January 10, 2014
The Trouble with Forecasting Chemistry
We’re 27 days away from the first pitchers and catchers reporting to spring training. We’re also 27 days away from the first reporters reporting,* which means we’re no more than 28 days away from reading quotes about team chemistry.
Chemistry is confusing, and not just for those of us who haven’t played professional sports. Even among players, opinions on its value fall along a wide spectrum from “essential” to “superfluous.” On one extreme, you have Eric Hinske, who believes that one can’t win without chemistry (and whose presence was, conveniently enough, perceived to promote it):
Then there’s the slightly more moderate stance of Hinske discipline Evan Longoria, who stops short of stating that one can’t win without good clubhouse chemistry, but still calls it “The most important thing.”
Finally, Jim Leyland, who when asked about the Blue Jays last spring exhaled and harrumphed:
It's possible, then, for a wide array of players who’ve put in plenty of clubhouse time to disagree about where chemistry comes from and how much it matters. But no one would dispute that, say, hitting more home runs, or allowing fewer, can make a team much better. As Leyland once put it, apparently: "I'll tell you what fucking chemistry is: Chemistry is a .300 batting average, 30 home runs, and 30 stolen bases."
So when Baseball Prospectus predicts team and player performance, as we do every spring, we tend to focus on homers, steals, and other things we can count. There’s a lot less ambiguity about those things. We know what they’re worth, and we can project with at least some degree of confidence how many each player (and by extension, each team) might produce. Our projections are far from perfect: we can’t predict luck, and we haven’t made much progress in predicting injuries. But as Bo Porter points out, there’s another nut we’re not cracking:
Well, shoot. We all want to improve our predictions, and if we could do that by accounting for insides, we would. Contrary to the stathead stereotype, not every numbers guy dismisses the potential impact of a team’s unquantifiable je ne sais quoi. We’ve written plenty about it at BP, and the third result you get when you Google “team chemistry” is an article by my podcast partner. We just don’t have a handle on how to tell which teams will have it when we’re putting predictions out, and GMs either can’t tell or won’t admit to being able to when they’re putting their teams together.
Ideally, we’d send a baseball version of the Trimates to live among the players and record their interactions. Because we can’t do that, we rely on less rigorous observation. According to the internet, the Chinese painter Gu Kaizhi once said, “What lies inside a man cannot be better revealed than through the eye.” (He also said, “What? I’ve never even heard of Cicero.) Kaizhi believed that the point of portrait painting was “to convey the spirit and give a vivid portrayal.” Many baseball writers subscribe to the same principles: that their job, when not needlessly tweeting play-by-play, is to look a team in its collective eye and tell us what lies within, using as many vivid one-sentence paragraphs as possible.
So come February, we’ll start seeing stories about which teams are hungry and full of competitive fire and on the same page and playing loose and pulling together. In theory, we should want to see those stories. No lesser authority on team chemistry than Torii Hunter believes that spring training is when a team’s identity forms. “That's when everybody comes together,” Hunter said last year. “When the season starts, you should already have your chemistry. Teams say that it's built in April and May. No it isn't. It's built in spring training.”
If spring is so crucial for chemistry, it’s the time when we should want someone around to chronicle the reactions (Gomes (s) + Inge (s) —> 24 wins). And who better to do that than the beat writer? We can look up and analyze stats and transactions from afar, but we have no hope of evaluating chemistry without the aid of someone on the scene. Beat writers spend more time in the clubhouse than anyone who isn’t paid by the team, so aside from the players, they’re our primary source.
What offends our sabermetric sensibilities is when chemistry is trotted out to explain success or failure after the fact. As Russell Carleton put it, “some of the common wisdom about chemistry is post hoc, ecological fallacy-driven narrative building that would fail an undergrad research methodology course. The Cardinals won and Molina is a good player and a hard worker. Therefore, Molina’s work ethic must have been the difference between the teams. Science just doesn’t work like that.”
Take, for example, an SI Orioles season preview from last March that began by mentioning the various bond-building clubhouse activities the O’s had engaged in the previous spring: table tennis, bumper pool, indoor golf. “That type of chemistry helps explain how the Orioles went 29-9 in one-run games and 16-2 in extra innings last season,” the next sentence said. It’s easy to draw that causal connection in retrospect—either for a surprise team like the Orioles or disappointing teams like the 2011 Red Sox and 2012 Marlins—but it’s far from convincing. Was Baltimore the only team playing table tennis? And where was the contention that the Orioles had a historically close clubhouse before they won those one-run games? To come closer to “sabermetrics for chemistry,” we’d have to have some data set down before Opening Day.
As I combed through columns from last spring in search of predictions for my article earlier this week, I came across a number of articles about chemistry. Eventually, I started to search for those specifically, and I came up with clubhouse quotes about almost every team dating back before the start of last season. See if you notice any similarities on the following team-by-team list:
In case you’re not counting, that’s 25 teams with great chemistry. And most of those quotes came from articles specifically devoted to assessing the state of a clubhouse: how the arrival of a new manager or new players had changed the mood; how returning players had grown together over time; how well team-building spring training exercises were working. I came up empty for only five teams: the Angels, the Brewers, the Mets, the Reds—who came close—and the Twins (and I might have missed something). Only the Mets’ clubhouse betrayed any publicized signs of unrest, thanks to incidents involving Jordany Valdespin and Zack Wheeler and Aderlin Rodriguez, as well as David Wright’s absence during the WBC.
So what should we make of the good feelings that seemed to sweep through every clubhouse last spring? One possible conclusion is that just about every team—with the occasional outlier on the low or high side—has pretty good chemistry, in which case clubhouse culture probably isn’t the separator it’s often made out to be. Another is that Hunter is wrong, and that spring training is just too soon to tell. Maybe every team starts out believing it has a great group of guys, but doesn’t discover its true nature until the going gets tough. But if spring is too soon, then winter is way premature, which means that teams would have no way to tell whether they’d assembled a good chemistry club until it’s too late to change course.
There’s another possibility: it could be that the players interviewed aren’t being honest—why air their dirty laundry and create further controversy?—or that no two or three of them can speak for a team. What strikes one player as the perfect clubhouse culture might rub another the wrong way, or affect his performance to a different degree. And while it seems like gauging group makeup would be one of the most obvious ways in which a reporter could add value, even writers who spend a lot of time around a team aren’t necessarily qualified to assess inter-player relationships when the clubhouse doors are closed. The chemistry column, after all, is an easy one for a writer who’s out of ideas to half-ass without substance on a slow day in spring training. Most players are happy to back up a narrative with boilerplate quotes if it gets the guy with the notepad to leave them alone.
Whatever the explanation, we’re not going to get any closer to the truth about clubhouse chemistry by bookmarking the dispatches filed from Florida and Arizona next month. It’s quite possible—even likely—that some teams have superior chemistry, and that that quality is worth some number of wins, just as it’s possible that some players who show up to spring training in the best shape of their lives will benefit from it. But when the available evidence suggests that almost every team has an enviable attitude, it doesn’t leave us with much useful information to work with. All it does is turn everyone into the team who cried chemistry, and ensure that when the real thing comes along, we can't adjust expectations accordingly.
*Except for Pirates beat writer Rob Biertempfel. He's already there.