Football coach may refuse to punt

by on August 14, 2012 at 4:20 am in Sports | Permalink

San Diego State head coach Rocky Long told Tod Leonard of the San Diego Union-Tribune that he’s considering not punting or kicking on fourth downs in 2012. Instead, Long is considering going for it on fourth downs inside an opponent’s 50-yard line in order to try and pick up a new set of downs every time.

Kevin Kelley, the head coach of Pulaski Academy in Little Rock, Arkansas has developed the strategy over the years. He claims punting is an offensive failure, and is essentially a voluntary turnover.

Here is more, hat tip goes to Mark Buckley, and for previous MR posts on this topic see here and here.

ibaien August 14, 2012 at 4:42 am

statistically smart yet counter-intuitive strategies like this can only succeed if the coach either a) has the full faith and support of his employers (and a multi-year contract) or b) is an absolute nobody coaching an absolute nothing team (so who really cares if many of the conversion attempts fail?). any other set of circumstances will promptly lead to tears, recriminations, and firings.

Andrew' August 14, 2012 at 5:35 am

It’s similar to my ponderings on “going for 2″ at all times. Doing it always you lose the game theory aspects, but you also get better at doing it. You only have to convert 50% of the time to approximate the results from kicking the extra point. If you did it more you might get good enough to convert at that rate, and that practice would also make you better at your short-yardage situations.

With the going for it on fourth down, it might change the play calling on the prior 3 downs. Would it allow you to be more aggressive or more conservative? Clearly you wouldn’t go for it every time in your own end of the field, and at other times a field goal is more appropriate. At others a Hail Mary can serve as a poor-man’s punt. In the end, you probably end up going for it a little more than the average coach. Not to mention the “public choice” aspects ibaien points out that coaches would have to surmount, and I’d add that you get to be a head coach by paying your respects to the traditions of the game for 10-30 years beforehand.

Artimus August 14, 2012 at 6:04 am

Well, except for Bill Belichick.

rob August 14, 2012 at 10:40 am

Not that it’s necessarily comparable to the real game, but I adopted the strategy in the PS’s NCAA football of never punting and always going for 2, and found it to be effective.

Artimus August 14, 2012 at 6:00 am

Good post Andrew, but one point I would like to make is that while the offense may get better at two point conversions so will the defense at defending them.
Just my opinion and a guess but I don’t see the two point conversion success rate increasing from its present rate.
It would be interesting to watch and find out though. On a related note, I notice
not many economists are football coaches.

Andrew' August 14, 2012 at 6:06 am

Good point. The main thing might be it that it reduces the offence’s pucker factor on those plays because they would become routine. Also, you might be like Georgia Tech and others in that people don’t have the time to prepare for your different play style one game a year whereas you practice every game. At any rate, it would make the game more interesting for the fans!

Artimus August 14, 2012 at 8:34 am

Actually I didn’t think about that. If a team (or 2 or 3) adopted that style of play in the NFL, all the teams would have seen and played against that strategy relatively quick.
However in the college system with more teams, there would be many teams that have not seen that style of play and are less prepared. Kind of like the wishbone offense.

The Original D August 14, 2012 at 11:29 am

Even if defenses adapt, the context could make a difference. I bet it’s harder psychologically to stop a 4th and 2 (essentially a two-point conversion) immediately after they just put 6 points on the board.

KLO August 14, 2012 at 11:25 am

I don’t really understand why offenses that try more two point conversion attempts and defenses that face more two point conversion attempts will get better at them. Two point conversion attempts are little different from other third down short yardage plays, particularly those around the goal line. Teams already practice these situations extensively and have developed lots of plays and schemes in the playbook to deal with them. It is hard for me to understand how a couple extra attempts per game would greatly affect the conversion rate. Believing that it would is akin to believing that kickers who get lots of chances to kick field goals (likely because their teams have garbage red zone offenses) are more likely to convert than kickers who get fewer chances. Kickers, like offenses and defenses, practice a lot, and so their performance in common game situations has almost nothing to do with their prior experience with those game situations.

Rohrer August 14, 2012 at 6:25 am

================

” Being in politics is like being a football coach. You have to be smart enough to understand the game, and dumb enough to think it’s important. ”

— U.S. Senator Eugene McCarthy

Anon. August 14, 2012 at 7:03 am

I know some of these words.

Anyone mind explaining what that stuff means?

bluto August 14, 2012 at 9:15 am

American football is a game where the basic unit of measurement is 10 yards. The ball must be advanced by at least 10 yards every 4 plays. At the offenses option (though in almost every case this happens on the 4th play or down) they may punt or kick the ball, turning it over to the other team. The punt usually nets about 40 yards, and it’s been standard operating procedure in football at all levels to punt on 4th down.

However, using statistical analysiss, the cutting edge is beginning to believe maintaining control of the ball is exceedingly valuable far more than was previously believed by conventional wisdom. The value is high enough that it’s worth a fairly high risk of losing the 40-70 yards that kicks return (this applies to punting and also attempting more risky kick-offs that offer a chance to keep the ball after scoring). Thus far the only team that’s applied this discovery has been a single high school in Arkansas (there may be others now but they’re the most famous), this announcement is the first time (I’ve heard of anyway) a college team attempting the expected optimal strategy.

It comes up in economics because at least one of the people who popularized the analysis was an economist.

Jan August 14, 2012 at 7:11 am

It is true that coaches opt to punt way too often. From a purely statistical point of view, I think I have read that teams should actually be going for it on fourth down something like 40% of the time. Being more aggressive on 4th is what teams should be doing, but I think there are problems with making a rigid rule like this. I wouldn’t think it makes sense to go for it on 4th and 9 at your opponent’s 48 yard line in very many situations.

ThomasH August 14, 2012 at 7:15 am

But why would a coach announce his strategy? Surely it would be more effective if the other team does not know that he will always try for the first down? Or is the strategic announcement itself part of his meta strategy?

Sol August 15, 2012 at 9:28 am

Presuming the coach is not playing on faking a punt / field goal, then there’s no real disadvantage to announcing the strategy, is there? They’d know what he was attempting to do probably before the team reached the line of scrimmage.

Anonymous Coward August 14, 2012 at 7:25 am

For reference, here’s Brian Burke’s fourth-down graph (assumes that everyone is NFL average, and that maximizing points is the goal, i.e. there are no end-of-half considerations):

http://farm3.static.flickr.com/2618/3688516023_07450826e5_o.png

Sol August 15, 2012 at 9:19 am

Why did they draw the beginning of the graph that way? They make it look like it makes sense to attempt a field goal if you’ve got (say) three yards to go on the two yard line, when of course you *cannot* have three yards to go on the two yard line… Seems like it would be cleaner to just say “never attempt a field goal if you’re within six yards of the end zone.”

Asdf August 14, 2012 at 7:46 am

As pass friendly nfl rule changes continue to juice yards per attempt, eventually all good offensive teams will find things like punting are useless.

Tarrou August 14, 2012 at 8:58 am

It would seem the math might be long, but relatively simple. If average # of opponent’s first downs X average yards per play 2.5, then going for it on Fourth Down every time is completely justified, and I’ve never understood why more teams don’t do it.

Nick August 14, 2012 at 5:04 pm

Because frequency of success, field position and game score matter. So a high enough expected yards per play number is not sufficient.

Todd August 14, 2012 at 9:22 am

Much like the resurgence of the Quarterback run-option offense (“The Wildcat”), I think that foregoing punting on 4th down at a higher rate could be effective in the short term. Over the course of the season in which it was utilized, defenses would be forced to spend time preparing to play four downs of defense even in games where the strategy was not employed. This could produce advantages above and beyond the higher rate of success on fourth down plays.

But eventually defenses would adjust (maybe by being marginally more aggressive with blitzes on earlier downs), and the overall utility of the scheme would start to approach the level of punting on fourth down.

NaG August 14, 2012 at 9:24 am

Always going for two makes sense if your chances of converting are 50% or greater. Generally, that means that you will tend to go for two against weaker defenses, and kick extra points against strong defenses.

I’m a big fan of not punting. I would not punt (barring needing 10 yards or more) between the 50 and the opponent’s 30, and would not punt between my 30 and the 50 unless we needed 5 yards or more. But as noted elsewhere, coaches matter an awful lot in the NFL, and coaches get a lot of heat when they make those calls and fail. Only well-established coaches could get away with it, and they are good enough to know how to win without having to resort to those tactics.

The Other Jim August 14, 2012 at 9:36 am

Sadly, the media has far more influence than it should. (Which answers the question of why this guy is pre-announcing his strategy — he wants media attention.)

If a Coach has a unique strategy that will allow him to win a few extra games per year, but will also lose him one extra game per year in horrifying fashion, this is a net gain. He should clearly do it. But he knows the media will absolutely shred him for that one loss, and maybe put him in an All Time Biggest Blunders compilation that will run on ESPN til 2040. So he won’t.

Roger Sweeny August 14, 2012 at 10:00 am

Gregg Easterbrook, ESPN’s “Tuesday Morning Quarterback”, has been pushing this for years.

Skip Intro August 14, 2012 at 10:35 am

That is a good prima facie reason not to do it.

derek August 14, 2012 at 10:51 am

I’m surprised no one has mentioned why the punt in the first place. It isn’t the odds of getting first down, it is the odds that you are giving your opponent. Even if you convert half the time, the other half you give the opponent 30 plus yards, increasing the chance that you get scored against.

Yancey Ward August 14, 2012 at 10:59 am

Bingo!!

Go Kings, Go! August 14, 2012 at 12:45 pm

Not all yards on the football field are the same. If you punt from your opponents 45 yard line and he recovers at his 20, those are the easiest 25 yards to pick up on the field, since your secondary is typically in deep coverage. Now if you go for it on 4/10 from your own 5, that’s nutty.

Andrew' August 14, 2012 at 12:59 pm

This is why I say “announcing you are always going for it on 4th down isn’t about always going for it on 4th down.”

Scoop August 14, 2012 at 11:29 am

Why limit the conversation to punting? He’s also suggesting no field goals, which I find that even more surprising. 4th and 10 from the 20 makes for an easy field goal but a really tough first down conversion. There’s no way to spread the D with so little room to work. Can your expected points from going for it really exceed the 2.9 points you’d expect from kicking the FG?

Nick August 14, 2012 at 4:57 pm

A couple points:

He’s a high school coach. Most high school teams do not have reliable kickers. There are many scenarios where an NFL team would clearly maximize expected points by kicking a field goal and HS team would clearly maximize expected points by going for it. There is now way a HS kicker is averaging 2.9 points per attempt on a 37 yard field goal, even in ideal weather conditions.

The lack of skill at the HS level also applies to punting decisions. The average punt will advance the ball fewer yards in the air and the average punt return will be larger, making the net benefit of the punt substantially smaller.

Turning the ball over on downs is also not as bad when you’re deep in your opponents territory bc you will have a field position advantage in subsequent possessions. There are people who have modeled the value of various game states and found that 1st and 10 from close enough to your own end zone is actually worth negative points because of this factor (and the threat of a safety).

Brian Donohue August 14, 2012 at 11:41 am

So, when in doubt…?

Go Kings, Go! August 14, 2012 at 12:42 pm

The incredibly great TMQ has been banging this drum, and highlighting the Pulaski Academy, for some time. He will be pleased.

Jim Nazyum August 14, 2012 at 1:28 pm

I say throw the long bomb pass on every down.
Every. Single. Down.

Charles LeGrand August 14, 2012 at 4:15 pm

Related idea. Pro tennis players should risk more double faults and hit two first serves instead of a safer and less offensive second serve. From a couple years ago in the NYT.

http://www.nytimes.com/2010/08/30/sports/tennis/30serving.html?pagewanted=all

John David Galt August 14, 2012 at 10:14 pm

The graph confirms my intuition that when inside the 35 or 40, the usual choice is neither to punt nor go for a first down, but to try a field goal. I’d expect this to be the best choice most of the time.

Still, I would want to plot average gain (in yards), or even the average outcome of the possession, against current field position for each of the three options before making a decision to change overall strategy like that.

The outcome of going for it will also vary with how far you have to go to get the first down. If it’s 4th and inches it may make sense — if it’s 4th and 25, forget about it. And time left on the clock will be a factor.

Reinholt August 16, 2012 at 9:06 am

A few notes:

1 – As mentioned above, kicker skill matters. If I have an NFL-caliber kicker, a college-caliber kicker, or a high school-caliber kicker, the decision points change. FG from inside the 20 is nearly (though not completely) automatic for the NFL kicker, yet could be quite an adventure in high school. So the “value” of kicking field goals is going to be lower in the first place for college compared to the NFL, and high school compared to college, meaning the “value” of going for it doesn’t have to be as high for it to be worthwhile.

2 – Kicking decisions are context dependent. At the 50 yard line, faced with 4th and 1, punting is a very different decision than at the 50 yard line, faced with 4th and 28. Also, again, skill matters. NFL teams pass better, on average, than college teams (though that has been changing), but most high school teams still lack reliable quarterbacks and focus more on running the ball. So down and distance should be considered.

3 – Therefore, each team, based on both match-up and personnel, are going to have different “optimal” states for what they should do in each game. There’s no obvious answer that can be applied everywhere, and there is career risk to using an outlier strategy in many situations. However, most strategies focus on the NFL (or sometimes college), and probably aren’t fully applicable to high school, where quality of kicking and QB play are significantly lower.

With all that said: punting less and going for it more is almost certainly an optimal strategy. Punts should occur when the probability of offensive success is low (4th and long), risk is high (long drop backs leading to sacks), or harm from failure is high (deep on your own side of the field).

However, never go for it / always go for it are both losing strategies. The correct move is almost certainly “go for it much more than currently, but not always”, again, dependent on personnel and match-up.

Comments on this entry are closed.

Previous post:

Next post: