I hope you enjoyed the game last night. How was OR used for the Superbowl? I enjoyed reading about the logistics of the 2005 Superbowl in Jacksonville, Florida–including using cruise ships to provide enough hotel rooms–but I didn’t see anything similar for yesterday’s game. Please share anything you’ve found.
Are the commercials worth the price tag now that viewers watch their favorites online? What was your favorite commercial?
Punk Rock Operations Research 
February 8th, 2010 at 12:40 pm
I came across an interesting blog on NFL stats that has some good info on the decision making during the game http://www.advancednflstats.com/2010/02/who-dat-gonna-kick-onside-to-start.html .
There were several critical plays during the game, the 4th and 1, the onside kick etc. that weren’t obvious on what was the best decision to make.
I know coaches use some probability to make certain decisions easier, i.e. when to go for two points, but I wonder on certain plays if they go more with their gut, or how they think their team will do vs. a straight probabilistic decision.
I think there was a study that basically said you should always go for it on 4th down no matter where on the field you were because the upside outweighed the downside. I think though that while in theory that may be right, there’s probably factors like the element of surprise that should get included. I don’t know the exact percentages but say there’s a 20% historical chance of people succeeding on 4th down, if a team went for it every fourth down their actual success rate would probably not mimic the historical rate because other teams would learn to expect it. These types of studies often make people think that conventional wisdom isn’t “smart” because they’re not using statistics to make rational decisions but there’s probably more that needs to be looked at such as game situation and the element of surprise.
It was also interesting during the game because if you looked at the one 4th and 1 play in isolation you might get one decision but if you look at a string of events it might change the decision. For example the Saints deciding to go for it on 4th and 1, knowing that if they didn’t get the touchdown they would still have a chance to possibly get the ball back and have a successful field goal or touchdown. So I guess that would be maybe like a Markov chain.
February 8th, 2010 at 1:10 pm
I happened to catch the coin toss. Apparently, the NFC team has won the coin toss for 13 straight seasons now. The odds are 1 in 2^13 (8192), which Jim Nantz pointed out. However, the odds that some conference eventually has a streak 13 games long is somewhat more than that, since there are 44-12=32 years for a 13+ game streak to start as well as two conferences, resulting in a probability of 64/8192. I felt like an opportunity was missed there.
What are the odds that probability of rare streaks will be put in context?