Forget Moneyball; Haifa U study shows how emotions can beat baseball odds

In the 2011 movie Moneyball, baseball nerd Peter Brand (played by Jonah Hill) almost wins the 2002 American League playoffs for the Oakland A’s by playing the percentages – looking at the records of seemingly has-been players and putting them in the right positions, getting them to produce far more than they ever thought they could. Now, researchers at Haifa University may have found another system to beat the odds for baseball players – and maybe everyone else.

Called it “Emoti-ball” – a system that identifies the emotions of baseball players, especially pitchers and hitters, based on their body language and facial expressions. The findings are part of a study that was conducted by researchers from Haifa University and the University of Amsterdam in the Netherlands, that was published in the latest edition of the journal Frontiers in Psychology. The aim of the study was to determine how the expression of emotions provides clues about what is likely to happen in social situations in general, and it used the medium of baseball to demonstrate the process.

That emotions affect social situations is clear; you don’t need a study to realize that an extremely angry or combative person can successfully intimidate people in a meeting or at a party.

But this study aimed to dig a bit deeper – to understand how more subtle cues set off responses, perhaps subconscious ones, and lead to consequences that can perhaps be quantified and even predicted.

“Other peoples’ emotions provide us with information,” said Dr. Arik Cheshin of the University of Haifa, who led the study. “Expression of emotions can mark for us what the person is thinking and what they are about to do. If we read those emotions well, we will be better able to anticipate others’ behavior and adapt our behavior as needed.”

Understanding the faceoff

Baseball might seem to be an odd choice for the demonstration of these principles, but it actually makes sense. Players are forced into a social situation where they may be staring at each other for long moments, trying to “read” what the other party is doing; for example, a pitcher needs to be on alert when an opposing player on first base takes too big of a lead, which means that he may be trying to steal second. Thus, the pitcher has to get the ball to second base before the runner can get there.

In such a game of cat and mouse, both the pitcher and the runner try to play it cool, not giving away too much about their intentions.

It’s the same in a pitcher/batter scenario, said Cheshin. The study assumes that the emotions displayed – including very subtle ones – in the moments before an action are the true indications of what is going to happen. “The players stand opposite each other in a traditional duel stance, scrutinizing each other for a clue as to what comes next. One makes a move, and the other responds to it. We wanted to see whether the expression of emotion offers a clue about this move in advance – and we found that it does,” said Cheshin.

The joint study used as subjects Dutch students from the University of Amsterdam – a conscious choice, since they, as most Europeans, don’t know much about baseball, thus freeing them up to concentrate on the emotional aspect of the interaction between pitcher and hitter (needless to say, none of them knew the identities of the players, or even the teams involved).

The subjects were shown 92 instances of interchange from three games that determined the World Series champion from two different seasons, selected because the camera had been able to get a close-up of the pitcher’s face before the throw. The video clips – which lasted for just two seconds – were edited so that it was only possible to see the pitcher’s preparations before the pitch, and the clip was frozen once the ball left the pitcher’s hand. The clip did not show the outcome of the game nor provide any additional information about it, nor did participants see how the hitter reacted – at least not right away.

Then, 213 of the study’s 250 participants were asked to evaluate the pitcher’s emotions – whether he was angry, happy, or worried. Thirty clips with the highest level of agreement were then chosen, and another group of 34 respondents were then asked to predict the outcome of the pitch from the batter’s side – how fast it would be, how accurate within the strike zone, and whether the player would try to hit the ball.

The results of the study showed that, as theorized, the subtle cues on the faces and in the actions of pitchers triggered specific responses in the batters.

“The participants predicted various properties of the pitches according to the pitcher’s emotion,” said Cheshin. “When the pitcher showed anger, this led to the prediction of faster and more difficult pitches. The expression of happiness by the pitcher led to predictions of more precise pitches and a higher probability that the batter would attempt to hit the ball. The expression of worry led to predictions of imprecise pitches and fewer attempts to hit the ball.”

While the observers were able to accurately predict the outcome of specific pitches, they were less accurate about the overall outcome of games. “In baseball, the distance between the pitcher and the batter is 90 feet, and we are dealing with an emotion that is expressed for just a few seconds, during a period of movement, with a cap covering part of the face and a large glove on one hand. So the conditions for identifying emotions are far from ideal.” While more work is needed to develop the idea as a system to predict baseball game results, it shows a lot of promise as a predictor of social encounters. “It’s reasonable to assume that the expression of emotions when people are closer together, without accessories or items concealing the face, will yield stronger findings,” said Cheshin.

The happy tip off

There was, however, one emotion that turned out to be a reliable predictor of a game’s outcome; when a pitcher was happy, it meant that his team won more often than not (over half of the time, according to the study, as opposed to more random results for other emotions). When a pitcher showed happiness, the batter was more likely to try and hit the ball – and that meant more opportunities to strike out or hit foul balls.

The researchers suggested that this was an important finding, since the expectation in baseball is that if the pitcher is happy just before and during pitching, he is liable to be trying to execute some kind of scheme or trick. “It is possible that the batter’s reaction is not conscious but evolutionary. There is a lot of pressure and tumult around the batter, and accordingly the batter sees the pitcher’s expression of happiness as a positive sign that encourages him to try to hit the ball,” said Cheshin. “Whether this is an authentic emotion or a strategy, the expression of emotions has a social impact in sports as in other areas. Controlling the expression of emotions and the ability to read emotions in order to predict behavior can make the difference between a strike and a home run.”