Home / General / Coaches make bad decisions in part because dum dum fans can’t understand simple stats

Coaches make bad decisions in part because dum dum fans can’t understand simple stats


It’s often been noted that coaches are much more likely to get fired for decisions that don’t work out that also violate the conventional wisdom than they are for decisions that don’t work out but at least adhere to the CW.

I give you Denver Broncos’ head coach Nathaniel Hackett:

Denver Broncos head coach Nathaniel Hackett almost made another awful decision against the San Francisco 49ers.

Hackett and the Broncos hired a time management coach of sorts in Jerry Rosburg this week, as the first-time head man had a tough time dealing with the ins and outs of in-game strategy. In due time, he will learn, but for now Hackett has an extra hand to help out when the situation calls for it.

Against the 49ers on Sunday Night Football, Rosburg convinced Hackett NOT to go for it on fourth down from his own 34-yard line. While some fans in Denver would’ve liked that aggressiveness, that’s far too close to his own goal line to make such a rash decision.

I don’t care if it’s fourth and inches. Going for it from your own 34-yard line is a rather large mistake that can backfire instantly, especially given how the Broncos were moving the ball on Sunday night.

The Broncos had fourth and two inches just short of their own 35, 12 minutes to go, down 10-5. Punting in that situation is completely idiotic. You know, according to the stats, which show teams are successful converting more than 80% of the time when they go for it in similar situations. The Broncos were at that point likely to get the ball back no more than twice, and they were going to have to score at least one touchdown, after scoring exactly zero in their first ten possessions. Giving up a possession because of a less than 20% chance of turning it over on your own 35 instead of on the 49er 25 (a punt is a turnover) makes no sense whatsoever.

This wasn’t the end of Hackett’s expertly advised horrible decisions though. When Denver got the ball on its own 42 with 2:06 to go and a one-point lead, Hackett decided that the smart thing to do would be to run the ball into a goal line defense three times in a row, so he could give the ball back to SF fifteen seconds later. Seriously SF had nine guys in the box on first down, TEN on second down, and backed off only slightly on third and nine. Calling three run plays there probably had about a 5% chance of getting a first down, as opposed to letting your $250 million QB take a shot at ending the game.

The sequence made me wonder whether Hackett, who calls the plays, wouldn’t give Russell Wilson the option of checking out of one or more of those plays in that situation. The second down play was a sweep directly into ten guys, i.e., a play that Wilson could obviously see from the pre-snap read was totally DOA. So I assume he wasn’t allowed to change it. The result of all this is that Denver used up 15 of the remaining 126 seconds before giving the ball back to SF, with the 49ers and their great FG kicker needing just three points to win, which they probably would have done if not for a great strip fumble from the Denver defense, that won the game pretty much by itself seven different times. I mean giving an NFL offense, or maybe even the Lions, almost two full minutes with a time out still in hand to get the 55 or so yards they would need to get in Robbie Gold’s FG range is a horrible gamble, especially since coaches and fans both tend to forget how much harder it is to stop an offense that is using four downs on every series instead of three.

But it all worked out this time so I guess that shows why you can’t trust statistics. Or something.

  • Facebook
  • Twitter
  • Google+
  • Linkedin
  • Pinterest
It is main inner container footer text