Tuesday, September 28, 2010
Debating McCarthy's endgame approach
By Kevin Seifert
In giving the Green Bay Packers a Free Head Exam a few hours ago, I agreed with coach Mike McCarthy’s approach in the final two minutes of a 20-17 loss to the Chicago Bears. The exam included this passage:
McCarthy said he did not consider allowing the Bears to score a quick touchdown to get the ball back with more time on the clock, and I'm on board with that. In my opinion, you always force the opponent to execute flawlessly under pressure in that situation. The chances of a miscue are higher than driving the length of the field with one timeout to score a game-tying touchdown.
As it turns out, some people much smarter than me have proven that final sentence inaccurate. Over at Advanced NFL Stats, Brian Burke’s historical research showed that teams have lost 97 percent of the time when they followed McCarthy’s approach by allowing the clock to wind down for a last-second gimmee field goal.
But had McCarthy instructed his defense to allow Matt Forte to score on a first-down run, thus preserving a timeout and getting the ball back with 1 minute, 40 seconds remaining, the Packers actually would have incrementally improved their chances of winning. Historically, teams have lost 90 percent of the time when put in the situation the Packers would have been in.
Burke: By not allowing CHI to score the TD on 1st down, McCarthy cut his chances of winning from about 10% to about 3%. Neither prospect is very appealing, but every little bit matters.
Meanwhile, our friends over at AccuScore addressed the question using computer simulations of the scenario. (Remember, AccuScore develops digital profiles of every NFL player, team and coach, using them to run simulated games and seasons to predict real outcomes.)
Of 10,000 simulations where Gould kicked the field goal, the Packers won 2.1 percent of the time. But in the 10,000 simulations where Forte was allowed to score on the first play, the Packers went on to win 15.4 percent of the time in overtime.
What does this information tell us? Allowing a team to score goes against every ounce of conventional football wisdom. But at the very least, that conventional wisdom can and should be questioned here. I’m still not willing to criticize McCarthy for the approach he took, but it’s interesting to realize it would have been a statistically defensible position to allow Forte to score.
Either way, the Packers’ chances of winning were remote after the Bears moved inside their 10-yard line. But if you believe in degrees of remote, the numbers are interesting.