This past weekend was an exciting week for college (American) football. During the game I watched, however, a thought struck me: was it really a good idea for these teams to be running out the clock?
Michigan was battling Notre Dame in Ann Arbor. Both of these teams have had some recent awkwardly bad seasons, but the rivalry is big enough that this was sure to be a good test.
The last quarter of the game was full of both sides scoring only touchdowns (usually worth 7 points) as each tried to get their offense past the opposing defense. In American Football, at any given time one team has their offensive players out, while the opponents have their defense out. The offense tries to advance down the field with the ball while the defense tries to stop them. A touchdown is scored by getting one of your players through the defense and down to the other end of the field. How's that for a simplification?
The fourth quarter consisted of Michigan's offense scoring a touchdown, putting them 11 points ahead (there are other ways to score points, but let's forget about that for a second). Notre Dame's offense then had the ball, and scored their own touchdown. Michigan had the ball again, but failed to score, thus Notre Dame had another shot and scored again, putting them up by 3 points. Michigan attempted to score and failed, then Notre Dame had another go, but failed as well. On their next turn to be offense, Michigan scored another touchdown, putting them on top by 4 points. With only 11 seconds left to play, Notre Dame did not score again.
The exact amount of time remaining is a big part of the strategies that I think are a little short-sighted, and it happens with every team. After both teams had scored their first touchdown of the quarter and with Michigan up by 4 points, Michigan had the ball with about 8 minutes left to go in the game. They failed to score on this drive, but they used tactics that spent more time on the clock. The logic is that since they were ahead, "running out the clock" would give the opponents less time in which to score.
Of course, after Michigan failed to score and Notre Dame responded by getting a touchdown themselves, the Wolverines were suddenly fighting against the clock. Now it was them who needed to score within the remaining time.
This pattern went back and forth a few times, with the clock obviously running out on Notre Dame at the end. This strategy continues to confuse me, however. For a large portion of the game, Notre Dame showed that they were able to move the ball very well, meaning that if they were on the offense, they had a very good chance of being able to score. With eight minutes left to go in the game, it seems more likely to consider: "How long will the rest of our offense last? If they get possession again, will they be able to score? If so, will they be able to eat up the rest of the time during their possession, so that the clock actually runs out on us?" Michigan barely scored their last touchdown in time; the running of the clock earlier was almost their undoing!
This really starts to look a bit like an impartial game. Instead of determining how to control as much of the time of possession as possible, how can we instead control the parity of the clock so that on our last drive towards the end zone, we will have enough time, but the opponent won't have much after we're done.
Naturally, there are a lot of probabilistic factors in there, ignoring the fact that I've over simplified things. Still, it seems like some better form of analysis could be employed there.
A Domino-Covering Problem
2 weeks ago