Tuesday, January 15, 2013

John Fox and "that" kneel down

I hadn't planned to write a new post until after the season was done, but I feel compelled to now because of something that happened this weekend in the Broncos/Ravens game, and the reaction of both fans and the media to it.

Yes, we're talking about the Broncos decision to take a knee on their 20 yard line with 31 seconds left on the clock and two timeouts remaining.
Since then Fox has been ripped apart as various analysts, journalists and a mountain of fans pile in to offer their explanations as to why it was a dumb decision.

Now let's be clear before we even start; I DO NOT like John Fox a whole lot. I think over the course of his career as a head coach he's made some incredibly poor decisions. But this isn't one of them. 

What has annoyed me the most is that a lot of fans are turning to statistics to explain their grievance. Statistics are to football what CrossFit is to fitness; a vastly over hyped, completely unproven, poorly understood phenomenon that survives mainly on account of its followers loyal devotion to it than any basis in genuine results.

Take a couple of the responses to an article on ProFootballTalk for example;

"Fox and Elway should be ashamed of themselves for not being able to figure out the statistical odds of how that only lowered their chances of winning the game."

"You don’t get to pick the time, chief. Clock’s winding down. Best time to go for the jugular also happens to be potentially your last time. Lucky you didn’t give up a touchdown that ended the game."

"you can’t win when you’re playing not to lose."

"It seems absurd to not even try to get a few quick passes to setup a field goal. If you go three and out, then so be it. Playing not to lose often results in losses."

All of the above responses show a profound lack of understanding about what really happens on a football field, along with a slavish devotion to the idea that statistics can solve the problems on a football field, which includes comments like "playing not to lose".

The fact is that statistics have utterly failed to improve the ability of people to predict football games or to make sense of why certain teams win a lot etc. The reason for this should be obvious to anyone who has even the remotest understanding of the application of statistics in the real world.

Statistical analysis requires three things to be of use; 1) a low number of variables affecting the outcomes, 2) a large quantity of data to draw theories from and 3) a good understanding of which variables are most important and how they affect outcomes.

Football lacks 1 and 2, while my experience has been that the people who study football statistics lack number 3. 

Football is not a maths equation. You have - at any one time - 22 human beings running around on the field, supervised by a another group of almost 10 human beings, with information and plans being fed onto the field by about another 20 human beings. That's before we get round to issues like weather, time, distance travelled, temperature, fatigue levels, etc.

Football is the anti-thesis of what statisticians are looking for. It has a huge number of variables on every play, there are only 16 games for one team in any regular season, at most 19-20 for a potential super bowl champion, and most of the people that study the numbers of football have demonstrated that they generally seem to know the precise sum of d**k all about how football is actually played. 

I get the distinct impression they spend more time looking at the numbers than they do actually watching the games. If they did then they probably wouldn't say some of the incredibly stupid things that they do and wouldn't need to have very obvious things explained to them to fill in the gaps that their data has no answers for.

John Fox's explanation for why he didn't go for it was that he felt his team was in shock after giving up such a late touchdown. He described his team with a boxing analogy as being on the ropes, which he said wasn't the right time to go for a knockout punch.

And mark your calendars people, because for once I'm going to agree with John Fox (I feel dirty now).

What people who study the numbers of football always deny is that emotion has an impact on the game. They play it down as being a fiction created by the TV talking heads to explain certain events in a football game and then dismiss it out of hand.

The real reason they reject emotion - and other human factors like it - is simple. It's because it can't be measured on a spreadsheet. And if it can't be measured on a spreadsheet then its affects on the game can't be used alongside their other numbers to help analyse football like it were some math problem.

To deny the impact of emotion on football players is to deny the fact that they are human beings. And you can sit there and argue all day long that they get paid millions of dollars etc, but that doesn't change the fact that they are human beings. Human beings f**k things up. They get nervous, they get excited, they get worn down, they get elated. Emotions affect the way that they play a game like football.

No matter how many spreadsheets you produce. No matter how many numbers you crunch. No matter how many formulas you create. You will never, NEVER, be able to dismiss the human aspects of a football game. It just doesn't work like that. You don't get to discount the huge impact of humanity on football just because you have no way of quantifying it for your little math model. 


All the people saying that Fox "reduced his chances of winning the game" or "played to lose" fail to understand that sometimes your team is just not mentally and emotionally up to the challenge right now. And nobody understands that better than the coaches who work with those players all season long.

To try and push a team that is mentally frustrated, down beat, in some cases despairing and generally in a very negative frame of mind, is to invite disaster onto yourself. 

Trying to force a miracle under those circumstances just because some idiot with a spreadsheet says that you have a slightly better percentage chance of winning if you go for it is to put your quarterback in a tough spot where he feels like he has to make something happen, which is a perfect recipe for a pick.

Let's not forget this; the Broncos were on their own twenty. Realistically speaking they would have had to have moved the ball 50 damn yards to give their kicker any real hope of making the kick. In 31 seconds. Even with two time outs that is ridiculous. You're not going to do that without pushing the ball down field.

And that's pushing the ball down field against a defense that is ready and waiting, that knows you have no choice but to throw the ball deep, because screens and dump offs are not going to get you 50 yards in 30 seconds.

What was I saying earlier about a recipe for throwing an interception? Sounds a lot like it to me. And if you give up the pick then in all likely hood you give up the touchdown, or at the very least hand the other team the field position they need to kick the winning field goal.

I give credit to Fox for a change (I'm gonna need a wash after this). He must have known that by taking the knee he would get chewed out by the fans and the press. But frankly given those circumstances, on that day, he made the best decision he could. 

I guaran-damn-tee you this as well, if he had said "right, we're going for it" and had tried to get the yards for the kick, and Manning had thrown a pick (or there was a fumble) then you know he would have been eaten alive even more voraciously than he has been already. People would be asking him why he didn't just take the knee and play for over time.

Now you could argue that maybe Fox could have tried one or two running plays, and I except that's a course of action that might have been considered, presenting as it does a lower risk of something going seriously wrong, but it would have been unlikely to get the yards needed and in all fairness wasn't really a viable alternative.

I think Fox made the right choice (for a change) and actually deserves some credit for making the best decision for his team under that set of circumstances, as opposed to being vilified because he didn't take the fantasy option that spreadsheet coaches are sure would have won him the game.

It just really annoys me when people talk about "oh this team has a 60% chance of winning when they do xyz, because that's how the league has averaged out over the last 8 years etc...".

Coaches don't get to play those sorts of averages. There is no such thing. Those results, those percentages you're talking about, the data that makes them up was generated by teams as widely different from each other as the Patriots to the Browns, the Steelers to the Chiefs, the 49ers to the Chargers. 

That is not a sound basis for making critical decisions in a game. Fox made the best decision he could, basically the only viable decision he could under the circumstances. People just need to deal with it.

No comments: