A few weeks ago, Boston pitcher Jake Peavy sat down to discuss his (then) most recent start against the Atlanta Braves with the Providence Journal’s Tim Britton. Peavy went inning-by-inning, and when looking back over the sixth frame of that game noted that his Red Sox had taken the lead immediately prior to him pitching, so that he had the chance to throw a “shutdown inning” where he could keep the hard-win lead intact.
Peavy said: “I always laugh at the term shutdown inning, but it is. It is because of the momentum of the game. People are crazy if they don’t believe in momentum because there is such a thing. The team that has momentum, it seems like the good things go their way. Look at how we lost 10 straight. We could not catch a break. Now it seems the opposite. It’s important when you do score a run and have the momentum on your side to not give that up.”
Momentum is one of those words that gets thrown around a lot in baseball (and other sports). It’s that extra boost that a team gets from success building upon itself. One of the easiest clichés to reach for in the announcers handbook is that whatever event just happened, it conferred this special force upon one of the teams. They’ve got momentum on their side! I’m trained as a clinical psychologist, so it’s easy for me to see the argument for why momentum might actually work. Happy events lead to hope, and if you have something to hope for, you try a little harder. If you’re on the wrong side of the momentous event though, sad events lead to sad feelings and that can lead to hopelessness and despair. That has to translate a little bit on the field, right?
Then again, it was the great Earl Weaver who said “Momentum is tomorrow’s starting pitcher.” I suppose, according to Jake Peavy’s definition, that makes Earl Weaver crazy (Peavy would not be the first to make the accusation.) But as someone whose job used to be to decide who was crazy and who wasn’t, let me suggest that perhaps Weaver might be completely sane in his skepticism. Before we accept any claim, we should ask for proof. If momentum exists in baseball, we ought to see evidence of it in the game. And for that … well …
Warning! Gory Mathematical Details Ahead!
Let’s look for cases where we might find momentum. If momentum is more than just a description that something good has happened, but is instead something that influences what is about to happen next, we ought to be able to see its effects.
But let’s be careful about what evidence we admit in court. For example, suppose that a team of MLB players was playing at full strength against a team of amateur rec league guys. We’d expect that the MLB team would probably get a bunch of hits in a row, and that streak would continue. Is that because the MLB players are building momentum? No, it’s because the MLB guys are just better and we expect more out of them. Runs of good fortune are more likely if a team is good to begin with and the good results that they produce afterward might be because they are good rather than in the throes of Big Mo. If momentum is a real thing, we would expect to find situations in which baseball teams and players would not only perform well, but better than expectations.
To take one example, let’s look, as Ben Blatt of the Harvard Sports Analytics Collective did, at whether a team that comes back to tie a game in the ninth inning is more likely to win a game that goes into extra innings, more so than if they had simply been tied since the sixth. Ben found that when the home team tied the game in the bottom of the ninth, they won the game 53.8 percent of the time. When they had blown a lead in the top of the ninth (and not scored in the bottom of the ninth), they still won 53.1 percent of the time in extra innings. (Re-running his numbers with more recent data, I found something similar.) Surely, the emotion of a big comeback should grant the comeback team greater momentum, but it seems that the difference between being the team that makes the comeback and the team that gives up the lead is all of seven-tenths of a percentage point of winning. That’s nearly nothing. To give some context, if the pitcher walks the first batter in the top of the 10th, we can estimate that his team’s chances of winning drop by something on the order of nine full percentage points.
Let’s take a different view though. Mr. Peavy suggests that the then-recent cold streak that the Red Sox had been on had an effect on his teammates. I found all cases in which a team had won its last three games in a row and looked to see whether the next day, hitters in general did better than we might expect by getting on base more. To figure out what to expect, I used a method known as the log-odds ratio method. If you want to do a little more reading on the subject, it is described fully here. It turns out that the answer is yes, hitters do perform better the day after a three-game win streak. Barely. The effect size is a couple of points of on-base percentage. When I looked at a slightly longer time frame (teams that had won 4 out of 5 games, or 7 out of 10), the effect disappeared.
In a previous article, I ran the same basic research method looking at how hitters do in a playoff game after their team scores a come-from-behind win in the game before. There’s nothing better for building momentum than a come-from-behind playoff win, right? Except that the next day, hitters were no better (or worse) than we would have otherwise expected.
Well, maybe we can find momentum in other places. Peavy talks about the shutdown inning, where the offense scores a couple of runs to take a lead and then the pitcher puts a goose-egg up on the board to preserve that lead. Peavy is correct that throwing a scoreless inning is always a good thing, but does it give an extra special boost to a team? Again, this is a topic I’ve researched before and found that yes, there is an effect and that teams win slightly more often than we would expect after a shutdown inning. Note the emphasis on slightly.
Well, then what about the momentum of a pennant race? Surely a team that plays well down the stretch is going to carry that momentum into the playoffs. Right? It turns out that the answer is “not really.” Jay Jaffe (then of Baseball Prospectus, now with Sports Illustrated) found that there was little connection between a team’s record down the stretch and their performance in the playoffs. There is some evidence that playing in a lot of meaningful games (win or lose) does help during the postseason, but again, the effects are rather small.
In fact, when I’ve read well-done studies on momentum in baseball, the conclusion that they all seem to come to is that if there is momentum (and results are mixed as to whether or not it exists), the effects are actually rather small, even for the big-ticket emotional items that should produce a lot of momentum, like a ninth-inning comeback or a big winning streak. Small enough that frankly, we shouldn’t even bother with it. If it were as real as sportscasters appear to believe that it is, we should see plenty of evidence that it makes a difference, evidence which just isn’t there.
Am I Crazy?
Right now, there’s a reader out there who is saying, “Yeah, but I’ve played baseball (or some other game) a bit and I know I’ve felt it. I felt the momentum turning and sure enough, we won!” Maybe it’s you. As a trained psychologist, I actually buy part of the momentum argument. I do believe that emotions can affect a player’s abilities or at least his level of effort. In fact, if teams and players did let the emotion of the moment wash over them, I think that we’d see a lot more momentum effects in baseball.
The reason that I don’t think we see very many momentum effects, and the ones that we do see are so small, is that we don’t give the players enough credit for not letting their emotions get the better of them. Consider a team that has just suffered a heart-breaking defeat and has to play the same team the next day. Sounds like the recipe for one team having major amounts of momentum and the other being ready to roll over. After that heart-breaker, the players on the losing team will probably be sad, but will probably find ways to deal with that sadness. A lot of teams have rituals for getting over losses the same way that all humans have ways of getting over losses. If a team does not have the ability to do that, then that’s something that needs to be fixed right away. If a pitcher has made it to the majors, he at some point probably learned how to shake off having a bad couple of hitters. If we assume that a simple turn of events will destroy a Major Leaguer, we’re certainly not giving him any credit for being a rational human being who can say “Yeah, that’s awful, but I can’t do anything about that now … better control what I can now.”
So do players feel good after a turn of events in their favor? Certainly! Do they try harder or step livelier as a result? Probably. Is that going to turn the entire complexion of the game? The evidence says not really. And if that makes me crazy, then so be it.