## Field Goal Kickers

I recently took a look at special teams (ST) and its importance in winning. One of the most important, if not the most important, players on ST is the field goal kicker. No one else has such a direct and solitary impact on points scored. In this post, I'll look a little closer at the impact a kicker can make on the win-loss record of his team.

The relative importance of each dimension of the game, including ST, is estimated in a regression on regular season wins. Each variable is in terms of team efficiency (yds per attempt) and is standardized. The ST variables are relative to the league average for similar situations. For example, the FG/XP scores are relative to kick distance and the average success rate for each distance (1).

The results are summarized in the table below.

 Variable Std.Coeff. P-Value O PASS 1.26 0.00 D PASS -0.81 0.00 O RUN 0.47 0.00 D RUN -0.47 0.00 O INT -0.63 0.00 D INT 0.60 0.00 O FUM -0.33 0.07 D FFUM 0.36 0.02 PEN -0.41 0.01 FG/XP 0.34 0.00 KICK 0.44 0.00 K RET 0.07 0.64 PUNT 0.23 0.12 P RET 0.27 0.10 r-squared 0.79

The results above can be interpreted as follows. Each coefficient indicates how many additional regular season wins a team can expect, on average, per standard deviation above average. For example, if a team is completely average in every facet of the game, it can expect to win 8 games. But if a team is average in every facet except offensive pass efficiency, in which it is 1 standard deviation above average, it can expect to win 8 + 1.26 = 9.26 wins.

The best kicker in the league (#1 out of 32) would typically be in the top 96th percentile, which is very close to two standard deviations above average. Therefore, the best FG kicker in the league would normally be worth 2 * 0.34 = 0.68 added wins in a season.

It's hard to imagine many other positions, other than the starting QB and RB, that have such a large individual impact on a team's record.

### 5 Responses to “Field Goal Kickers”

1. Anonymous says:

Very interesting. A lower p-value means greater significance, right?

Two thoughts. One, it makes sense that excellent field goal kicking will win more games. But I wonder how well you can predict field goal ability. That is, the top field goal kicker each season might win .7 extra wins, on average, but his true talent might only be .2 wins above average -- it's just that it's likely SOMEBODY will have an excellent season.

Second, Football Outsiders has preached the importance of deep kickoffs as a significant portion of a kicker's value. Your KICK value backs that up (although some of that is obviously kick coverage.)

Good stuff.

2. Brian Burke says:

Thanks.

Yes, p-value is the probability that the effect of the variable is actually zero, but due to sample error it appears to have an effect by chance alone.

I agree with your point on the difficulty of predicting future FG performance on past performance. The sample sizes are so small, but it's the best we have.

I used FO data for kicking, so unfortunately this isn't independent confirmation. But to me, the strong importance of kickoff performance makes a lot of intuitive sense.

Consider it this way: The average series success rate (getting a 1st down) in the NFL is 65%. So for every additional 10 yards of kickoff depth achieved, the opposing team has about a 35% less chance of scoring on the upcoming drive, because that's one additional first down they'll need.

10 yards is a lot, but some teams are able to be 10 yds better than others in kickoffs.

The effect is not linear, however. A 5 yd difference in kickoff depth means a (0.35)^2=0.12 difference in the probability the opposing team will score on the next drive.

Suddenly it makes a lot of sense for teams to carry a kickoff specialist and invest in coverage specialists!

I plan to do more work on series success rate soon.

3. Brian Burke says:

Sorry, that should be 1-(0.65^0.5)=0.19 for a 5-yd difference in kickoff depth. Not a 0.12 difference like I said in the previous comment. Probabilities are so hard to wrap your head around.

4. Tarr says:

It only makes sense to have a seperate "kickoff specialist" and "field goal specialist" if the field goal specialist is consistently better at field goals than the kickoff guy. If, however, field goal accuracy (like recovering a fumble) is pretty much the same for everyone who is capable of place kicking in the NFL, then you may as well just have one kicker, and pick him based on how good he is at booming the kickoffs.

And again, I see absolutely no conflict between this research (which says that accurate field goal kicking has a large impact on winning games) and the FO research (which says that NFL place kickers who have been above/below average accuracy given the distance and environment don't tend to remain so in the future). The two sets of research are addressing different things. It's entirely possible for something to be both very important, AND "random" (non-predictive) at the same time.

One issue with kick coverage is that teams are worried about using their best players in coverage because they might get hurt. For the same reason, some teams avoid practicing coverage at full speed during the season. This research may suggest that it is worth the risk, though. I think the Pats are known for using more starters on special teams.

5. Brian Burke says:

Tarr-I agree almost completely. My only difference with your take is the assumption that all FG kickers are equally accurate. My intuition is that not all FG kickers are the same in accuracy. I've watched Matt Stover for 11 years out kick anyone in the league. The guy is money.

I'm sure part of the accuracy rates is affected by luck and small sample sizes, but at the core there is a baseline accuracy skill level for each kicker.