"Their analysis is based on some simplifying assumptions, but the conclusion that this will be a close game is supported by the data."
Needham, MA (PRWEB) January 28, 2015
Using a mathematical tool called Bayesian statistics and prior data from each team, a group of students at Olin College of Engineering has come up with the likely winner of the February 1st Superbowl. The stats account for field goals and touchdowns, but not safeties and two-point conversions or the possibility of overtime. Still, the students are confident they can predict the winner and the likely point spread.
Hint, it’s New England.
Now, you might think that because Olin College is in New England that might bias Alex Crease and Matt Wismer’s conclusions. But here is how they generated their results: “We considered scoring to be a single Poisson process, and have a separate distribution that represents our belief about the probability of a team scoring a touchdown or a field goal. We then calculated the probability of scoring a certain number of touchdowns and field goals based on the binomial distribution. The single Poisson process is related to the ability of the offense to get into field goal range, and the percentage of touchdowns is related to the team’s red zone efficiency.
“We also keep track of offense and defense separately. To calculate the distribution of points scored, we average the offense’s scoring rate with the rate at which the defense allows scoring, and average the offense’s touchdown percent with the percentage of touchdowns that the defense allows.
“The points distributions for each team were generated as described above, so we then calculated the probability that one would be greater than the other, or, in simpler terms, the probability that each team would win.”
The net result is that Crease and Wismer predict a 54 percent chance that the New England Patriots will win and just a 46 percent chance that the Seahawks will win.
The students worked with Prof Allen Downey, who added, “Their analysis is based on some simplifying assumptions, but the conclusion that this will be a close game is supported by the data. But please don’t blame us if you lose a bet!”
The rest of the data – including a predicted point spread – can be found here.