Punt Returns: An Analysis

[tims4wins]

Here is what I don't understand. BB is way ahead of the curve in terms of understanding risks in football. 4th down, etc. It seems to me that punts have like 4 outcomes. Most common is probably fair catch. Next common is probably a short return. Then there are long returns and turnovers. Turnovers might happen more often than long returns, not sure. And of course holding and block in the back calls can happen on fair catches, and on any types of returns.

What I am trying to say is that, overall, there doesn't seem to be a benefit in returning punts, as opposed to just fair catching them. It seems like a net negative to the returning team.

Am I way off base?

Edit: I understand taking risks when you have a middling offense, just doesn't seem worth it when you have the GOAT at QB

 

[tims4wins]

Another point on returns: I get putting Jones out there in situations where the other team is somewhat backed up. Say punting from their own 20-25 or inside of it. Situations where you have a chance to bust a return for a touchdown. But when there is a good chance the punt is going to have to be fielded inside your own 20, that is when I want the sure handed guy out there returning. There is a time and a place for Jones, fielding punts inside his own 20 isn't one of them.

I am hoping the Pats can wrap up the 1 seed prior to week 17, so he can get out there and maybe gain some confidence back.

 

[gammoseditor]

Here is what I don't understand. BB is way ahead of the curve in terms of understanding risks in football. 4th down, etc. It seems to me that punts have like 4 outcomes. Most common is probably fair catch. Next common is probably a short return. Then there are long returns and turnovers. Turnovers might happen more often than long returns, not sure. And of course holding and block in the back calls can happen on fair catches, and on any types of returns.

What I am trying to say is that, overall, there doesn't seem to be a benefit in returning punts, as opposed to just fair catching them. It seems like a net negative to the returning team.

Am I way off base?

Edit: I understand taking risks when you have a middling offense, just doesn't seem worth it when you have the GOAT at QB

The analysis should compare where you are likely to have the ball returning it vs where you are likely to have the ball fair catching it. This should factor in the possibility of a turnover and account for the strength of the teams punt coverage. Of course, a fair catch does not completely remove the possibility of a muff and a turnover. I don't think the strength of our offense has a lot to do with the decision making. Belichick is not a fan of plays that are not competitive. He's the one that proposed pushing back the extra point. He's not going to forgo the chance of improving the teams field position if on average returning the punt has a positive result.

 

[tims4wins]

Right, if the metrics show that on average you start with better field position than if you just fair caught every return, then I agree he wouldn't employ the strategy. But I think the #1 goal on every special teams play where you are receiving the ball is to possess the ball. Ball security is clearly trickier on special teams than regular plays. Fumble rates are far higher (note: this is an assumption - I could be off base).

 

[gammoseditor]

Right, if the metrics show that on average you start with better field position than if you just fair caught every return, then I agree he wouldn't employ the strategy. But I think the #1 goal on every special teams play where you are receiving the ball is to possess the ball. Ball security is clearly trickier on special teams than regular plays. Fumble rates are far higher (note: this is an assumption - I could be off base).

I agree with where you're coming from. I just don't think it's how Belichick thinks. The league is so competitive that you need to look for every advantage you can. If you're fumbling punts the first, second, and third solution is to fix the problem. Giving up on the play by fair catching every single one is a last resort and in a way admitting failure.

 

[Stitch01]

I don't think its generally correct that returning punts is bad, but it is analyzable. Not sure if anyone has done the analysis.

 

[tims4wins]

I agree with where you're coming from. I just don't think it's how Belichick thinks. The league is so competitive that you need to look for every advantage you can. If you're fumbling punts the first, second, and third solution is to fix the problem. Giving up on the play by fair catching every single one is a last resort and in a way admitting failure.

Agreed with everything you are saying. I would just be very curious to see the average gain in field position in comparison to simply fair catching, when you take into account the penalties. Not sure how you would factor in turnovers though.

Edit: and obviously turnovers could happen on fair catches. But I would argue that if you are coaching fair catching punts, that would eliminate penalties like block in the back / holding. And it would probably also cut down on muffs.

 

[IdiotKicker]

I don't think its generally correct that returning punts is bad, but it is analyzable. Not sure if anyone has done the analysis.

By the end of this season I'll have full charting done for the last two seasons and I'm hoping to have a big enough sample size to make some preliminary conclusions. But we're tracking for everything including field position, hash, punt direction, hang time, and the whole lot. So we've got the data, I just don't know if it really shows anything yet.

 

[Stitch01]

By the end of this season I'll have full charting done for the last two seasons and I'm hoping to have a big enough sample size to make some preliminary conclusions. But we're tracking for everything including field position, hash, punt direction, hang time, and the whole lot. So we've got the data, I just don't know if it really shows anything yet.

Interesting. How are you dealing with turnovers and penalties? That seems like the biggest complicating factor.

I don't support cutting him for now. But he should not have been used on punt returns in the second half.
Did you see the punt just before the one he touched and fumbled?
He got dangerously close to that one.
Why run him out there after he almost made a mistake... so he can really make a mistake?
Bill and the staff should know better.

Im very glad messageboard posters don't get to coach the team.

 

[IdiotKicker]

Interesting. How are you dealing with turnovers and penalties? That seems like the biggest complicating factor.

Not sure yet other than noting the plays in which they happen. I'll probably run a few different sets but my guess is stripping out plays with penalties and keeping those with turnovers will be the route I end up going for my initial purpose, which is determining the value of different punt attributes such as hang or direction. But for something looking at the relative value of returns, it's going to be a different process.

 

[Stitch01]

Yeah, I think for returns you'd have to do something like a game state concept for turnovers and penalties.

 

[IdiotKicker]

Yeah it's way more complex on that side which is why it's probably the second thing I'll tackle. But we have data from like 5000 punts that we are open-sourcing if anyone wants to dig in on their own as well.

 

[tims4wins]

Yeah it's way more complex on that side which is why it's probably the second thing I'll tackle. But we have data from like 5000 punts that we are open-sourcing if anyone wants to dig in on their own as well.

I'm bored at work and would love to play around

 

[Jnai]

Yeah it's way more complex on that side which is why it's probably the second thing I'll tackle. But we have data from like 5000 punts that we are open-sourcing if anyone wants to dig in on their own as well.

Any chance you have field goals / xpat data? Distance, make / miss, defense, score, quarter?

 

[IdiotKicker]

So I believe this is only updated for this season through Week 11 or 12 since I still haven't done a dump for the last month, but this is every special teams snap for the entire season if anyone wants to go through it.

http://www.insidethepylon.com/2016-special-teams-play-charting/

Edit: Actually it's through Week 9 since I'm getting married in two weeks and have fallen behind. But I'm wrapping up Week 14 charting today and should hopefully have it all up tomorrow then.

 

[IdiotKicker]

Any chance you have field goals / xpat data? Distance, make / miss, defense, score, quarter?

Score and time I have not been tracking myself, though in hindsight that might have been a mistake. I can go back through PFR and do a pull and then merge things in during the offseason though.

 

[IdiotKicker]

Ok, just loaded in all the data through Week 14 except for last night's game, so should be good to go.

 

[tims4wins]

My browser is having a tough time with the script

edit: got the data through week 9. What is TDP?

 

[tims4wins]

Went through the punt data from the first 9 weeks. Overall there were 1,201 punts. Some quick and dirty stats:
- The average return for all punts - including fair catch and touchback - is 4.42 yards (standard deviation of 9.8)
- The average return for punts that are returned (787 punts) is 6.73 yards (standard deviation of 11.5)

Can't tell too much without the turnover and penalty data.

 

[IdiotKicker]

Yeah it's not just your browser, there are like 5000 special teams plays in there with between 8-12 entries for each so it's not the quickest load. TDP is my stat to correct for field position in punting by weighting relative to the target distance, so punters who punt from the 50 all the time aren't penalized. Pretty much it takes the NFL average on punts within a team's own 40-yard line as the target and then the opposing 10-yard line as the target for all punts from the 41-yard line onward. Formula can be found here.