There’s no sugarcoating it: this has been a hellish week for the Yankees and their supporters. The bats were nowhere to be found in the recent losses to the Orioles. Those games in particular highlighted a strategy that has frustrated Yankees fans during Aaron Boone’s tenure as manager. By that, I mean the inflexible adherence to rest schedules.
For the record, I’m not placing the blame for this methodology on Boone. In fact I believe he has very little control over it at all. The increasing reliance on analytics to dictate team management has taken a lot of the decision making power out of Boone’s hands. More likely than not, a carefully crafted algorithm has been derived for each player and shows what schedule of rest helps that player achieve peak effectiveness.
I’m also not knocking the transition to analytics-driven baseball. On the contrary, I love it. I’m all for leveraging the most cutting-edge resources to maximize a team’s success in the present and the future. As a fan of the game, I have never had more access to information at as granular a level as is made available today. I would argue, however, that analytics do not tell the whole story.
A problem with applying these analytical models is that they exist in a semi-vacuum, built upon seasons worth of aggregated data that is run through simulations thousands if not millions of times. They operate at the macro level, ignoring the context of the specific micro-level situations to which they are applied.
Back to the matter at hand, I think prioritizing player rest this season is dumb. Here’s the issue: the framework of regularly-scheduled rest days was constructed in and for a 162-game schedule. And for good reason — giving guys days off over the slog of a normal season allows players to preserve a level of freshness come the postseason. However, attempting to apply said framework to a 60-game sprint is like trying to fit a square peg in a round hole.
Any accumulated amount of energy hypothetically saved by getting days off will, in my opinion, never outweigh the detriment of removing that player from the lineup. Guys like DJ LeMahieu, Luke Voit, and Gleyber Torres will always contribute more value to the Yankees by playing than whatever fractional value they might gain for the postseason by getting rested.
Let’s look at LeMahieu specifically. Is it really worth resting him — ostensibly to maximize future wins — at the expense of possibly missing the playoffs this year? Need I remind them that his contract expires after this season and there is no guarantee he is a Yankee next season? Normally you try to extract every drop of production out of guys in contract years (see: CC Sabathia), and not worry about preserving them for the future. Is it possible that by resting DJ, Boone has given fans one less game in which to see DJ in pinstripes?
This brings me to my biggest issue with scheduled rest. DJ seemed miffed that he was rested. He seemed like a guy who had been told to sit out, rather than one who was tired and needed a day. He was feeling good and wanted to play.
Whenever a player talks about why they’re doing well, you rarely, if ever, hear them say, “well I optimized my launch angle,” or “I’ve been working on my exit velocity.” No, instead you hear the same references to the immeasurable aspects of the game. “I’m really feeling in a groove; I’m seeing the ball well; The game is slowing down for me; I’ve really hit my stride; I’ve gotten into a good rhythm; I’m carrying good momentum,” etc., etc.
The one common thread of all these observations is they are borne out of repetition and routine. And the best way to interrupt that routine is to have a guy sit out in the midst of a productive stretch. The last thing a guy who is seeing the ball well wants is to have an off-day. A batter on a hot streak wants to keep the good times rolling, not ride the pine. At the extreme end of the spectrum, giving a guy a day off could actually hinder their performance in the next game(s).
And that’s the biggest problem with the application of analytics in this regard. Analytics will never account for the human aspect of the game. They can never put a value on confidence, muscle memory, belief, or the hot hand. They ignore the very real effects that cognition and emotion have on behavior and performance.