Professor Unveils Soccer's Hidden Stats
GUY RAZ, host:
Now, is it really possible to figure out the best soccer player in the world? It's a little easier in baseball where you've got loads of stats to make the case for a Babe Ruth or a Ty Cobb or a Lou Gehrig. But up until recently, regular stats were not kept for soccer. So players have pretty much been judged by...
Professor LUIS AMARAL (Chemical and Biological Engineering, Northwestern University): How much they make people go wow.
RAZ: That's Luis Amaral. He's an engineering professor at Northwestern University and a devoted fan of Portugal's national soccer team. And he's created an algorithm that can take the official stats from this year's World Cup and give us a list of the best players in the world.
Prof. AMARAL: Nowadays, they keep track of a lot of things, including how much a player runs during the match, how many shots he makes, how many shots on goal, how many passes he receives, from whom, how many passes he makes, how many of those passes are successful, how many tackles. Lots of things of this sort.
RAZ: Now, you have, obviously - you and I and everybody listening to us now has access to all of those statistics. You have done something different. You have taken them and you plugged them into a computer program that you created to kind of determine who the best players in the world are. First of all, how does the program work?
Prof. AMARAL: The work is based on some of my research which falls under discipline called social network analysis. Essentially you can see a football team as a network of individuals that are collaborating towards a common goal.
And in soccer, what you want to do is you want to gain possession of the ball. You want to keep possession so as they create an opportunity for shooting, hopefully on goal and hopefully to score a goal. So what we try to do is, based on the data that is being collected, build a social network of the players and try to see how likely that social network is to enable a particular team to shoot on goal.
RAZ: So, you've been obviously been looking at the first round of the World Cup. Who've been the best players so far, according to your calculations?
Prof. AMARAL: Argentina is playing at the very high level and Lionel Messi has been the player that, according to our ranks, has played the best. Maicon, which is the Brazilian defender, Gilberto Silva, which is a Brazilian defensive midfielder and Juan Veron, who is an Argentinean midfielder have been the top...
RAZ: I'm noticing a pattern here - Argentina, Brazil, Argentina, Brazil, Argentina, Argentina, Brazil, Brazil, Brazil, Argentina. What about Americans? I don't see any Americans in your top 20.
Prof. AMARAL: Yes. But you also have to realize that there are many, many, many, many players in this World Cup. So if we continue...
RAZ: I think there's a flaw in your calculations there, Luis.
Prof. AMARAL: Probably.
Psrof. AMARAL: But - yeah. And I'm open to suggestions on how to improve my algorithm.
RAZ: So knowing what you know, right, about which teams have the best players now, what do the numbers tell you about who is going to win the World Cup?
Prof. AMARAL: Our numbers are not a prediction. After the fact, you go and measure how people perform. And you can never be sure that people will keep performing at a high level or that someone will not get injured. But if we would extend this level of performance, we would expect, I would say, that probably the most likely outcome is that Argentina is going to meet Brazil in the final.
RAZ: God help us.
Prof. AMARAL: It would be a wonderful game, even though Portugal is not there.
RAZ: My sympathies in advance to you.
Prof. AMARAL: And to you too.
RAZ: That's Luis Amaral. He's an associate professor of chemical and biological engineering at Northwestern University. He's developed a mathematical formula for determining how good a soccer player is.
Professor Amaral, thanks for joining us and enjoy the rest of the Cup.
Prof. AMARAL: Thank you. Thank you for having me.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.