Going for it!

Disclaimer: this post talks about economics, statistics, and football. If you are a curmudgeonly old man who thinks statistics have no place in sports, and are scared of phrases like "expected points value," don't read any further. Just go back to calling David Eckstein the best player in baseball and listening to Joe Morgan make fun of Moneyball.

Tonight, December 1, 2007, the top two college teams in the country are playing football. Missouri is number 1 (I don't mean this like their cheerleaders mean it. I only mean they are ranked number 1). West Virginia is number 2. Remarkably, they both had the same situation early in the game. Both had fourth down and goal to go from the two yard line. Both kicked field goals. Both did the wrong things, at least if you think scoring more points than your opponent is the right thing to do in football.

Here is an article by an economics professor at Berkely (or, as it's known in the college football game, Cal): http://www.econ.berkeley.edu/users/dromer/papers/PAPER_NFL_JULY05_FORWEB_CORRECTED.pdf

The article contains statistics showing that football teams going for it on fourth and goal from the two score a touchdown about sixty-four percent of the time (the figure in on page 16). That means, on average, the expected point value of going for it is 4.46 points, since they will get seven points if they score a touchdown, and 64 percent times seven is 4.46. If they go for the field goal, they have a nearly 100 percent chance of scoring three points, so three is their expected points value. 4.46 is more than three.

But the case for going for it gets more attractive. If you go for it and fail, your opponents will take the ball at the two yard line. If a team starts a drive at the two yard line, they average scoring -0.2 points (because of the possibility of a safety). If you kick a field goal and make it, the other team will take the ball over at around the twenty-five yard line. Teams score an average of 1.86 points from the twenty-five yard line. So the real expected value of going for it is 4.66 points--4.46 plus -0.2--and kicking the field goal drops to 1.14--three minus 1.86.

This guy gives plenty of reasons why coaches usually kick it in the situation Missouri and West Virginia found themselves. But the biggest one has to be this: coaches always go for it in that situation. Commentators call it "a smart play" to take the points you know you will get, and call it "rolling the dice" for taking a sixty-four percent chance of scoring seven points, and a far worse situation for your opponent even if you fail. Fans--even smart fans--usually don't think about these percentages. So if no one is going to call you out on making a far worse call, and they will hassle you for making the far better call, you go for the far worse call.

And, as of this writing, Missouri is losing by eight, and West Virginia only leads 4-7 Pitt by four.

--Paul

Update: Both teams lost; Missouri by a lot, West Virginia by less than a touchdown.

I guess Kirk gets frustrated easily

Watching ESPN can be frustrating sometimes. They hire terrible commentators like Skip Bayless, John Kruk, and Lou Holtz. Their NBA shoot-around--to which they dedicate a half an hour, three times a week--features Steven A. Smith and Bill Walton.

That said, Kirk Herbstreit is great. He co-hosts College Gameday with Lee Corso (who knows nothing about college football, but who dresses up like the mascot of the team he likes to win the big game of the week.) The show airs once a week, on Saturday mornings, and reviews and previews college football. Kirk is generally well prepared and knowledgable about the subject he gets paid to be knowledgeable about: college football.

But this morning, gameday decided to veer from the game of football, and instead to talk about race; specifically race in college football coaching. The conversation went about how you'd expect. All lamented about how few black men hold the title of head coach for division I football teams. All said we should do more to fix the problem. So far, I had about the same reaction I have when Barbara Streisand endorsed Hillary Clinton. "That's great, but I pay you to sing/discuss football. Your opinion on the subject carries as much weight as do my thoughts on gardening."

Then Kirk said this: "It's not just the head coaches. There are 240 offensive and defensive coordinators [the second rung on the coaching ladder] in division I football. Of those, only 28 are black. That's whats so frustrating to me."

First, kudos on actually giving statistics, Kirk. Most discussions of race and how little progress we've made go like this:

Person: "We have made little progress on race."
Me: "Really? Richard Parsons heads the largest entertainment company in the world. Barak Obama leads polls in Iowa. Billy Martin is, if not the best private lawyer in America, at least in the top ten. They are all black. It seems to me making progress must mean seeing racial minorities in top positions in business, law, and politics."
Person: "You are a racist."

But here is the problem with what Kirk said. 28 of 240 is 11.6 percent. According to the U.S. census, African-Americans make up 13.3 percent of America's population. That is under-representation of 1.7 percent. Or, as the economists call it, statistically insignificant. Is this really enough of an under-representation to make Kirk frustrated? To be fair, I started this post by saying I get frustrated because a sports station sometimes put people on the air I don't like. But I get frustrated easily. I guess Kirk does too.