Yes, you read that title right. Not only is today about punters, guess what? Tomorrow will be, too. Today, I want to dive into punting statistics. The two key numbers the media focuses on with punters are usually net punting average and gross punting average. But both numbers are pretty heavily influenced by field position.

Both gross and net numbers are higher if the punter is between his own 20 and 40 yard line. But as a punter moves closer to his opponent’s end zone, his gross average goes down. On the other hand, the gap between gross and net punting goes down significantly as a punter gets closer to the opponent’s end zone.

Take a look at the graph below, which is generated from all punts made during the 2014 regular season.  The blue line displays gross punting averages; the dotted blue line is a best-fit line.  The red line displays net punting averages with the dotted red line being the smoothed line.  The X-axis represents the line of scrimmage for the punt, so 20 = the punter’s own 20, 40 = the punter’s own 40, and 60 = the opponent’s 40 yard line. ^[1]For those curious, I also included the formula for the best fit lines and the R^2 data.

But looking at gross and net punt data doesn’t get us all the way to understanding how well a punter is at his job. That’s because the closer you get to the opponent’s end zone, you care more about where the other team gets the ball than how long the punt is.  Whether you punt at the 50 or the 45, you really just care about pinning the other team inside their own 10.

The graph below shows the percentage of punts in 2014 ^[2]Including whatever happens on the return. that went for touchbacks in blue, that landed inside the 10 in red, or inside the 19 in green.

So while it makes sense to use net (or gross) punting data when a punter is far from the opponent’s end zone, those numbers are much less important than where the opponent will start the next drive the closer move the line of scrimmage to the opponent’s end zone.

But what about the reverse: perhaps we should always keep our focus on where the opponent’s next drive starts?  Let’s get to our next graph.  As before, the red line displays net punting averages, while the orange line displays where the play ended.  So if a team punts at its own 45, it should expect the opposing team’s next drive to begin at the 85 — i.e., that team’s own 15.  This average includes all touchbacks, which obviously give the opponent the ball at the 20.

That’s pretty close to a linear relationship, although at the tails that description be off.  ^[3]A linear line would overstate what should be expected from a punt inside the 15 (it’s going to likely remain on that side of the field) and what should be expected when a team punts on the … Continue reading  But a polynomial trend line works nicely.

And here’s the data in table form. I don’t want to clog up the page with 70 rows of numbers, but note that the table below does contain the expected yard line where the opposing team will take over for punts at all yards lines between the 1-yard line and the opponent’s 30 yard line.

Using this data, we can figure out where the opponent will take over following the average punt. Which means we can tell, over a large enough sample size, if a punter is doing better or worse than average. That’s what we’ll get to tomorrow.