Plagued by head injuries

Journalists predicting the end of the NFL cite lawsuits by current and former players suffering concussions and head injuries. Alongside these lawsuits are reports of violent controversies in players’ personal lives.

What’s worse is these two trends are related: according to vice sports, doctors examining former players at Boston University linked their football-related brain injuries to depression and impulse control. Heeding these injuries, parents of potential NFL players are forbidding participation in the sport, and conscientious fans are second-guessing their devotion to the game.

However, in spite of all this, I continue to watch football, as do an increasing number of Americans (and the English! American football is enjoyed by the mayor of London, Boris Johnson).

Ever-more powerful players

An aspect of sports that as always been of interest is the continuous advancement of athletes in the same sport over time. Athletes, over time with advances in nutrition and training seem to get better and better. How good they get and how long will it take to get there?

In American Football, players are getting ever more powerful. This has translated into a greater probability of conversion for any given play over time. This is shown in the scatterplot below, using data from . If you take this to its extreme where every play results in a first down, the game itself becomes too predictable to enjoy. One of the reasons sports (and sports betting) is so popular is because of the high degree of unpredictability.

If players are getting more and more powerful over time, they could become so powerful they are able to drive the ball to the endzone every possession.

The evidence agrees with the theory offensive plays are gaining more yards per play than the year prior. This relationship has a high degree of statistical signifance (the probability the relationship is random being 2.183e-07). Admittedly, the advancement is very small, about 0.026 yards per year.

Correspondingly, a strong and positive relationship was found between the likelihood of a play resulting in a conversion over time. The likelihood a play results in a first down, year over year, is increasing by 0.128% (in 2013, the percentage of plays resulting in a first down was 27.8%). Again, this is a rather trivial quantity, but the relationship shows a high degree of statistical significance (the likelihood this relationship is actually random is 1.782e-06).

So, long term, as football players get better and better, the game will get more and more predictable and less fun to watch.