Contestants Get Measurably Exhausted During the Home Run Derby

For years, analysts have talked about the “home run derby hangover”. This hangover manifests in derby contestants performing noticeably worse in the second half of the season. If you compare home runs per plate appearance from before and after the derby, the contestants nearly universally perform worse.

How can this be? Some might say that the players try to change their swing to win the derby. They swing for the fences – literally – to try to grab a derby championship belt and in the process ruin the beautiful swing that got them there in the first place.

The article linked above claims the homer run derby hangover is due to normal statistics. The players that overperformed in the first half of the season look really good on paper and so are selected for the derby. Then, in the second half they tend to perform worse which is actually more reflective of their true talent.

While there might be some element of truth to this, I claim there may be a third effect that lurks. Tiredness. Maybe players perform worse just because they get more tired throughout the season. And to boot, participating in the derby itself doesn’t help.

The correct way of testing this theory would be to grab the data from multiple seasons and see if home-runs per at-bat decreases for everyone throughout the season. This would be evidence of a “tiredness epidemic” as the season goes on.

But that’s boring. We’re going to look for evidence of home run induced tiredness by looking at the derby data itself. Do the contestants get measurably more tired as the derby drags on? Is the derby really that exhausting? Luckily, the statcast data from recent years helps us answer this question.

To receive email updates when new articles are posted, use the subscription form below!

Statcast Data

Old school MLB data was all box score counting statistics. Runs, hits, walks, strikeouts, home runs, and on, and on. These things are valuable and do in fact tell most of the story of what is going on in a baseball game. If you score more hits, you record more runs. If you get more strikeouts, you allow fewer runs.

But statcast data is a relatively new development that goes a step further. The “statcast” era started in 2015. The point was to track how everything moves around the field. We track

• How fast the ball comes off the bat
  
• The spin rate of the pitcher’s throws
  
• The reaction time of an outfielder
  
• How fast a baserunner is

And many more things. Basically, statcast is tracking data. I’ve written at length before about how important tracking data is for the future of sports analytics. Any way to track players and equipment and how it moves gives us higher fidelity information to analyze the game with.

For the analysis contained here, we are going to look at some statcast data from the home run derbies over the better part of the last decade. If players get more tired, their home runs should travel further. If they get more tired, the ball might come off their bat slower. Tired players might not launch the ball at the optimal angle. The point is this: we can use these measurables to see how tired players get over the course of just a short home run derby round.

Home Run Derby Contestants Tiring in Each Statcast Measurable

First, let’s look at the most obvious stat: how far a home run travels as a measure of tiredness. We would expect that as players get more tired, their home runs would travel less distance. The plot below shows a scattering of home run distance as a function of how far into the round they are. The red line shows the regression.

While the relationship is not strong, what we do see is a slight downward trend throughout the round. The slope of the regression line is .08. This translates to an average home run at the end of the round being 8 feet shorter than an average home run at the beginning.

Lets look at exit velocity now.

Exit velocity throughout the round decays less quickly than home run distance. The exit velocity only drops by about 1 mph towards the end of the round. However, this represents a 1% drop. This is still pretty significant for just two minutes of exertion.

Finally, we look at launch angle as the home run derby round progresses.

Again, we see a very slight decrease as the round goes on, but a decrease nonetheless. Here, the slope is about -1 degree throughout the round. While this might “again” feel like a not-so-significant decrease, think again. One percent here happens to be about 3% of the total!

Putting it all together tells a coherent story. Exit velocity decreases by about 1% through a home run derby round. Launch angle suffers a bit more, going down by 3%. Finally, total distance – being a function of both launch angle and exit velocity – decreases by about 5%.

Conclusions

The data shows that players do noticeably get more tired as the home run derby goes on. The average home run distance drops by about 6% throughout the course of a home run derby round. Home run distance can be decomposed into launch angle and exit velocity. Both of these statcast stats decrease throughout a home run derby round. However, launch angle is responsible for more of the drop-off. Launch angle drops by 3% while exit velocity drops by only 1%.

To me, this means players are swinging just as hard through the round, but their accuracy drops as they get more tired.