Punters are people too

[SteelCityColt]

Greetings fellow posters, I’m back again with another TLDR post. Not so long ago it was announced that Pat McAfee wouldn’t be back as a Colt next season. Understandably I’d say the vast majority of posters were saddened by this as he was one of the more popular Colts for his performances, both on and off the field. However, there were a few voices of dissent saying we wouldn’t miss him one iota next season. So, it got me wondering, how much impact does a punter have and does it actually matter about having a “good” one. The two thoughts are somewhat unrelated in some respects. For example, an average or even below average QB will be in high demand still purely because of the singular effect a QB has on a game. Hardly rocket science I know, but conversely if a punter doesn’t have too much effect on a game it doesn’t matter as much about having a good one if an adequate one will yield similar results. 

 

Therefore, the two main questions I want to explore are:

 

·         Do punters have a significant effect on a game at any point?

·         Was Pat McAfee a good enough punter that he will be a significant loss to the team?

 

Let’s begin by thinking about what the traditional stats recorded for punters are and seeing if they are useful metrics for us to use as yard sticks:

 

Total Punt Yards

Urgh…. As I think I’ve made clear volume stats in my opinionare pretty much useless for comparing players as they lack so much context by their inherent nature. Lots of punting yards could be down to being on a awful team that’s punting lots and punting from deep. It’s the same as people holding up total tackles as a “good” stat for assessing defensive players.

 

Average Yards/Average Net Yards

Ok a little better here, we’ve got numbers that shows us a per play average so could be viable as comparators. But our football knowledge will tell us it’s not the be all and end all as there will be situations where a punter will end punting “short” if they say punt from their own 45 rather than their own 10. It’s worth noting too that Net Yards might be more likely to show the performance of the punt unit as a whole. Overall common sense tells us here that the longer a punter can kick it the better.

 

Inside 20

Now we’re onto something a bit more interesting, this number being tracked came from a somewhat arbitrary origin in that a touchback puts the Opp on their 20 to start so anything inside that and the punter has beaten the baseline so to speak. While punt distance is important it’s important to recognise that leg power isn’t everything. Much like a QB who has a cannon for an arm, if he can’t make the touch throws as well he’s never going to be a top-level player. Again, common sense would also tell us the closer we can pin back the Opp to their own goal line the better the likely outcome for our D.

 

Using the above than as the basis for our expectations, if we were to plot the oppositions starting field position against various outcomes (TDs, FGs, Turnovers etc.) we’d expect to see a linear relationship between these things and the starting field position. If it’s a linear relationship throughout than average punt distance is most important stat for us to look at, and depending on the rate of linear change it will show us how much impact it has on the game. If there’s no linear relationship, then there’s an argument to be made that punting has no significant impact at all.

 

However, what would it mean if we saw a partially linear relationship which radically changed at a given starting position? This would indicate that pinning back the opposition within a certain number of yards to their goal line has a significant influence on the outcome on the ensuing drive.

 

Without further waffle let’s look at some pretty pictures:

 

The graph plots much like the 3^rd scenario described above. We see a nice strong relationship between Opp starting position and metrics such as the % chance they score a TD, % chance to score a FG etc. Put simply the further you punt the ball the better the outcome is likely to be for your D. Surprising huh?

What is surprising (or perhaps not), is the change in that relationship when we start pinning the opposition back within their 10. We see a marked decrease in the chance they will score and an increase in positive outcomes such as them being forced into punting. Interesting we don’t see as marked an uptick in the chance that our D will force a turnover until we start pushing back the opposition into their own 5.

To illustrate slightly more clearly let’s look at the average number points the opposition scores for a given drive starting position:

As points allowed is going to be a product of the TDs/FGs the opposition scores on us it’s no surprise the behaviour is much the same shape as the first graph. But it’s nice to see we can confirm this with a very strong R squared value for drives that began outside the 10 yard line. By extending the trend line to plot the expected values if this behaviour continued within the 10 yard line we can see just how much the real life outcome is different, significantly different.

What does all this mean?

·         The longer on average your punter can kick the ball the better a player he is but there isn’t a huge difference between a punter who say can on average punt the ball 40 yards and one who averages 45. As also noted a punter can end up with a lower average punt distance purely from the field position they are asked to punt from.

·         We see a significant change in the expected outcome of opposition drives when we force them to start within their own 10, therefore this should be a key metric in assessing out punter.

Armed then with this knowledge let’s take a look at how Pat did throughout his Colts career. I’ve broken it down into looking at the following metrics based on the above; Average Punt Distance, Average Net Punt Distance, % of Punts inside the 10, and % of Punts leading a to a Touchback (higher is worse). For each Metric I’ve also added Bryan Anger as a comparison to show what a 3^rd round draft pick can do, as well as the mean number between 2009-2016 for the #1 punter, #10 punter and #16 punter for each criteria to give us baselines for comparison.

Average Yards per Punt

 

 

Not all that surprising to see that he’s been in the top 10 for the last 5 seasons. What is probably more interesting is we see how the gap widens between the top 16 punter and the top 10 punter vs the top 10 and top punter. This would suggest that much like QBs there is an “elite” tier near the top end of punters who perform at a much higher level than their more mediocre peers.

Average Net Yards per Punt

 

 

Not quite as strong a showing for Pat, but we speculated that net yards could be affected by how our punt unit as a whole performs. After all what normally affects net yards is a punt return, so if you’ve got good gunners, you’re more than likely going to limit the net yard loss. We can explore that by examining the difference between the 2 numbers:

 

Note here the Y axis has been inverted as a lower number is better for our purposes. It does not make for pretty reading for the Colts special teams unit, aside from 2014 where they had a very very good year, most other years they were below average at best. It’s what I expected to see based off the “eye test” of watching the games. We always seemed to let returners spring a big one against us!

So far, so good for Pat, but as we established in the opening the more telling stat is what percent of his punts ended up inside the 10:

Pretty darn good, although there was a large drop off last season, which doesn’t seem to fit really with his career best Avrg Yards/Punt. Maybe he was punting longer but was punting on average from worse field position?

 

Nope, as we can see above on average Pat punted when the LOS was on our own 36, which is in line with the previous 4 years where he was really nailing those punts inside the 10. On a side note it’s a nasty reminder of just how bad the team was in 2011, but I digress.

If it’s not the average field position, and his average punt distance has actually improved then what’s the deal? Here we see the limitations of examining stats. They can only tell so much of the story, I can think of at least 3 occasions where a punt should have been downed inside the 5 but was spilled over the line for a touchback. When you’ve only punted 55 times in a season those 3 punts can have a marked drag down on the numbers, if they’d been executed correctly we’d probably have seen Pat well north of the 10% mark.

Speaking of touchbacks, in my opinion a touchback is a negative outcome for a punt. Granted it’s not always down to the punter, but we don’t split hairs when we look back at QB’s INT numbers either. For the sake of completeness:

 

Again I’ve reversed the Y axis as a lower number is better and we can see that he’s not as consistently good as with the other metrics. However, we can see the sharp rise in Touchbacks this season which supports somewhat my theory that it was inept play by other members of the punt unit turning possibly very good outcomes into bad ones.

 

We finally arrive at the summary, maybe I should add a TLDR header for this part so people can skip straight to it! Based on the above then it’s my opinion that:

 

• ·         There isn’t much difference between punters outside of the “top 10” and the lower level punters. The actual drop off even down to the worse punter in a metric isn’t as astronomical as you’d think. It would be reasonable to say at that level of play punters are a much of a muchness.
• ·         Inside the top 10 and above  however we see punters start to have a more marked effect on games and a distinct gap opening up in the metrics between those in it and those outside of it.
• ·         Long punts are good, but field position is king.
• ·         Poor special teams play can really ruin a punter’s day

 

All the above is pretty obvious and not exactly surprising, though it is nice to see the data support it. But for those shaking their heads and wondering why I’ve gone to these lengths to tell you something you knew already, I’d counter with this. Pat McAfee was a top 10 player at his position by pretty much any metric you choose to measure by, yet we’re questioning if he’s going to be missed by the team? Losing any player of the calibre hurts, regardless of position. That’s before we’ve even considered the intangibles (such a lovely oxymoron), his locker room presence, his chemistry with AV and Overton on FGs, his passing ability (!). These should also not be overlooked.

 

To Pat then. Thank you for your years as a Colt, you were a pleasure to watch on the field and a fantastic contributor off the field. In your own badly paraphrased words… “All-Pro Punter please and thank you”.

 

 

All data for the above was sourced again from the excellent Pro Football Reference, as well as the equally useful Sporting Charts website. As ever I add the rider, stats are not the be all and end all, merely information. But with a greater amount of information we can be greater informed. 

 

Who knows, one day I might learn about football so I can actually look at film

[Yoshinator]

The main thing that will suck about losing Pat is that he can't bail the offense out when they go 3 and out in the 1st half so much on the slow starts, the opponent's offense will get better field position, and our defense will have less room for error to stop them. The defense is already bad to start with, and now they will probably have to stop the opponent 10-15 yards closer than avg this year. This is not going to be good. Pat will be sorely missed.

[GoColtsWin]

I agree Patty Mac will sorely be missed. But with that being said, the Punter position seems to be becoming alittle more competitive and in the spotlight more often. Hopefully this translates into a replacement that shares the same passion for the game, coupled w/our already great special team corp whom should be able to help out our new punters progression. Absolutely appreciate the time it took to put out your very well thought out post OP. 

[csmopar]

9 hours ago, SteelCityColt said:

Edited for neatness

 

To Pat then. Thank you for your years as a Colt, you were a pleasure to watch on the field and a fantastic contributor off the field. In your own badly paraphrased words… “All-Pro Punter please and thank you”.

 

 

All data for the above was sourced again from the excellent Pro Football Reference, as well as the equally useful Sporting Charts website. As ever I add the rider, stats are not the be all and end all, merely information. But with a greater amount of information we can be greater informed. 

 

Who knows, one day I might learn about football so I can actually look at film

In other words, we are screwed without Pat

[braveheartcolt]

If we had Football Forum Oscars, this would win ''Best Post'...........sorry, really sorry, Superman is 21, you win, Superman is 21 wins best post.........

[SteelCityColt]

2 hours ago, braveheartcolt said:

If we had Football Forum Oscars, this would win ''Best Post'...........sorry, really sorry, Superman is 21, you win, Superman is 21 wins best post.........

 

Obviously I'm upset but I won't make a song and dance about it. 

[ReMeDy]

::Me before reading the topic::

"Nah, there are plenty of punters. We'll be fine."

::Me after seeing the graphs::

"We're screwed."

[SteelCityColt]

16 minutes ago, ReMeDy said:

::Me before reading the topic::

"Nah, there are plenty of punters. We'll be fine."

::Me after seeing the graphs::

"We're screwed."

 

Mm I wouldn't say screwed, screwed is Andrew Luck retiring to become a children's author. 

 

It is still going to be a loss however, I don't think people realise how much Pat helped mask a lot of the holes on our D. He was literally the first line of defense. That's a really worrying statement.

[teganslaw]

7 hours ago, SteelCityColt said:

 

Mm I wouldn't say screwed, screwed is Andrew Luck retiring to become a children's author. 

 

It is still going to be a loss however, I don't think people realise how much Pat helped mask a lot of the holes on our D. He was literally the first line of defense. That's a really worrying statement.

Luck retiring as a children's author? Please don't give him any ideas!

[jvan1973]

On 2/27/2017 at 6:42 PM, Jared Cisneros said:

The main thing that will suck about losing Pat is that he can't bail the offense out when they go 3 and out in the 1st half so much on the slow starts, the opponent's offense will get better field position, and our defense will have less room for error to stop them. The defense is already bad to start with, and now they will probably have to stop the opponent 10-15 yards closer than avg this year. This is not going to be good. Pat will be sorely missed.

The best net avg punter and the worst net average punter is 7 yards

[Yoshinator]

Just now, jvan1973 said:

The best net avg punter and the worst net average punter is 7 yards

Thanks for the info. Based on who we had, and who we could have, 7 yds could still make a difference though.

[2006Coltsbestever]

18 hours ago, braveheartcolt said:

If we had Football Forum Oscars, this would win ''Best Post'...........sorry, really sorry, Superman is 21, you win, Superman is 21 wins best post.........

Actually you win with that Post, you looked at the wrong envelope.

[SteelCityColt]

8 hours ago, jvan1973 said:

The best net avg punter and the worst net average punter is 7 yards

 

Granted, but as we explored in the OP yards isn't the be all and end all and the % of punts within the 10 drops off a lot more the lower down we go down the rankings.