I have a theory that football in particular, isn’t actually a very scientific sport. It seems to mostly be governed by lore, rule-of-thumb and tribal knowledge. Nothing about the NFL play or personnel choosing screams that the teams are operating in a science or evidence based fashion or anything like that. Except maybe the Patriots- I wonder how much is Brady vs. how much is Belichick doing something fundamentally different than other teams that we don’t know about.
Case in point- there have been studies showing that statistically in pro football, there’s some hay to be made by being more aggressive on fourth down instead of almost always punting. Yet the pros nearly ALWAYS punt on fourth down. Why? Because that’s the tribal knowledge. Personnel decisions seem to be made in an equally arbitrary fashion- concentrating on height, weight, speed, etc… and less on intangible stuff like “do they win games?”, “do their teammmates respect them?” and so forth.
So I think this is the same thing in part- a combination of the tribal knowledge of “3rd and 2 - run the ball” combined with a lame attempt to confound that tribal knowledge by doing the opposite and trying to pass in the middle/short range of the field, when that’s a lot less helpful than just running the ball- at least that way, you have a pretty good chance of getting at least a yard, which sets you up for 4th and 1… oh wait, they CAN’T run again on 4th and 1. Oops.
There’s no coaching schools to go to, certifications, or anything along those lines that companies valued at many billions of dollars would use to evaluate their chief operating officer candidates. Most places on that scale would require an MBA from a good school, as well as good performance. But there is nothing similar in the sports world, particularly football. There aren’t researchers writing papers on how to coach, or how to pick players. There aren’t coaching students learning this and writing articles for the journals in the off season.
Football coaching seems to be a system entirely dominated by connections and imperfect evaluation of on-field performance, where again, due to that tribal knowledge, a promising young coach could well be penalized for doing the smart thing and going for it on fourth down, because “everyone knows” that you don’t go for it on fourth down.
It’s all well and good that going against the grain (like, say, going for it on 4th instead of punting) might give you a slight statistical advantage, but the NFL isn’t a video game and it’s got an extremely small sample size of 16 games. Put yourself in the coach’s shoes:
a) You have a 50% chance to win playing traditionally, and win or lose both ownership and the fans understand and accept your decision making; you can’t win every game.
b) You have a 55% chance to win never punting, but if you lose (45% chance) there is a major outcry from the fanbase and ownership might fire you.
I think you should never punt when it is 4th and reasonably short and you are in your opponent’s half of the field. Either do a field goal or go for it. But on 4th-and-longs or when pinned deep in your own half, sure, nobody could fault you for being conservative.
It’s not Brady. He’s great but he’s not the whole team. When Brady was suspended following Deflategate and they had backup QBs, they did fine. There’s no question that he’s a big reason for their success but it’s mostly due to Belichick.
Right. Coaches don’t actually optimize for winning, they optimize for not getting fired. Those two goals are mostly the same, but in certain corner cases, like the one you outline, they don’t.
Marshawn was getting yardage almost every time he carried the ball that game, and there was time for at least 2 plays. Just because the defense is in a particular alignment doesn’t mean you can’t still run it right at them and gain a couple yards.
