A New Analytical Philosophy

 

 

When trying to quantify football performance, you would be hard pressed to have any trouble locating any one of many football analytics, metrics, stats, etc. From EPA, DVOA and QBR, to Points Per Game, Yards Per Rush, and QB Rating. There is a stat or metric for just about everything. What you may notice the longer you follow us here at Football Behavior, we don’t utilize any of them. Now, let’s pause so I can be abundantly clear: this is NOT an anti-analytics take. All of the aforementioned stats have their place in helping to provide some level of understanding to what we’re seeing on the football field.

 

 

Rather, this is more about how analytics are VERY important, contingent upon you using the right ones at the right times. We don’t use the ones mentioned in any of our analysis or our modeling because there are two very distinct things about each of them that go against our laws here at FBx.

 

  1. They rely on historical averages.
  • In our opinion when assessing across multiple unique environments, historical averages, and averages in general, are stistcically meaningless when analyzing human behavior with the goal of making future predictions about it. Behavior is so dynamic and under specific controls that historical averages as a baseline amount to nothing more than a statistical guess when making predictions.
  1. They treat the NFL as a singular universe where all participants are participating in the same environment.
  • The only thing for every player that is uniform in the NFL is the dimensions of the field, the ball, and the rule book. That is it. In reality the NFL is a multiverse of madness made up of 32 independent and unique playing environments that really can’t be standardized when trying to assess and predict human behavior. The above stats don’t provide enough environmental context, which matters for our third law in particular.

 

This is not to say that they aren’t useful. They just aren’t useful for what WE do as behaviorists. Human Behavior is so variable on just an individual level, but then to try and predict the behavior of a group, nay, 32 groups, we need something that truly captures the essence of what it is we are trying to analyze.

Using other metrics will yield us some correlative context. But it likely won’t tell us what it is we actually need to know. And in some cases they aren’t really telling us things we didn’t already know.

 

 

The Three Laws Of Football Behavior

 

Like Asimov’s Robots, Football Behavior Has Three Laws

 

 

Which brings us to why we developed the philosophy of Football Behavior. If you have subbed with us for a while now, you have heard us use terms like “Scoring Behavior” or “Score Prevention Behavior”. You’ve also heard us refer to “bounce rate” and “bounce envelope”.

These are all measurements we use in our analysis based on the principles of B.F. Skinner’s “Behaviorism” and Ogden Lindsley’s “Standard Celeration Chart”. Simply put, we are applying the principles of behavior to a complex area of human behavior like football, a sports where 53 individuals must compete in practice and in games, performing different behaviors, but towards a common goal.

 

 

Skinner’s Radical Behaviorism Is The Foundation For Football Behavior

 

 

To understand howe do that, and why, we must first understand what we are calling the “Laws of Football Behavior”. Much like Asimov’s “Laws of Robotics”, this is a set of jumping off points that illustrates what we’re doing and gives us place of common ground to start to discussion from. These laws are something we must all agree on as fans and observers of football. If we can’t agree on these, nothing else makes sense.

 

 

Now, the laws:

 

 

  1. The offense’s primary job is to score points, and it can only score points when it possesses the ball.

     

  2. The defense’s primary job is to prevent the opposing offense from scoring points, and it can only prevent the scoring of points when their opponent possess the ball.

     

  3. All behavior is repeatable and must occur across units of time. Behavior does not happen in a vacuum. We refer to this as Repeatability, Temporal Locus, and Temporal Extent. Football is a time-capped sport, and there is finite amount of it for each of the offense and defense to perform its primary jobs.

     

If you can agree to those three laws, understanding the rest of Football Behavior is much simpler. Basically, time and what with you do with it on a football field, is a strong indicator of team behavior. You never heard of the “two-down” offense or the “four-down” offense.

It is the “two-minute” and “four-minute” offense for a reason.

Time is the key indicator of environmental context for our behavior. And that is especially true in the sport of football because it is a timed sport. You only 60 minutes, and maybe a 10 minute overtime period. But then that is it.

So, before we move along with this in the coming weeks, I want to hear from you. Do you agree with the three laws of Football Behavior? If so, would you add anything else?

More importantly, if not, why not? Reply to this email with your answer, and make sure to answer the poll!

Do You Agree With The Three Laws Of Football Behavior?