Other Critical Analysis of 4th Downs

[exe ex machina]

To Punt or Not to Punt? The Debate Continues

Stephen J. Dubner
09/11/2012 | 2:48 pm

(Photo: Ben Stanfield)
A reader named David Stokes writes to say:

Last night’s Raiders – Chargers game gave one team a unique opportunity to implement the no-punt strategy.

With the Raiders’ long-snapper hurt, the Raiders coach had a much less risk-averse reason to try always going for it on fourth down. Especially after the first punt was blown and the punter tackled with the ball, who could blame the coach for going for it on fourth every time?

Alas, he proceeded to attempt more punts, and three in a row were blocked or otherwise blown.​

FWIW, I think someone should make a documentary about long snappers. I am not kidding.

1. Shal says:
   September 11, 2012 at 2:57 pm
   I wrote a paper on Kevin Kelley for a sport management class. What really struck me about his strategy isn’t that he figured this out. Once you look at the numbers (esp in high school football where punts dont travel as far), its a pretty obvious winner. What really got to me is that the reason this works is that he’s gotten the school and fans on board with what he’s doing. I can just see people flipping out because their team blew a 4th and one ‘gimme’ and gave the other team great field position. Not at this school– everybody knows what they’re doing and why they’re doing it.
   
   
   2
   
   0
   Reply
2. Tony says:
   September 11, 2012 at 3:25 pm
   The Redskins’ long snapper reportedly played most of the game with a broken arm. That’s the degree to which teams want to avoid having to use a back-up long snapper. I’m almost dumbfounded that the Raiders kept trying to punt with a back up long snapper.
   
   
   2
   
   0
   Reply
3. Jesse says:
   September 11, 2012 at 5:37 pm
   If there is a documentary to be made, let it be dubbed “Snapdragons”
   
   
   2
   
   0
   Reply
4. Ricardo says:
   September 11, 2012 at 5:50 pm
   Agreed. I saw this debacle. So who is this new dumb coach and why is he not an economist like you can can parse the details and the #s? As in, why are coaches whores to the status quo?
   
   
   0
   
   0
   Reply
5. Gary says:
   September 11, 2012 at 7:27 pm
   The choice is very understandable for the Raiders. They have the best punter in football history. And they have the kicker with the strongest leg in history. They have a huge advantage in the kicking game over other teams and are built to take advantage if that. It isn’t that ready to just switch.
   
   
   0
   
   0
   Reply
6. David L says:
   September 11, 2012 at 9:27 pm
   I am more surprised that coaches in the NFL aren’t laying down the law that any kickoff that reaches the goal line should not be returned, but downed for a touch back.
   Between the risk of injury, and the consistent lack of productivity, it just doesn’t make sense to try to run it out. Particularly when you factor in how many holding calls and other penalties are called. From the games I have watched I doubt if the average return is much beyond the 25 and the median is probably inside the 20. Save you team the effort, spare the special teams, and take the ball at the 20. But I am sure there are statistics out there to figure this out.
   
   
   2
   
   0
   Reply
   • Nic says:
     September 11, 2012 at 10:59 pm
     A lot of the time the kick returner is not out on the field any other time. A rookie WR who doesn’t know all his routes yet or a new RB. This can be their time to shine and show the coaches how much speed and agility you have in game situations.
     
     
     0
     
     0
     Reply
7. A.J. says:
   September 11, 2012 at 9:28 pm
   Why is there a picture of Penn State on this article?
   
   
   0
   
   0
   Reply
   • Jay says:
     September 11, 2012 at 10:59 pm
     Go Hoos!
     
     
     0
     
     0
     Reply

1. • http://www.freakonomics.com/2012/09...e-debate-continues/?replytocom=330344#respond

 

[exe ex machina]

Sean Payton is the NFL’s most aggressive coach on fourth down
Posted by Michael David Smith on April 4, 2015, 12:17 PM EDT

AP
Saints coach Sean Payton goes for it on fourth down far more often than any of his colleagues around the NFL.

In fact, in 2014, Payton went for it on fourth down more than twice as often as an average NFL coach would have, when faced with the same circumstances. FootballOutsiders.com analyzed every fourth-down decision in the NFL in 2014, adjusting for the distance needed for a first down as well as where the team was on the field and the score of the game (filtering out times when a coach goes for it because his team is getting blown out), and found that Payton had an Aggressiveness Index of 2.01.

Many statistical analysts say that NFL coaches should be more aggressive about going for it on fourth down, but Payton says he relies less on statistical analysis than on gut feeling. Payton explained during the 2014 season that when the Saints face a fourth down, he usually just makes his decision based on whether he has a feeling that the Saints’ offense can call a play that will work in that situation.

“Fourth downs maybe is sometimes a feel, and a lot of times it is not having a play,” Payton said. “There are some times where you have got a play you cannot wait to run, so you would be more apt to be aggressive.”

Whatever the reasons, Payton feels like going for it more often than other coaches. Analyzing fourth-down data going back to 1989, Football Outsiders found that there have only been eight seasons when a coach was more than twice as aggressive as his peers — and two of those seasons were Payton-coached Saints teams. In addition to the Saints’ Aggressiveness Index of 2.01 last year, the Saints had an Aggressiveness Index of 2.22 in 2007.

Unfortunately for Payton, those also happened to be his only two seasons with a losing record. Payton isn’t afraid to go for it when his team needs it, but he’d be better off if his team didn’t need it.

 

[rumply]

That last line says it all doesn't it & could apply to many coaches (esp me in Madden ) - 'Payton isn’t afraid to go for it when his team needs it, but he’d be better off if his team didn’t need it.'

 

[JeffDunne]

One reason why our 4th down attempts have been so high in the past has been our poor kicking game.

4th and 1 or attempt a 45yd+ FG attempt? The odds of making the 4th down were generally greater than making the FG attempt.

 

[Dirty Bird]

One reason why our 4th down attempts have been so high in the past has been our poor kicking game.

4th and 1 or attempt a 45yd+ FG attempt? The odds of making the 4th down were generally greater than making the FG attempt.

Then perhaps you COULD round out a good team rather than throwing 50% of your salary cap at 3 players.

 

[JeffDunne]

Then perhaps you COULD round out a good team rather than throwing 50% of your salary cap at 3 players.

Rather general and inaccurate statement.

Are you suggesting our kicking problems have been a result of cap pressure?

Think about that for a moment. Sure you want to run with that?

 

[ashley12]

The problem with 4th downs, and other "risky" moves in sports, such as playing a more attacking soccer style, playing star basketball players when in foul trouble etc... is that the current accepted play (punting a lot, defending a lot, resting the stars for fear of a foul out) is considered normal. If you lose a game after doing "normal" things, generally speaking everyone says "eh, what can you do" and the coach gets a lot less hate compared to when he tries something risky and it doesn't work (such as Belichick going for it vs the Colts on 4th and 2).

As well as that, coaches for obvious reasons value "not getting firedness" more than "increase win chance %" in terms of an objective, so as such can make conservative decisions that increase the former at the expense of the later. A solution to this would be the measure coaches by a different metric, but the owners and decision makers are generally speaking, not that sharp. It's just another small contributing thing as to why the Patriots just crush the rest of the league by giving Belichick control to make decisions without worrying for getting sacked if things don't work.

 

[exe ex machina]

Belichick only survives so long because he has 4 lombardis. Had such a great win percentage for so long. But he didnt START at the Pats with such total power. He had to earn it gradually over time. He wasn't so ballsy early on. And had people around him whom he delegated aspects of football operations to.

Some coaches get given a level of power higher than others (like just the 53 man roster control) and they end up getting fired fairly quickly after say 2 years of failure.

For instance....Chip Kelly has risen fairly quickly with more power inside the Eagles org. If he bombs out this year, he probably runs the risk of losing his job.

 

[ashley12]

True, but the trust and power given to him was given over a far quicker period than it was to Harbaugh for example, which led to the eventual untenable situation and the hiring of someone who could pass for a pornstar in the Bay Area.

The issue isn't necessarily power to make decision, but just trust from management to rely on other ways of measuring success than just wins or losses.

This is a good example of what I think would never happen in the NFL

http://www.theguardian.com/football/blog/2015/feb/22/brentford-mathematical-modelling-denmark

Another example is the ridiculous contract given to Flacco, based predominantly on his winning the SB in his contract year. Rahim Moore doesn't fall over on pretty much the last play of regulation, Ravens don't win that game, and Flacco doesn't get nearly as much money. Flacco going forward is going to be pretty much the same quarterback either way, except now because of one circumstance for the most part outside his own control, he gets all the cash and puts a lot of pressure on the Ravens going forwards with his cap numbers.

Generally speaking, the sooner the sports world stops using simple but inaccurate measures for what makes a player good or bad (points in basketball, goals and assists in soccer, W-L when evaulating coaches and teams) the better the decisions dictated by these measures will become.

 

[exe ex machina]

True, but the trust and power given to him was given over a far quicker period than it was to Harbaugh for example, which led to the eventual untenable situation and the hiring of someone who could pass for a pornstar in the Bay Area.

The issue isn't necessarily power to make decision, but just trust from management to rely on other ways of measuring success than just wins or losses.

This is a good example of what I think would never happen in the NFL

http://www.theguardian.com/football/blog/2015/feb/22/brentford-mathematical-modelling-denmark

Another example is the ridiculous contract given to Flacco, based predominantly on his winning the SB in his contract year. Rahim Moore doesn't fall over on pretty much the last play of regulation, Ravens don't win that game, and Flacco doesn't get nearly as much money. Flacco going forward is going to be pretty much the same quarterback either way, except now because of one circumstance for the most part outside his own control, he gets all the cash and puts a lot of pressure on the Ravens going forwards with his cap numbers.

Generally speaking, the sooner the sports world stops using simple but inaccurate measures for what makes a player good or bad (points in basketball, goals and assists in soccer, W-L when evaulating coaches and teams) the better the decisions dictated by these measures will become.

Nice one about Tomsula there

Agree about Flacco.

However, i'm not sure about a whole new complex mathematical measure to determine good/bad player. It's like ESPN's Total QBR. Which had Tebow rated better than Aaron Rodgers that year. You can make statistics paint any type of picture to suit an argument. There's still a ton of stuff that doesn't always show up in stats. The importance of a player, how he affects those around him, leadership, could be like a general on the field barking orders, assignments, but other players make the plays/stats.

In terms of 4th downs, owners/gms are always going to judge a coach's playcalling based on the success/failure of those plays. It's a result-driven thing. No other way to assess it but did it work or not. If a coach fails too much, goes for broke too much, and it costs a team a win or two or more, he WILL get fired. Risk aversion. But also hedging your bets, smarter playcalling.

Like Sean Payton said, a lot of the time he's just going off feel, no analysis/maths of trends etc that a lot of other coaches might rely on to make all their down-and-distance decisions.

There are definitely many examples every year, where a coach takes the chicken-shit option of punting on 4th down at the opposition's 45 yard line, who are losing, and the net yard gain is like 10 yards, and in the end the other team scores quickly and kills off the other team. Seems to me a LOT of HCs are lacking the ability to FEEL the momentum in the game to make a ballsy call. Payton's surprise onside kick in the Super Bowl one of the best examples of a HC with a finger on the pulse of the game, and making an incredibly brave/tough decision. Who dares wins.

 

[ashley12]

Tebow vs Rodgers thing isn't true,at least over the full sample of the year.

http://espn.go.com/nfl/qbr/_/year/2011

Rodgers 1st 87.1
Tebow 30th 29.2

Generally speaking, people using W-L to measure QBs actually ends up working often in a backwards kind of way, but thats only because having a good or bad QB is the number 1 thing to determining whether your team is good or not. With a couple of small exceptions, rating teams 1-32 based on W-L record would give a pretty accurate rating of QBs, but would give a bad rating of coaches, RBs and offensive lines. McCarthy and Pagano would be the third ranked coachs, but would be bottom 12 in the league generally considered based mostly on under-utlisation of their awesome QBs.

While I agree with your point on Payton generally, and that onside kick, I think when he talks about "feel", its not a case of "well, am I feeling lucky, punk?" and making decisions based purely on the rumblings of his loins at the moment of decision, but more a case of making decisions that aren't based on a binary yes or no based on distance, score and time remaining, but also a feel for what kind of defence he expects to face, and whether they have a good enough play to exploit any particular defence he expects to face. I completely agree that for the most part coaches are pretty timid when it comes time to gamble, and since Payton is more aggro than most, he is therefore a decent decision maker every thing else being equal.

For example, THAT play from the most recent superbowl. If you decide that Carroll is a peanut for passing rather than running as if it is purely a decision between A) run and B) pass, then you are going to make mistakes in your analysis. Various other factors are in play, and the same applies for BB and his not calling timeout prior, which was a pretty awesome leveling war between the two coaches. Saying

"passing the ball is bad and stupid"

is bad analysis, while saying something like the initial decision to pick a formation that had only one TE vs a goal-line defensive set with Marshawn Lynch in the backfield, while pointing out his success with 6 blockers as opposed to 7+ is quite good analysis, even if you come to the same decision of the simple analysis

TL;DR, long rambling post, but generally speaking casual fan analysis = bad, objective based statistical analysis = good, provided stats are used correctly.

 

[Simple Jack]

TL;DR, long rambling post, but generally speaking casual fan analysis = bad, objective based statistical analysis = good, provided stats are used correctly.

A billion times the bolded. People stay 'statistics can be used to show anything' (a la David King ), but that's only if they aren't used appropriately.

Statistics that are collected and analysed properly reveal the actual state of something. The problem is idiot fans (and presenters) just assuming because something has a decimal point or % sign after it that it's of equal value to every other 'statistic'.

 

[juss]

4th and yes

 

[TheGreatBarryB]

Belichick only survives so long because he has 4 lombardis. Had such a great win percentage for so long. But he didnt START at the Pats with such total power. He had to earn it gradually over time. He wasn't so ballsy early on. And had people around him whom he delegated aspects of football operations to.

Some coaches get given a level of power higher than others (like just the 53 man roster control) and they end up getting fired fairly quickly after say 2 years of failure.

For instance....Chip Kelly has risen fairly quickly with more power inside the Eagles org. If he bombs out this year, he probably runs the risk of losing his job.

BB made 'ballsy' decisions as DC with Parcells. In his first HC job he made a ballsy decision to dump Browns QB Bernie Kosar who was a huge fan favourite, I think for Vinny Testaverde. Copped a lot of flak but it was the right decision.

 

[exe ex machina]

NFL teams averaged 3.7 punts per game in 2020, the lowest in league history.
Many NFL coaches are embracing what analytics advocates have said for years: punting -- especially on fourth and short at or near midfield -- is counterproductive if a team's goal is to put points on the scoreboard. From the mid-80s to 2017, NFL teams averaged around 4.8 punts per game -- a rate that has dropped in each of the past four seasons. Fourth down aggressiveness has taken center stage during the 2021 postseason. While Steelers head coach Mike Tomlin elected to punt on fourth and short in a Wild Card Weekend comeback attempt against Cleveland, Kansas City head coach Andy Reid called a game-winning fourth down play against Cleveland in the Divisional Round, directing backup QB Chad Henne to take the snap -- instead of trying to draw the Browns offsides -- and throw a short pass to Tyreek Hill. Titans head coach Mike Vrabel decided to punt on fourth and short while trailing the Ravens in the Wild Card round -- a disastrous decision that doomed Tennessee to a first round loss. More analytically minded coaches could make punts from midfield a relic of an ultra-conservative NFL past.
SOURCE: NYTimes.com
Jan 25, 2021, 9:22 AM ET

 

[ashley12]

ooooh, nailed the comment on the Flacco contract being an albatross...

 

[drd23]

 

[aussiedude]

Just NFL coaches acting like NFL coaches.

College coaches tend to coach to win, NFL Coaches are more often coaching not to lose.

 

[Dawggin It]

Someone explain to me number 10 on that list.

Opening drive of the game, 4th and 3 from their own 32. Who isn't punting in that situation? That's the 10th worst call over 12 games??

 

[aussiedude]

Someone explain to me number 10 on that list.

Opening drive of the game, 4th and 3 from their own 32. Who isn't punting in that situation? That's the 10th worst call over 12 games??

i think it took the 55yd return into account. so ended up a net 10yd loss

 

[Dawggin It]

i think it took the 55yd return into account. so ended up a net 10yd loss

Thanks for the explanation, not sure why they would take that into account in a list of worst decisions to kick though.

Surprised green Bay is only 3rd on the list as well!

 

[aussiedude]

Thanks for the explanation, not sure why they would take that into account in a list of worst decisions to kick though.

Surprised green Bay is only 3rd on the list as well!

taking hindsight into the decision is always stupid.

 

[samFFC]

Someone explain to me number 10 on that list.

Opening drive of the game, 4th and 3 from their own 32. Who isn't punting in that situation? That's the 10th worst call over 12 games??

#3???

Where the heck is this list from?

Bills 4th and 1, on THEIR 15, down 10-7 in the 2nd quarter. Punt was 58 yards (net 44). ESPN has it listed as Colts 53% to Bills 56% after the punt.

 

[drd23]

Where the heck is this list from?

Ben Baldwin is an economist, writer for The Athletic and one of the more prominent NFL Analytics Twitter members

That list is from his 4th down decision website, which is explained in this article

NFL fourth-down aggressiveness

Our numbers guru creates our own fourth-down model and explains where the numbers come from.

theathletic.com

There are a number of sites which provide win probabilities, but a number of them are slightly different from each other - ESPNs one is far from infallible, as that screenshot aussiedude has posted from a CFB game involving UCLA attests to. #3 appears to be another where his 4th down calculator is taking the results of the punt into the calculation (which seems stupid)

 

[samFFC]

Ben Baldwin is an economist, writer for The Athletic and one of the more prominent NFL Analytics Twitter members

That list is from his 4th down decision website, which is explained in this article

NFL fourth-down aggressiveness

Our numbers guru creates our own fourth-down model and explains where the numbers come from.

theathletic.com

There are a number of sites which provide win probabilities, but a number of them are slightly different from each other - ESPNs one is far from infallible, as that screenshot aussiedude has posted from a CFB game involving UCLA attests to. #3 appears to be another where his 4th down calculator is taking the results of the punt into the calculation (which seems stupid)

Fair point.
Just ridiculous though - wasn't even a huge punt return.