As of today, there are 10 teams left in the World Cup, and Brazil is favourite. Never mind that they have won just two of four games so far (in regulation time, not including the match they won on penalties), and conceded goals in three matches (including an embarassing own goal.)
In fact, Brazil’s two wins, one draw, and one win on penalties is equivalent to Costa Rica’s performance. But nobody thinks the Ticos will win.
Here are the odds as presented by Sportsbet.
Germany, who have won all four of their matches, surely look to have better form than Brazil. France too.
Colombia, who have won four matches, scored 11 goals, conceding only two, would appear to be in even better form. The team ranked eight in global soccer at the start of the tournament doesn’t seem to be high on anyone’s list.
So why is everyone backing Brazil?
I propose that a cognitive bias is affecting prediction markets: The availability heuristic. The heuristic is that if people can imagine something clearly they believe it is more likely. The most famous proof for the availability heuristic goes like this:
“In one experiment that occurred before the 1976 U.S. Presidential election, some participants were asked to imagine Gerald Ford winning, while others did the same for a Jimmy Carter victory. Each group subsequently viewed their allocated candidate as significantly more likely to win.”
What’s available to our minds when we think about Brazilian football is scenes like this, from 2002:
And what comes to mind when we think of Brazil hosting an event is scenes like this:
Rio is synonymous with celebration. Can you really call to mind a picture of Rio de Janeiro full of glum Brazilians moping? Neither can I.
The availability heuristic probably means teams like Brazil are over-rated (the same to a lesser extent is likely true of recent World Cup finalists France, Germany and the Netherlands.) So should we let that affect our betting? Probably not.
But there is one exception.
We must acknowledge that referees play an important role in deciding the outcome of Soccer matches. Dodgy penalties come at crucial moments, like the one awarded to the Netherlands to effect their defeat of Mexico.
And referees are just as subject to biases as the rest of us. Research shows they are susceptible to crowd noise, home ground advantages and racial biases.
So if, with scores tied, a Brazilian player sprawls on the turf inside the penalty area in the 90th minute of the World Cup final, the incredible surge of noise inside the Maracana could well blend with the availability bias inside the referees head, and encourage him to point to the penalty spot.
If that happens, the team in green and gold will get to hold the trophy aloft, and all this lofty pontification about controlling predictions for cognitive biases will have been a waste.
Thomas the Think Engine 