Baseball Therapy: Bumping the Grind

I know, you all just want to talk about the trade deadline. It’s fun because you get to see general managers pretend they aren’t sweating over the fact that they just mortgaged two really good prospects to pick up a guy who may or may not even get them into October this year. Of course to get to October, the GM—and all of the other 24 players who weren’t picked up in a trade at the end of July—have to get through August and September. The dreaded dog days of summer. By Friday, we’ll know where almost everyone is going to land for the stretch drive. After Friday, teams actually have to go play the games.

If you’re not a player who is changing addresses this week, the trade deadline might seem like just another day, other than getting an extra hug or two. It’s still hot. You still haven’t seen your wife all week because you’ve been on a road trip to the West Coast. Your iPad charger has been acting weird. Something in your suitcase smells, and you have no idea why that is. Your shoulder’s been aching a bit … not that you’ve told the trainer that. And really, do we need to play again tonight on a Tuesday in front of maybe 15,000 people? When’s our next day off? Okay, when’s our next day off where we aren’t required to take a cross-country flight? Three weeks? Three @#$%ing weeks?

Baseball is a game that on both the micro and macro levels takes patience and perseverance. Playing right field can be lonely and dull. There are long periods when there’s not a lot to do, and it’s not like you can just sleep through those times. Some time ago, I looked at the effects of The Grind on players and found that over time, they show small effects in their ability to make good decisions at the plate. Small in the sense that we might not notice them with the naked eye, but they turn out to be large is the sense that they add up pretty quickly into lost value for a team. I also found that there appear to be managers who have some sort of repeatable talent at keeping their players from falling away because of The Grind. In fact, I estimated that a manager who is particularly gifted in this skill might save his team 30 runs (!!!) over the course of the season compared to a merely average manager.

But let’s ask a different question. Are there certain players who are better at handling the stress of the long season? Their bubbly and effervescent personalities never let them get too far down over the course of a season? Or maybe their stoic and steady personalities never let them get too out of sorts? Something like that. Baseball players are human beings and human beings differ in their abilities to handle boredom. Surely every team has that one guy who’s always so @#$%ing positive all the time. He must be able to fight The Grind. Right?

Warning! Gory Mathematical Details Ahead!
Similar to my work on managers, I used a simple metric for the basis of my analyses: strike-avoidance. I coded each pitch for whether or not it resulted in a strike. Called balls were good. Balls in play were good. Two-strike foul balls were good. One-strike foul balls, not so much. I also calculated how many days it had been for the player since his team’s Opening Day. He might not have played in that opening game, and in some cases, might not have been on the roster, but it’s a decent proxy for how long he’s been playing baseball of some kind.

For each player, I constructed an individual regression. I even controlled for the quality of the pitcher he was facing in each plate appearance (some pitchers are better at inducing strikes than others). But the important variable was the number of days it’s been since Opening Day.

When I lumped every player and every pitch into one giant regression, the regression showed a slight, significant downward trend (more strikes over time, compared to expectations), which makes sense. But, of course, each individual had his own pattern. What I was interested to see was whether that pattern would be stable over time. I ran a regression for each player-year and looked at the slope of the line for each guy. I looked to see whether the patterns were similar over time. Is he the kind of guy who fades quickly over the course of a year? Does he stay pretty level? Does he actually improve? I wanted to see whether the regressions found the same “personality type” from year to year in the same person. To look at that, I used an AR(1) intra-class correlation (kind of like a year-to-year correlation, but with more data points) on all players who saw more than 2,000 pitches in a season. My data set covered 2010–14.

The answer: There was no consistency over the years. The intra-class correlation topped out at .07. (I tried it a couple of different ways statistically.) What that means is that if a player has a history of being the kind of guy who finishes strong or who has a problem with fading out at the end, that history has very little predictive power over the future.

He’s the Kind of Guy …
There are a lot of statements in baseball that start with “He’s the kind of guy who …” We are quick to ascribe performance to a personality trait rather than … a performance. Sabermetrics has debunked a few of them (clutch hitters being the most famous). I think baseball has a problem understanding that people and situations change much quicker than we realize. Maybe there is some set of circumstances that can lead to a player not drifting off as the season goes along. (My research suggests it’s the manager’s doing, or perhaps that’s just a proxy for the general atmosphere around a clubhouse. Hard to tell.) This is another one of those cases. There’s not really such a thing as a guy who is immune to The Grind. At least as measured by a nice straight line.

Maybe there’s some local variation that’s explainable (he was happy during August, but sad during September), but the point there is that to predict what will happen next, you need much more information than just his past performance. At that point, it’s more state than trait. It could also be that these findings are more of a sample-size problem. My work with the manager found a consistent effect, but he gets to “shepherd” 130 or so pitches per night, while a batter might see 15–20. Maybe there’s just too much noise to straighten the lines appropriately. Still, even regulars who play almost every day don’t show much consistency from year to year. One other possibility that I didn’t account for was the team’s record or playoff positioning status. As a team fades from contention, do players fade from being interested?

What it means is that your favorite team, whether it’s the proud new owner of a veteran starter or just middling along, is going to have to fight through The Grind. And while it’s tempting to say that your guys are the sorts who don’t mind the long hours and the hot weather in its fifth month, it appears that The Grind catches up to everyone sometimes.