- Thread starter
- #1
TDs3nOut
Well-Known Member
An old axiom in football is that holding on to the ball, getting first downs, keeping the clock running, and keeping the opposing team’s offense off of the field are all important for a team’s defense keeping the opposition off the scoreboard. Of course, that axiom ignores just how good a defense is to begin with, regardless of how long it’s left on the field.
In order to investigate the relationship between a team’s time of possession and the number of points that it gives up, I used data from every team in the league last year to estimate the following linear regression model:
Estimated Points Allowed/Game = 51.2-55.2* Average % Time of Possession.
For example, the team that most closely fits that model is TB, a team that gave up 24.3 points per game and possessed the ball 49.3% of the time in its games. So, the estimated model predicts that the Bucs would give up 51.2-55.2(.493)=24.1, which is quite close to the 24.3 points per game that the Bucs actually gave up.
Some teams, however, do not fit the model at all. For example, Seattle gave up only 14.4 points per game, the fewest of any team in the league, yet they possessed the ball only 50.8% of the time, which, according to the model results in an estimated average points against of 23.2 points per game. At the other end of the spectrum are the Bears, who had possession of the ball 51.7% of the time, yet gave up 29.9 points per game.
So, what do you think? Does time of possession systematically affect the ability of a team to prevent the opposition form scoring, or is it just a matter of the particular characteristics of the team that matters?
In order to investigate the relationship between a team’s time of possession and the number of points that it gives up, I used data from every team in the league last year to estimate the following linear regression model:
Estimated Points Allowed/Game = 51.2-55.2* Average % Time of Possession.
For example, the team that most closely fits that model is TB, a team that gave up 24.3 points per game and possessed the ball 49.3% of the time in its games. So, the estimated model predicts that the Bucs would give up 51.2-55.2(.493)=24.1, which is quite close to the 24.3 points per game that the Bucs actually gave up.
Some teams, however, do not fit the model at all. For example, Seattle gave up only 14.4 points per game, the fewest of any team in the league, yet they possessed the ball only 50.8% of the time, which, according to the model results in an estimated average points against of 23.2 points per game. At the other end of the spectrum are the Bears, who had possession of the ball 51.7% of the time, yet gave up 29.9 points per game.
So, what do you think? Does time of possession systematically affect the ability of a team to prevent the opposition form scoring, or is it just a matter of the particular characteristics of the team that matters?
Last edited by a moderator: