In sports, athletes and coaches alike are forced to make decisions. There are several different factors taken into account in those decision making processes, not the least of which is what the other team (coach, player or team as a whole) will do. This is evidenced particularly well in football and baseball.
Kenneth Kovash and Steven Levitt (one of the authors of Freakonomics) published a paper in September of 2009 looking at the game theory implications in major league baseball and football. Their conclusion: Pitchers throw too many fastballs and NFL coaches don’t dial up enough pass plays. They predicted that if pitchers were more creative and less predictable in their pitches thrown, it could mean at least 2 more victories per year for a team. This doesn’t seem like much, but when, for exapmle, a team like the 2014 Mariners miss the playoffs by 1 game, those 2 extra victories are a big difference. Levitt and Kovash also consluded that if coaches mixed up their play-calling and ran more passing plays, that could amount to half a win extra per year. Again, this isn’t a huge difference, but in sports, any legal and fair advantage is one worth taking.
This post will focus on the football side of things, and culminate with a look at the final plays of the most recent Super Bowl, which ironically ended with an untimely pass-play called.
In football, running the ball is statistically much safer and more consistent. My high school football coach would always say “God wants 4” (I went to a Catholic school). What he meant by this was simply that if we advance on average 4 yards on every play, we will score every possession and presumably win most games. What he didn’t think about though, was that running the ball with hopes of 4 yards every time, though a small amount, gets a whole lot tougher when the other team knows, with certainty, that you’re running the ball. They don’t have to defend the pass; so they load the box, blitz often, and the whole “God wants 4” line doesn’t seem all that simple anymore.
But the same logic, though obviously more advanced, applies in the NFL too. Handing the ball to a running back who is paid to not fumble and also gain yards is a whole lot less risky than putting the ball up in the air for a reciever to make a play on. This is why a majority of plays called in the NFL are run plays. But what if coaches switched it up a bit?
Levitt and Kovash used a study of 125,000 plays from 2001 to 2005 to put together their statistics for the study. They concluded that “a pass on average gains .55 yards more than a run, is 9 percentage points more likely to yield a first down, and leads to scores with a 3.8 percent probability.” As for scoring, “runs have only a 2.8 percent scoring probability,” though they conceded that runs also have a lower probability of a turnover. According to Levitt and Kovash, if a team passed 70% of the time (as opposed to the then average of 56%), they could score an additional 10 points over an entire season. That would be an extra 3% of their scoring. But these stats could go both ways. Either an offense could start calling more pass-plays, or a defense could start playing the run more often.
Then the authors looked at decisions following a pass call:
“A team that has passed is 10 percentage points less likely to pass on the next play. After a passing play with a poor outcome, a team is 14.5 percent points more likely to switch from a pass to a run on the next play (or vice versa), even after controlling for the down and distance.”
Using this information, a defense could alter its play calling and would give up, according to Levitt and Kovash, an average of 10.8 fewer yards per game. They anticipated this would be about half a victory more per season.
So did Pete Carroll, and/or Bill Belichick make the right decisions?
Let’s fast forward to after Jermaine Kerse made his improbable catch to bring the Hawks within the 10-yard line. For this analysis, we must assume that Carroll was playing for a potential 4th down if he didn’t score on 1st, 2nd or 3rd.
The first obvious call would be a run with the most reliable running back in the league, which they had dialed up. But the team was slow and disorganized getting to the line, forcing an unnecessary 2nd timeout called by Carroll.
After Marshawn was stuffed at the 1, Belichick had a choice to make: Call a timeout assuming the Seahawks score so Brady and company have enough time to try for their own touchdown, or let the clock run. He let it run.
Now, moments before the ball is snapped on 2nd down, Carroll has some considerations to make. If he calls a timeout before this play, his team will have about half a minute to run 3 plays. This means at least one, likely two, will have to be a pass play. That is generally bad play calling on the 1-yard line. He could call a run on 2nd, call a timeout after if they don’t score, and probably have time to run two more plays IF the 3rd down call is a pass that would stop the clock if incomplete. But in a game theory lens, Carroll knew Belichick knew these options. Belichick could then play a hard pass defense on 3rd and dare Wilson to throw a perfect pass, or run with Marshawn Lynch and risk not scoring and the clock running out. Belichick knew Lynch was a force to be reckoned with, and thus sent his run stopping defense out on 1st down, most of which then had to stay on the field for 2nd down given no timeout called. Carroll liked the idea of a pass here vs. on 3rd, especially with a run defense package on the field. That’s what he called.
Now, I’m not going to argue whether the particular pass called was a good decision, because it wasn’t. A shallow slant is the last thing you call on 2nd and goal. Throw it to the corner to Matthews, or to the back of the endzone; roll Russ out and let him be an athlete and make a play.
But what I will argue is that Pete Carroll did make the right choice to pass on 2nd. He was acting on all the information he had at his disposal, and operating on the assumption that the Patriots defense might force a 4th down. He wanted enough time on the clock to run a final play on 4th and the only way to do that was to pass on 2nd or 3rd. If they don’t pass on 2nd, they have to on third and then it becomes that much harder with Belichick knowing that as well. And with a run defense on the field for 2nd, the cards aligned, but the particulars did not.
Had a fade to the corner been thrown, who knows what would have happened…
References: http://www.nber.org/digest/oct09/w15347.html
Whitman ECON 215: Behavioral Economics 