Dear Hitters: It’s Okay To Take a Break

The other week, I was talking to a friend as she fretted over her lack of productivity. Her struggle wasn’t with laziness or a lack of motivation, but rather a severe case of burnout, following her around the way No-Face follows Chihiro in Spirited Away. In the moment, I could see her desperately reaching for more energy to get through the day, like when the chip bag is nearly empty, so you bring it to your mouth and tilt your head back to suck down whatever salty goodness remains.

But instead of encouraging her to power through, I went a different way.

“Y’know, it’s okay to take a break.”

She laughed and said, “That’s what my therapist keeps telling me.”

If you’ve ever described yourself as a perfectionist or a people pleaser, or tied your self-worth to your measurable output, you know taking breaks can be hard.

Athletes learn from the beginning to idolize hard work. Hard work is the salve for every ailment. Wanna get stronger? Work harder. Wanna play better? Work harder. Wanna go pro? Work harder than everyone else. The hardest workers earn their own dedicated titles and recognition, separate from their actual production. Cal Ripken Jr.’s consecutive games played streak earned him the Ironman title. MLB gives out a Heart and Hustle award to whichever player’s heart tells him to hustle the hardest. And every year, we read stories about players and coaches who are the first to arrive to work each morning and the last to leave at night.

But the universe demands balance, and hard work is no exception. Hard work only provides benefits when paired with recovery. Without recovery, we slowly put ourselves in a hole. While in the hole, our performance degrades. As we keep digging, the hole gets more and more difficult to climb out of, until eventually it becomes our grave.

Yikes, that got dark.

All of that is to say, it’s important to balance out work with rest, but in the context of a team sport with a 162-game season, scheduling rest while rolling out competitive lineups every day is tricky. There are a lot of ways to go about measuring fatigue and its relationship to performance. Public-side methods will never compare to what teams can do with input from medical professionals, strength coaches, and access to the players themselves. But the more we can track the way players move their bodies and how that movement fluctuates throughout the season, the more likely we are to find a measurement that works as a reasonable proxy for when those bodies are doing too much.

For some time now, we’ve been able to keep tabs on pitchers’ release points and how hard they’re throwing as possible indicators of fatigue. Until recently, hitters’ movements could only be analyzed via the eye test, but with Statcast’s new bat tracking metrics, we can more objectively and efficiently analyze what’s happening with a hitter’s swing. Bat speed and swing length both have the potential to tell us something about how a hitter is feeling, but since it’s relatively easy to identify good bat speed compared to bad bat speed, and the same cannot be said of swing length, we’ll stick to bat speed for now. First, we’ll need to establish whether or not bat speed actually works as a proxy for fatigue, then we can explore what it tells us about the frequency and effectiveness of resting hitters.

The theory behind bat speed as a measurement of fatigue rests on the assumption that, for the most part, hitters are swinging as hard as they can without sacrificing the barrel control needed to make contact. They might ease up in certain situations, but if their overall average starts to dip, it’s probably not a result of active decision-making. With that in mind, if a player’s average bat speed starts to trend downward during a lengthy span of games played on consecutive days, then that number should rebound following a day off. If we don’t see a player’s bat speed tick up, then we might conclude that rest does not impact bat speed, or that the player didn’t really rest, instead using the time off to do more rigorous drills or lifting not appropriate for a game day, or perhaps that the player got a day off due to some nagging injury that’s interacting with all of this in some unknown way.

The analysis begins by rounding up players who played at least seven days in a row prior to a rest day. I grabbed each player’s swings from the five days before their day off and the five days after their day off, and calculated their average bat speed for both samples (minimum 30 swings required). I went with five days because my prior reliability analysis determined that approximately 30 swings are needed for a representative sample. Most starters will hit that number in four or five games, but measuring days made more sense in the context of an analysis dealing with fatigue over a span of consecutive days played. Admittedly, a five-day sample for players who only played seven consecutive days is still going to include some swings where players are probably still feeling pretty fresh, but thus are the compromises we make when working with real world data.

For this first comparison point, I did not require the player’s bat speed to have slid below their season average prior to the off day. Rather, I wanted to determine if a day off is beneficial, regardless of how things are going for the hitter beforehand. In 507 instances meeting the criteria described (across 252 distinct players), hitters saw their average bat speed improve by 0.03 mph following a day off. A positive return, but given the variance in the sample, not enough to make it statistically significant, meaning the difference in average bat speed is more likely due to randomness than any actual effects of the day off.

Perhaps the number of days with consecutive games played makes a difference. Intuitively, the longer one goes without a break, the more fatigue sets in, and the more beneficial the rest. The chart below shows the average change in bat speed, bucketed by consecutive days played. The improvement in bat speed is more pronounced after a longer streak of days in the lineup, but the difference still isn’t large enough to attribute to the rest day:

The change is minimal (and even slightly negative due to the muddiness of the sample) in the seven to nine day bucket. The largest improvement happens in the 10 to 12 day bucket. Though we still see positive gains in the 13 days or more bucket, the impact is much smaller than at 10 to 12 days. I’m comfortable attributing this to survivorship bias. Given that everyone reading this is probably familiar with the phrase “load management,” we find ourselves in an era where players are rarely asked to play that many days in a row (whether by their manager or the schedule), and the ones who are probably feel pretty good, or they wouldn’t do it.

Nothing conclusive came from comparing before and after bat speeds from the full population of players, so next we move on to players who likely are experiencing some level of fatigue. Does an off day provide a clearer return on investment? For this comparison, I filtered the sample to players whose average bat speed in the five days prior to a day off had dropped below their season average even a tiny bit. Among the 276 instances meeting this narrower criteria (across 194 distinct hitters), the average improvement in bat speed following an off day was 1.04 mph, and as shown below, the improvement is fairly consistent, no matter how many consecutive days a player posted up. This suggests players fatigue at different rates, so rather than assuming fatigue after a certain number of games, it makes more sense to keep an eye on the player’s movement and strength indicators, such as bat speed:

In this case, the measured improvement in bat speed is statistically significant, meaning we aren’t getting bamboozled by randomness. That said, an average is still a summary of a whole bunch of players, so any individual hitter’s mileage may vary, and as shown, the magnitude of the return isn’t necessarily tied to the number of consecutive days posting up. Perhaps the increase in bat speed following a day off is more closely tied to how fatigued the player is prior to the rest. The chart below shows the negative gap between each player’s full-season bat speed average and their five-day average bat speed just before getting a day off, and plots that against the change in bat speed observed in the five-day average following the off day:

The graph shows that regardless of how far in the hole a player is relative to their season average, that number will typically bounce back to the full-season number following a day out of the lineup. The more fatigued a player is, the larger the return on investing in that player’s rest and recovery.

The players situated farthest from the trendline don’t offer an immediately obvious pattern for their non-conformity. It makes sense that catchers and DHs would have a less straightforward relationship with rest. Beyond that, the non-conformers are likely impacted by factors trickier to observe — things like longer-term nagging injuries, purposeful swing adjustments or drill work, extra medical treatments, etc.

But what is the actual performance return on re-gaining a bit of bat speed? xWOBA models tell us that hitting the ball harder generally yields better results and faster bat speeds lead to harder-hit balls, but how impactful is one or two ticks of bat speed when it comes to impacting the baseball? To get a feel for the relationship, I plotted qualified hitters’ wOBA on the season against their full-season average bat speed. The middle 50% of players form a largely linear relationship between the two metrics, but the more extreme hitting profiles create some weirdness on the fringes. Some hitters find outlier success as slow, purposeful hitters (hello, Steven Kwan and Luis Arraez), while diminishing returns on swinging harder start to kick in around 76 mph bat speeds if not supplemented by other skills (see: Giancarlo Stanton, Oneil Cruz, and Jo Adell):

Within the linear chunk, adding 1 mph of bat speed is good for a five point improvement in wOBA. Five points probably isn’t a super noticeable difference over the span of a few days of the regular season, but for players with bat speeds lagging their season average by 2 to 4 mph, the 10 to 20 points of wOBA left on the table are worth recouping. Obviously, context matters. Roster construction may limit the options for getting certain players rest. A series against an opponent a game or two up in the standings may mean that playing a diminished version of your best hitter is better than the alternatives. But as circumstances allow, resting a hitter who seems to be swinging a bat with the donut still on should provide a nice jolt of energy upon re-entering the lineup.

It’s difficult to re-route the hardwiring that tells us the only way to get better and make a positive contribution is by doing, and that doing requires action with tangible results. Instead, maybe we should look for tangible results in places we didn’t have access to before (like by using some shiny, new metrics). Re-gaining two ticks of bat speed and 10 points of wOBA is more tangible than vaguely feeling better after some time off, especially if you don’t actually feel that much better because you spent your whole day off being anxious about not getting anything done. With measurable gains in hand, spending an afternoon parked on the couch with a smoothie and then cheering on your pals from the dugout might feel less like being put in time out and more like productive action.

And so, for the benefit of hitters and anyone else feeling the weight of weariness, I will reiterate what I told my friend: It’s okay to take breaks.