Tuesday, February 5, 2013

HARBAUGH'S FOLLY AND HOLDING FOR FREE; Observations on the 2013 Superbowl


Two thoughts about the Superbowl (the “Har-bowl”) just concluded:

First, 49er's coach Jim Harbaugh made a fundamental game-playing error in the second half.

I know that football coaches, unlike other types of coaches, are rarely game-players. They don't play backgammon or chess or poker or Risk or other types of strategy games. They have more in common with military generals in that their primary task is to manage, and control, and inspire a bunch of young men and get them to achieve certain group goals. It's a very difficult job, especially considering the psyche of the modern pro, and anyone who can keep fifty immature, imperfectly-educated, testosterone-fueled, filthy rich world-class athletes with no manners on something like “the same page” has no time to learn how best to play games and make the correct strategic decisions at crucial moments. I understand that.

But why can't they hire somebody to do it? Why won't they hire some numbers geek to tell the Jim Harbaughs of the world when to punt, when to kick a field goal and (above all) when to try a two-point conversion? These are not “football decisions” like whether to blitz on a particular play or whether to replace your left tackle because he can't walk anymore, and they don't depend on how a particular coach “feels” at that moment. They are objective, numbers-based determinations to be made about what should be done in a specific situation in the game of football. And since they are objective, there is one correct course of action. And Jim Harbaugh, like many coaches, has no idea what that correct course of action might be. In fact, it is unlikely Harbaugh is even aware there is a correct course of action.

So why won't the 49ers hire somebody to tell him?

With 14:49 left in the 3rd quarter, the Ravens led 28 – 6. Jim Harbaugh was down by 22 points. What, at this point, were the 49er's chances, and what could they hope to accomplish? How could they possibly get back in this game, and win it?

The obvious answer, the starting point, is obvious: score three touchdowns and four extra points. That's the 22; that ties it up. Then you have to shut down the Ravens as well. This had to be what Harbaugh was thinking and hoping and planning for. And in fact, the almost-successful 22-point comeback was the entire story of the second half, up until the last couple of minutes.

And as luck would have it, 14:49 in the third quarter was the moment the San Francisco fortunes turned around. Seven minutes later, they scored their first touchdown of the game and presented Jim Harbaugh with a critical game-playing decision. Assuming their dreams would come true and they would score the three touchdowns they needed, how should they approach the question of the four extra points? When do they go for two?

There is one right answer to this question.

For the analysis that follows, for simplicity's sake, we will assume the chance of kicking a single extra point is 100% and the probability of success for a two-point conversion is 50%. The real probabilities, from years of NFL data, are very close to those numbers. A missed kick on an extra point is extremely rare, and NFL teams succeed on two-point tries almost exactly half the time.

Since the 49ers actually scored the three touchdowns they needed, we got to see Jim Harbaugh's decision on the critical strategic question of when to go for two. (A field goal by each side at the beginning of the fourth quarter did not change the 22-point comeback equation.) He chose to take single points after the first and second touchdowns and try for two only after the third touchdown. Since the chance of making a single two-point conversion is 50%, this meant that Harbaugh's strategy had only a 50% chance to tie the game even if his 49ers succeeded in scoring the three touchdowns and holding the Ravens. This was not the optimal strategy.

In this situation, down 22 points, when a coach needs one two-point conversion to go with his three touchdowns, he must go for it at the first opportunity, after the first touchdown. If he makes it, he need only kick single points after the next two scores. But even if he fails, he has an additional chance. He can still get his four extra points with two-point conversions on both the second and third touchdowns, and that possibility would have provided an additional 12½% chance of success. By pursuing his optimal strategy and going for two on his first touchdown, Harbaugh would have given himself a 62½% probability of scoring 22 points on the three touchdowns. By waiting until the third touchdown to try for two, however, he limited himself to only a 50% chance of tying the game. And he gained exactly nothing by doing this. His failure to understand the situation (or his failure to have somebody on the sidelines capable of making the right decision for him) simply cost the 49ers that 12½% chance.

This is not to say the 49ers would have tied the game even if Harbaugh had done the right thing, of course. And even if the 49ers had tied up the game with ten minutes left (when the third touchdown was scored), there was no guarantee they would ultimately win it. But that cannot excuse Harbaugh's decision. This was a fairly simple math and logic problem, and he got it wrong in the most important game of the year and possibly the biggest game of his career. There is no logically-defensible reason for what he did.

Elsewhere, I have written about the 15-point deficit in pro football, and why it is best for the trailing team to go for two on the first touchdown in its comeback. The reason is that 15 points is not “two scores.” It's either two scores or three scores and you won't know which it is until you try the two-point conversion. Since it is essential for the trailing team to know whether the 15 points is two scores or three scores, the team should go for two as soon as it can. If it fails in the attempt and trails by 9, it will at least know that it has failed and will have to score twice more to overcome the 9-point deficit. There is really no reason to wait.

If a team is trailing by 15 points, however, and waits until its second touchdown to go for two, it does not actually reduce its chances of tying the game. A team down by 15 needs one two-point conversion and their probability of getting that two-point conversion is the same whether they try it after the first or the second touchdown. This is not true when the gap is 22 points (or 29 or 36 or 43), however. What Jim Harbaugh did significantly reduced the 49er's chances of tying (and ultimately winning) the Superbowl.

*

On the Ravens' intentional safety at the end of the game, the ball was snapped with twelve seconds left. The punter then ran around a bit before stepping out of the end zone with four ticks on the clock. It should have been easy to run out the clock on this play and the Ravens failed to do so.

The Ravens had ten blockers. The 49ers had, at most, ten guys rushing the punter. All the Ravens have to do on this play is assign a player to each rusher and have him wrap his arms around the guy until the clock expires. If necessary or desirable, each Raven could simply pick one guy, tackle him and sit on him. There would be flags all over the field, but (as Hillary would put it), what difference does it make? The penalty for offensive holding in this situation is to award the defense a safety.

In effect, on an intentional safety, there is no penalty for holding.

Copyright2013MichaelKubacki


No comments:

Post a Comment