Is there any statistical evidence for “momentum” in sports? All the announcers of the World Series seem to think it is of paramount importance. It seems to me that sports momentum has very little inertia, if it exists at all.
I personally don’t believe in it. The Yankees should have had momentum up the hoop after blowing out the Sox in game 3 of the ALCS, but it didn’t seem to do them much good.
When a team supposedly gets the momentum, they will continue to do well unless the other team gets the momentum. But, I ask you, if it’s possible for the other team to take the momentum away, then how could the first team have had it in the first place?
Momentum is just commentators way of explaining streaks that will just naturally happen anyway, without any mysterious force to cause them.
Years ago, I read an article that debunked the “hot hands” hypothesis in basketball. The article was based on the following famous paper:
Gilovich, T., Vallone, R., & Tversky, A. (1985). The hot hand in basketball: On the misperception of random sequences. Cognitive Psychology, 17, 295—314.
It’s surely overrated, and the cites against shooters with the “hot hand” (I’ve seen that before) are on the mark. Still, there’s something to be said for the psychology involved in sports. If one team is nervous or back on its heels, they’re more likely to make mistakes and more likely to get outside of their comfort zones, causing more problems.
I think momentum does exist in sports but it’s better described as confidence. If a team/individual has had a number of good results they are more confident and are likely to perform better. Having a very bad result can shatter that confidence and cause the team/individual to have a bad run of games.
Besides having the hot hand in basketball, momentum can also probably be aligned pretty closely with the concept of being a clutch hitter or the guy making a great play in the field leading off the next inning. In small sample sizes it may seem apparent or we may notice it because it happened, but it really doesn’t have much basis in fact in the long run.
But, really, in all these cases, we’re dealing with people, not numbers. And people are more than just probabilities.
For instance, the “hot hand” in basketball might be mathematical, but it also could be based on other factors. Say the player breaks his hand, and every shot causes excruciating pain, but, for some reason, he stays in the game. It might look like it was just mathematics, but if he misses eight shots in a row, is it because of the pain, or the pattern? Similarly, if a player thorougly outclasses his defender, he’s going to have a better chance of hitting eight in a row. If I were to try to guard an NBA player, I’ll bet he’d shoot better than average. If the defender as a broken foot and stays in the game, the offensive player is going to do really well – and it won’t be because of the math. What happens can be explained as just a regular mathematical pattern, but it also could be due to factors having nothing to do with math.
It’s the same with momentum. If you suffer a frustrating loss, it’s much harder to come back the next day. You start pressing – trying too hard – to compensate. Once the Red Sox won those two games, the Yankees started to get frustrated and started pressing.
Another indication that it’s not simply math is the fact that, in baseball, at least, if a team loses the first three games of a World Series, it is overwhemingly likely to lose the forth. In all World Series from 1903-2004, there were (quick count, so you can double check) 21 series where one team won the first three games, and only three where the opponent won the 4th game (I don’t have time to check the Championship series, but there’s a strong chance that the Red Sox were the first to win game four after losing 1,2, and 3). Considering that these involve the top teams in the two leagus, this .142 winning percentage indicates that the results are not what pure math would predict. The psychological effect of being three games down (and three games up) factor into the results.
Bill James did a few studies back in the 1980s that showed that momentum was bunk. A team’s likelihood of winning their next game is best predicted by the winning percentages of the two teams over the last 100, 150 games, not what streak they’re currently on.
I can believe that a team can go down in the dumps or ride a hot hand and gain momentum at LOWER levels of play. At the professional level, I don’t buy it; you’re dealing with athletes who by and large handle pressure as well as anyone alive.
RealityChuck has an interesting point about the amazing frequency with which teams that go down 0-3 lost Game 4 as well. Until this year’s ALCS it was something like 20 out of 25 teams (counting LCS play.) However, that’s certainly not impossible to have happened by sheer random chance, and interestingly enough most of the teams to recover and win at least one game have been recent ones, suggesting that maybe it was just a fluke.
Here’s a similar thread I started about the phenomenon of being “In the zone”. The summary was that there was no statistical evidence for it, but many people swore anecdotally that it was true.
The classic saying about momentum in baseball was said by Orioles manager Earl Weaver.
“Momentum is tomorrow’s starting pitcher.”
In other words, each game is a new event and has little bearing on how the next one comes out.
There could be exceptions, but that would really only happen if there significant changes in the personnel.
Say you are the San Francisco Giants and you win a game 7-2 and Barry Bonds hits 2 home runs and gets on base 5 times. But Bonds hurts himself and can’t play the next day.
The momentum may very well be thwarted, but that’s because the Giants personnel the next day will not be optimal.
Momentum might not be something that can be determined to exist statistically, but I think the psychology of it does exist. Momentum and morale are kind of the same thing in sports. When I played sports in high school there were games where as soon as I suited up I could tell whether or not the team had it that day. If there’s a good “feel” in the lockerroom, the warm-up goes well, etc. Though it’s entirely possible that I have a case of selective memory.
This also reminds me of a study I read about a couple of years ago: some economists determined that football teams would be better served by always going for it on fourth down. It had something to do with opportunity cost; I don’t remember specifics. Of course, no teams always go for it on fourth down. If that were the best idea, football coaches would have started using that strategy by now. Why haven’t they? They probably don’t want to have the momentum shift. 