Psychology matters in sport, at least in soccer. That's the unambiguous implication of a new paper by Jose Apesteguia and Ignacio Palacios-Huerta.
I watch more than my fair share of European soccer, and often hear the commentators speaking about a form of first-mover advantage. Scoring first, for example, provides a clear strategic advantage ( it is easier to defend than attack). It may or may not pile on psychological pressure, and disentangling psychological advantage from strategic advantage is no simple matter. A similar claim is made with regard to playing first in the round for teams in the title race: if you win and take the lead in the race, you supposedly put pressure on the teams that are chasing you. I find that a bit more dubious, but it could be true.
"Choking" is a form of failing under pressure, and this effect could be present in both individual and team performance. The flip side, "clutch" performance under pressure, has been studied extensively and is something of a holy grail among baseball stat geeks. Nobody's found robust evidence of clutch performance, including me, and I've spent months of my life sifting through data just in case everyone else has "missed it."
Apesteguia and Palacios-Huerta find a form of choking, in penalty kick shootouts. The data come from knockout competitions invovling a variety of club and international federations. The results are truly remarkable: the team that shoots first wins 60%, and the team that follows wins just 40% of the time. The coin flip is thus a major shift in the probability of winning (more so than the controversial coin flip in NFL overtime). Moreover, whether they know the numerical statistics or not, players and teams behave as if it matters, overwhelmingly choosing to shoot first when given the option. The inference is that the first team, by scoring, imposes a psycholgical burden on the second team. Here's the concluding paragraph, which sums things up nicely:
Lastly, from the perspective of the recent behavioral economics literature, we find a signifcant and quantitatively important type of psychological effect not previously documented. From the perspective of rational choice theory, we find that individuals are aware of this effect and they rationally respond to it.
The paper is forthcoming in the American Economic Review, and is full of all the technical bells and whistles that it is fashionable to criticize these days. Reporter Julian Guyer has a nice summary (hat tip to Julian) and discusses a simple reform proposal with Ignacios (hint: copy tennis), that makes a ton of sense to me.