Kickoffs and the NFL's New Rule, Part I: Touchbacks

A great many people have had their knickers in a twist this season over the new kickoff rule which places the ball at the 35 yard line. The putative reason for the rule is player safety.

This article by the fine folks at Advanced NFL Stats demonstrates that the difference between the 2010 injury rate on kick returns (2%) is higher, but not vastly higher, than injuries on other sorts of plays.

The lowest injury rate, strangely, was 1.3% on punts. The highest injury rate on plays other than kick returns was running plays, at 1.6%. (This doesn't take into account field goal attempts, where the injury rate was something like .25%. It wouldn't surprise me if there were almost that many injuries when the team runs out of the tunnel.)

The article concluded that the improvement in player safety is extrapolated to be about .2 injuries per game. It's hard to imagine that the NFL is willing to give up the excitement of that play to avert an average of three injuries per team per season. After all, an injury can be anything from a mild sprain to a broken limb.  So the question is, what sort of injury? I suspect the hypothesis is that special teams injuries are more likely to be head injuries. That assumption, combined with the new emphasis on preventing concussions, is what made them decide on the rule change.

This study indicates that the position group most likely to sustain a concussion that keeps a player off the field for more than seven days is the defensive secondary, followed by the kick team. You can't do anything about the DBs, I suppose, but the reasoning was presumably that you could reduce the incidence in the kick teams. However, given that only a fraction of the injuries were concussion-related, that still doesn't get us very far in understanding the rationale of the NFL higher-ups.

I don't know what data the NFL has, but this study indicates that for high school athletes about 20% of kick team injuries are head injuries, as compared to about 10% for all other types of plays. If we assume that this is a similar rate to the NFL figures, then about 20% of the special teams injuries will be concussions. That translates to less than one half of one percent of all special team plays resulting in a concussion injury.

The new rule isn't going to completely eliminate special teams injuries, no matter where they place the ball, unless they eliminate kickoffs altogether and just start everyone at the 20. I noted that on one kickoff last year there was a player injured, even though the kick was a touchback. And as we can already see this season, a substantial percentage of kickoffs are still going to be returned.

So in fact the injury rate is not going to go from 2% to 0%. It will be quite interesting to see where it does end up. It wouldn't greatly surprise me if it actually stays about the same, because of the effect of returners being willing to risk bringing the ball out from deeper in the endzone than before—but more on that in a moment.

I'm certainly not trying to minimize the effects of concussions on player's lives, or suggest that the league should not do everything in their power to decrease the incidence. But I suspect that there are other, better methods at hand that they aren't mandating—better helmets, for example—for not entirely noble reasons.

But to return to touchbacks, it isn't at all clear at this point that the new rule is going to vastly increase the number of touchbacks when the season is considered as a whole, for a variety of reasons. The first, mentioned in the Advanced NFL Stats article, is the effect of weather on kick distance. As it gets colder the ball isn't going to carry as far. Football Outsiders calculated that there is generally a decrease of between three to five yards in kick lengths during the course of a season, and there is a more pronounced effect in cold-weather games. (Cold temperatures make the ball denser and harder, reducing both the distance it travels and the height it reaches.)

Furthermore, it seems unlikely that coaches are going to order their kickers to kick touchbacks whenever possible. Unless a team has major problems with their special teams coverage, I would suspect that we will see an increasing number of kicks intended to land short of the goal line. There is a big advantage to trapping a team behind their 20 yard line. Even drives that begin at the 20-yard line have only a 15% likelihood of ending in a touchdown.

Sebastian Janikowski of the Raiders, who had the second-highest number of touchbacks last season, notes that not every kick will be intended to be a touchback, even on teams with a kicker capable of doing so. It will be surprising if coaches don't come up with strategies to mitigate the effects of the new rule.

One of the unexpected results so far this season is that returners are taking chances on bringing the ball out from far deeper in the end zone than they typically did in previous years. Antonio Brown commented that you can take teams by surprise when they expect you to take a knee. That won't last long, of course. And kicks that end up in the stands (like one of Robbie Gould's kicks in the Bears' preseason game in Nashville) aren't going to be returned. Unless the NFL decides to let fans have a go if they catch the ball in the stands. 

The touchback rate for all kicks last season was 16.4%, and that decreases to about 16% if you remove kicks that didn't begin from the 30-yard line. 

The touchback numbers varied greatly by team, as you might expect. Billy Cundiff of the Ravens had an other-worldly touchback percentage of over 50%.*  During the Week 17 game vs. the Bengals Cundiff tied the NFL record of 40 touchbacks for a season, in 70 kickoffs. He had two more in the first playoff game vs. Kansas City.

The next closest kicker wasn't very close. Sebastian Janikowski of the Raiders kicked 29 touchbacks last season in 88 kickoffs, around 31%. Tampa Bay's kicker, Connor Barth, had the lowest percentage in the league, at 1.28%.

Pittsburgh was well below the average at 8.74%—22 teams had better touchback percentages. Kickoffs were never Jeff Reed's strong suit, and Shaun Suisham joined the team after the weather turned cold, thus reducing the likelihood of touchbacks. It will be interesting to see where Pittsburgh ranks this season.

An ironic footnote as we look at these issues is that the NFL introduced "K" balls in 1999 to make the game more exciting. They do so by making it more likely that the ball will be returned, as this 1999 article sets forth: 

The idea behind the K ball was to keep kickers from "cooking" the balls -- placing them in microwaves, squeezing them in vices anything to make them softer and rounder. The NFL wants more points scored, and one way to get them is to force more kickoff returns.

Obviously, it's not having an effect yet on field-goal kicking, which usually is no farther than 55 yards. But kickoffs, which require 70 yards to reach the goal line, does show a K-ball effect.

In 1997, 17.2 percent of all kickoffs were unreturned (touchbacks or kicked out of bounds). The figure for 1998 was 19.7 percent. This season, it's down to 13.1 percent. Of course, the kickers with the strongest legs are holding their own. Mare, Hollis and Richey lead the NFL with 13 touchbacks. That's 32.5 percent efficiency for Mare, 38.2 percent for Hollis and 36.1 percent for Richey.

To be continued, at my usual length, no doubt. I noted down the result of every single kickoff from last season, and will be throwing out some of the things I found during that marathon. Some of the questions to be explored are what the percentages are on returns from within the endzone, the effect of a kick return for touchdown on the opposition, and other matters of interest, to me at least. So don't touch that dial!

*I have yet to see a good explanation for Cundiff's amazing showing last season. As a Football Outsiders article noted:

Going into 2010, Billy Cundiff was a run-of-the-mill kicker who had bounced around the NFL. Between game rosters and preseason, he had suited up for nine teams in nine seasons. Cundiff had just 11 touchbacks in his career, on over 200 kickoffs.

So it was a shock, to say the least, when Billy Cundiff had 40 touchbacks on 79 kickoffs in 2010.

The article went on to note that there was no obvious reason why he should have made such a massive improvement. Cundiff himself claimed that he had made some changes to his technique, and felt more confident. But the biggest previous improvement that a player attributed to a technique change was Michael Husted, who increased his touchback rate by 22%. Cundiff went from a career average of about 5% to 50%. 

Football Outsiders came to the conclusion it was a fluke. They might be right. The results after Week 2 have Cundiff at a 70% touchback rate. Pittsburgh is at 57.14%, for the No. 10 spot. Cundiff is actually only No. 4 in the league, and Denver and Oakland are tied for No. 1, at 100%. 

Steeler Nation naturally seeks a more sinister explanation than "a fluke."  Hopefully it will be positively Faustian. If any of you have one, I would love to hear it.