This is my reason and one that may supercede the actual immediate causes to which the NFL does not allow technology to pervade the sport. The theory is based on protective reasoning and that the NFL wants to ensure the monopoly is held as a status quo. Anytime we find that the technology and innovation create a competitive advantage or an advantage that cannot readily be seized by those in power, we see resistance.

The NFL has banned the use of certain technologies to be used prior the game. Specifically, iPads cannot be used two hours prior to a regular season game. This is odd and instead of iPads on the sidelines, reviewing and understanding the previously plays, we have archane still shot print outs from laser jet printers. Laser jets people, geez.

The point is, “NFL no like technology”. Why?

Applying my extremism theory, we come to the following application, that, if I had the capability to create, would create.

I would create the Artificial Intelligence Coaching Assistant, otherwise known as AICASS. AICASS would be a system of algorithms that can accruately predict the play that the offense or defense would be running. This system would include the following inputs that would lead to calculations that would statistically predict the next play within a certain degree of error. The following inputs will be written from the point of view of the defense.

Input 1: Prior game film: Each of the previous games leads to percentage chances that a certain play will be run based on prior down, distance, quarter and time left in the specific period. These can be calculated and estimated as a predictor of what play is to be called in the similar situation in the current game. E.g. 31.4% chance this team throws the fade (long pass) on first down when the formation has four wide receivers and the outside receivers are lined up inside the hash/numbers.

Input 2: Football game theory: Along with prior playing calling statistics, there are general rules of the game that are used for additional explaining power. It is estimated to have less predictive power, but it would be influential. E.g. 65% chance that the defense will allow the offense to score a touchdown if there are 56 seconds to play in the game and the defensive team is winning by four points.

Input 3: Specific tendacies based on formations, down and distance, time on clock: Most of us know that the offense typically runs the ball on first down, but what about when it is third and four and they are on the fifty yard line? This is where the intersection of input 1 and input 3 could be combined as the team’s tendacies would carry over from prior games and estimate the current play probablity. A team’s formation, down/distance, time, location on field and prior plays would create a prediction for the current play. E.g. Having three running backs in the back field would indicate a run play with fourth and one on the fifty yard line. (A computer is not needed for this example, but extrapolated and we can see the potential impact).

Input 4: Past and Real Time Individual player indicators: Did you know that a player’s gait (the way they walk/run) can indicate if that player is likely to be the focus of the pass? Yes, even in Division 3 football, we would watch enough film to pick up on little indicators that the receivers would show to the defensive backs. The computer and camera system would see these indicators and highlight on the LCD display this behavior. The computer could also estimate the tireness of a player and provide a status percent of their effort level on the next play. E.g. Receivers would sometimes hold their hands higher or lower depending on the play being a run or pass. Additionally, the amount of weight an offensive linemen places on their hands (knuckles white) can indicate the play type. (tired and morale)

Input 5: Real-time coach indicators based on facial and body indicators: Coaches sometimes exhibit behavior that can be gleaned from their face or body posture that could predict the next play call. A coach standing or appearing aggressive may indicate that on second down and short, there is an aggressive play call to gain more yards than required by the first down. The human body can be a large signal and produce surprisingly predictive indicators.

That is probably enough inputs for now, given that there is a serious amount of testing and verification of which inputs actually have the causality to produce accrurate probablities. But the prospect is entertaining and inevitable. I do hope that someone from Stanford or MIT or Missouri Rolla, reads this post and feels inspired. (I will help you formulate the research…as best as I can).

This type of analysis will lead to an artificial intelligence that would surpass the ability of humans to accurately predict the next offensive play. Such that, it would force the offense to use the same system to predict the defensive plays and take similarly calculated counter measures in order to disrupt the same tendacies that were being created by the offensive history and individual players. Suddenly, the route or technique of running is not soley depend on the six seconds of actual play, but the way the player behaves prior to the play. Coaches will no longer show their faces on the sideline for fear of facial readings (not unlike the current practice of covering your mouth while calling a play).

The impact of such a system would lead computer scientists creating algorithms with behaviorial scientists and psychologists, which would lead to a war of algothrims. Mathematicians would fall under the salary cap and players would become secondary pieces on the field of football as eletronic calculations become a key strategic hire.

In fact, the technology would expand and each player would have a translucent LCD display over their facemask to show the calculated play to run at the calculated time. This would put football on the same level of air combat, perhaps even more complex given the number of football plays called at any point in time is greater than the number of defensive air combat manuevers that take place during a dog fight (S curve left Maverick).

Have you seen the iconic 1980’s movie called War Games? Perhaps teams will just come to a giant TV screen (such as the one in Dallas) and allow their algorithms to play each other. Why even have the players? Why bother with offseason training or forcing players to maintain inhumane weight? We could simulate the entire game based on electronic inputs and algothrims. Such is the Theory of Extremism.

Coaches, players and the like are now no longer needed. Okay, this last portion may be a bit of an exaggeration, but the point is proven, technology would, as in other industries, render the current status quo inept and inefficient. In conclusion, I plan to take the Robotics course from Udacity this month.

-EricEricOut

discussion can be had at hackernews