as Heyward’s data fit my initial theory, Cespedes’s data is completely opposite.

 

Bryce Harper (linked articles)
Neck/Shoulder (mid-June)
2015 Average EV: 89.9
2015 Standard Deviation EV: 16.3
2016 Average EV: 88.7
2016 Standard Deviation EV: 14.7

Harper’s drop in exit velocity has already been covered by MLB.com’s Mike Petriello. A couple of notes. First, the SI article states the injury happened two months before (ran on August 12). I wonder if the source was a little off and the time frame was closer to a month and a half. Second, Harper had one the biggest downward spikes from April 23 to April 26. I could not find a cause, but he hit nine straight balls under 80 mph.

 

Elvis Andrus (article link)
Groin (Early May)
2015 Average EV: 86.5
2015 Standard Deviation EV: 13.2
2016 Average EV: 86.4
2016 Standard Deviation EV: 13.6

The source stated early May, but I wonder if Andrus’s injury was going before then and just got worse and worse.

 

Buster Posey (article link)
Thumb (early-June)

2015 Average EV: 88.4
2015 Standard Deviation EV: 13.2
2016 Average EV: 90.3
2016 Standard Deviation EV: 13.0

As a catcher, I would not be surprised if Posey is dealing with constant injuries, so I am not surprised to see his 2015 exit velocity higher than his 2016. One late, manager Bruce Bochy was concern Posey would lose some power from the injury:

“It may affect some power,” Bochy said. “If that hand is not right, you’re going to lose some power.”

 

Carlos Beltran (article link)
Knee (mid-June)
2015 Average EV: 90.5
2015 Standard Deviation EV: 13.2
2016 Average EV: 90.3
2016 Standard Deviation EV: 13.0

Beltran showed some drop near the injury, but not as big as the drop to start the season. I can’t draw any conclusions here.

 

I will say it was a nice start and here are some references I drew from the small sample set.

• The players’ average exit velocity dropped on average -.74 mph from the healthy to the hurt season (median = .30). The standard deviation change was .45 (median = .44). Some change is seen, but also this may be just the aging factor for exit velocity. I hope to run a basic EV aging curve once the season is over.
• There seems to be a huge drop in exit velocity in some cases which matches when the player was hurt. The drop is about 5 mph to 7 mph lower than the player’s average seasonal velocity. I will create a basic finder for this and report later on it.

That is it for partial answers. Now here are the questions which I need to explore.

• I would be great to have more data. Is Markakis back to previous exit velocity? The public will never know. Doing this experiment in a couple of years will be extremely informative.
• Even though I noticed large drops with injuries, the amount and rate of decline will need to be further measured.
• On issue I immediately found I didn’t like was the averaging value I did. I noticed in a few instances, the player’s exit velocity stayed high after the injury. I may make the averaging value a little more forward looking.
• I need more data. The above players were used because I was easily able to find a reference to the injury. Right now when I am setting the baseline values, I don’t want to use possible or rumored data.
• One other issue is the extent of the injury and the player’s production playing through it. If Bryce Harper is at 80%, his production is better than most other players, so the team and he will keep him playing through the injury. If a Quad-A player gets hurt in any way, they are likely replaced.
• Players can lie about their injury to help explain why they are sucking.
• All players are hurt in some way, especially catchers, so it is tough to get a good healthy baseline value.

I feel like I have some ideas to test and plan to come back to this subject next week with more players examined and few ideas tested. Until then, let me know if you have any ideas or theories on the subject.