Watch Live

Tweet TWEET

Super Bowl Quant Firm Won’t Quit Day Job as Giants Win Puts End to Streak

The New York Giants’ Super Bowl victory ended an eight-year winning streak by the quantitative money management firm whose model for picking the team to bet on went with the favored New England Patriots.

Analytic Investors LLC in Los Angeles predicted a Patriots victory by more than the three-point spread set by oddsmakers. It based the prediction on what it said was a historical tendency by Super Bowl bettors to overvalue the team that was more profitable to bet on during the regular season.

The Giants won 21-17 at Lucas Oil Stadium in Indianapolis Feb. 5 for the National Football League championship.

Nothing’s wrong with the model, whose success rate over the past eight years exceeds its long-run success rate, said Matt Robinson, a portfolio analyst for global and Japanese equities at Analytic.

“If we were getting it 100 percent right we would probably all quit our day jobs,” Robinson said in a telephone interview yesterday. “The model will not in the long run identify the team that’s going to win against the spread 90 percent of the time.”

He said that the model instead identifies teams that have a relatively higher probability of success. Betting against it would be like betting tails on a coin that’s weighted to come up heads three out of four times, Robinson said.

While disappointed in the outcome, Robinson said he was consoled by the fact that the game was close in the end, with the Giants’ victory at no point looking assured. A last-gasp pass by the Patriots’ Tom Brady fell in the end zone.

“We had a chance to tie the spread and push on the day if Brady had completed that Hail Mary,” he said. “There was a chance to pull it out.”

For Related News and Information:

To contact the reporter on this story: Elizabeth Stanton in New York at estanton@bloomberg.net

To contact the editor responsible for this story: Michael Sillup at msillup@bloomberg.net

Bloomberg reserves the right to remove comments but is under no obligation to do so, or to explain individual moderation decisions.

Please enable JavaScript to view the comments powered by Disqus.