Saturday, February 2, 2013

Super Bowl Sims and Championship Characteristics - 2013

Here is an excerpt from our analysis and numerical simulations of this year's Super Bowl, between San Francisco and Baltimore, which is published in the New York Times.  Our quant facts have been correct more than 60% of the time.



Consistency: Ball control remains one of the more important offensive indicators studied in both professional and college football. The team with a better running game, as measured by average yards per rush, has won 57.8 percent of the Super Bowls. The 49ers averaged 5.1 yards per rush this season, compared to 4.3 for the Ravens. Edge: San Francisco.

Monte Carlo Simulations: Probabilistic models like Monte Carlo techniques can be used to solve complicated problems. Similar to our analysis for the B.C.S. national title game between Alabama and Notre Dame, we used regular-season statistics in combination with our championship factors to simulate thousands of football games between San Francisco and Baltimore.

Read more here:
http://fifthdown.blogs.nytimes.com/2013/02/02/keeping-score-what-stats-and-simulations-say-about-the-super-bowl/ 

In addition, our popular analysis of Super Bowl Square Odds -- by Quarter -- is published here:
http://ireport.cnn.com/docs/DOC-917541