When Do You Try a 50 yard+ Field Goal?

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,297
AZ
The Patriots passed twice on long field goals yesterday -- they would have been 54 and 55 yard attempts. Both times, they opted instead for a punt.

Net on one punt was 26 yards, so they gained 33-34 yards relative to opponents' starting field position if they had missed, and they gained 26 yards in field position relative to an unsuccessful fourth down offensive attempt for first down (though it depends on whether they would have gained or lost yards on the fourth down attempt).

Net on the other punt was 18 yards, so they gained 25-26 yards relative to opponents' starting field position if they had missed, and they gained 18 yards in field position relative to an unsuccessful fourth down offensive attempt.

So, the question is when do you attempt a field goal of more than 50 yards? For all we know, Gostkowski told the staff that his range yesterday topped out at 52 yards, in which case the answer for the two scenarios yesterday would have been "never". There just isn't enough information to second guess yesterday's decision, so I'm more just asking the question in the abstract. We know that the Patriots are not opposed to going for 50 yard plus field goals as a matter of course. Gostkowski is 19 for 26 on them.

But what are the factors? Weather, for sure. Score. Down and distance. A fourth and short probably brings going for it back into the expected value conversation in a way that overcomes the field position benefit, especially when taking into account that you don't lose the 7 to 8 yards of extra field position. And, probably most important, consequence of missing. If the clock situation is such that the other team can't really do anything with the field position even if you miss, then it makes the decision easier. There is also the possibility of a block or a return, but on the other side of the coin, there is also a possibility of a penalty that continues the drive with a first down. (And punting is not free of risk.) But, more generally, what is the expected value of the field position, is, I guess the ultimate question that I'm asking. If you view most 50+ field goals as something around a 50/50 proposition, there's a 1.5 point expected value to the attempt. Is 25 yards of field position worth giving away those points?
 

Bowhemian

Member
SoSH Member
Nov 10, 2015
5,701
Bow, NH
I think that part of the decision to punt had to do with the defense. BB didn’t have enough faith in the D to be able to stop Houston with a short(er) field. My guess is that he thought they had a better chance to stop them if they made Houston go longer distances. It was all about field position.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,069
Hingham, MA
I think that part of the decision to punt had to do with the defense. BB didn’t have enough faith in the D to be able to stop Houston with a short(er) field. My guess is that he thought they had a better chance to stop them if they made Houston go longer distances. It was all about field position.
I agree with this - BB wanted Watson to have to drive the field. Simply put Ryan Allen has to do a bit better here. If he pinned them inside the 15 or 10 on those punts we would probably view the decisions more favorably
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
There's also the choice you mention but that isnt highlighted in the post intro or title....they can go for the 1st down

There's also a somewhat steep decline curve in success rate after 50 yards, so lumping 50+ yard field goals may not be a great way to bucket it

https://fivethirtyeight.com/features/kickers-are-forever/

Game states and field position are one of the building blocks for many advanced football metrics, which builds off of Pete Palmer's The Hidden Game of Football. The metric usually used is expected points for a team given field position and what 4th down bot used to try and calculate.

This link (slightly dated) has some graphs showing expected points from different game states.

http://www.advancedfootballanalytics.com/index.php/home/stats/stats-explained/expected-points-and-epa-explained


Those graphs are usually league wide samples, so they need to be adjusted for factors like weather or strength of opponent.

In general, teams are too risk averse based strictly on the mathematical calcuations of the value of field position (i.e. excluding ideas like momentum)

One comment: If your defense sucks and your offense is good, you should generally be going for first downs more often on offense, not less often. Also punting on 4th and ~5 from inside the opponents 40 with a good offense is going to be pretty dumb in most game state situations.
 

DennyDoyle'sBoil

Found no thrill on Blueberry Hill
SoSH Member
Sep 9, 2008
42,297
AZ
It's good stuff stitch. I guess there are just so many variables that it's difficult to come up with iron clad rules one way or the other -- it's not quite like baseball run expectations matrices.

Whenever I'm watching the NFL, it always seems like there are these expected plays that everyone seems to accept as the right things to do, but that always seem to be worth discussing. Maybe the expected is just what scared coaches do, but even BB does them most of the time so that it's tough to know. Maybe they are expected because they are correct.

In general, though, it seems like many decisions are made to avoid a result that ends in a likely loss right away. It seems that most coaches, when given the choice, will decide on the option that prolongs the feeling "oh shit we lost" as long as possible. The classic one is down 11, in the red zone late, with fourth and short. Pretty much every coach will kick the field goal there. It rarely seems like the right move to me.

One yesterday that bothered me was the Patriots punting from their own half of the field on fourth and short with 5:55 left. It felt to me at the time that giving the ball back might end up being the end of the game, because either the Texans would run out the clock or would score a TD and go up 9. If it had been fourth and long, it would be different, but it felt as though, even if everything went right, and they got the ball back, they very likely were going to face a situation worse than 4th and 1, and the only thing that would be different is that the clock would leave them no time but to try it. I wasn't exactly right, though if you had offered me fourth and one on the 20 when the Patriots were in 3d and 18 and the 30, I would have traded. But even not judging it in hindsight, I don't think it was as much an off-the-table option as I think many fans and coaches would think. Most would simply shoot down the idea of going for it there as crazy out of hand. Maybe that's right.

I dunno, I'm hijacking my own thread, but with field goals and fourth downs especially it seems like there is way too much concern paid to the downsides and not enough weight given to the relative upside of field position -- even with a great punter, which ours is not.
 

Toe Nash

Member
SoSH Member
Jul 28, 2005
5,600
02130
I don't really buy that Bill was worried about the defense.

The first punt was at the beginning of the second quarter. To that point the Texans had gone:
-3 and out
-6 plays and a FG (with a long pass interference after Watson heaved a ball into tight double coverage)
-3 and out
-2 plays and a TD on a short field after the mediocre Allen punt from his own goal line (after the self-sack by Brady)

They were down 10-7. At this point I don't see a good reason to be scared of the Texas offense. As it was Watson threw a pick two plays later.

After that punt we had the Texans drives:
-3 plays, interception (pass to no one)
-10 plays, FG - pretty nice drive, though helped by the unsportsmanlike late hit or whatever it was on Gilmore

Pats up 14-13. Sack fumble and Clowney return happens here.

-Then 7 play drive, punt

2nd half

-10 plays, TD - another nice drive

Here's where the second punt happened. The Texans had put together two good drives out of 8 and even though they had scored 27 points I don't think they were in any way dominating. Punting is defensible but not because you're worried about the Texans going 60 yards; more because of the possibility of pinning a rookie back against his own goal line.

Not knowing the details of the wind, which direction you're facing, (open end of Gilette) etc., I would have gone for the first down on the first punt (3rd and 8) and tried the FG on the second punt (where a successful kick actually puts you ahead).

Allen does definitely need to be better.
 

ElcaballitoMVP

Member
SoSH Member
Nov 19, 2008
3,932
I thought I'd have a quick answer, but there really isn't one. Opponent, weather, field position, time of game, and on and on we go.

I'd say for yesterday's game, he was relying on his defense to stop a rookie QB from driving the entire length of the field. He was also relying on Ryan Allen to do his job. If he pins the Texans inside their 5 on one of those punts, it makes things even tougher on Watson (who deserves credit, he played really well). If they get them to go 3 and out and punt, the Pats get a short field.

While Ghost has been pretty solid from 50+ in his career, he was 2 for 4 last year, so maybe Bill decided to punt vs taking a 50/50 chance of getting 3 points or turning over field position.
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
It's good stuff stitch. I guess there are just so many variables that it's difficult to come up with iron clad rules one way or the other -- it's not quite like baseball run expectations matrices.

Whenever I'm watching the NFL, it always seems like there are these expected plays that everyone seems to accept as the right things to do, but that always seem to be worth discussing. Maybe the expected is just what scared coaches do, but even BB does them most of the time so that it's tough to know. Maybe they are expected because they are correct.

In general, though, it seems like many decisions are made to avoid a result that ends in a likely loss right away. It seems that most coaches, when given the choice, will decide on the option that prolongs the feeling "oh shit we lost" as long as possible. The classic one is down 11, in the red zone late, with fourth and short. Pretty much every coach will kick the field goal there. It rarely seems like the right move to me.

One yesterday that bothered me was the Patriots punting from their own half of the field on fourth and short with 5:55 left. It felt to me at the time that giving the ball back might end up being the end of the game, because either the Texans would run out the clock or would score a TD and go up 9. If it had been fourth and long, it would be different, but it felt as though, even if everything went right, and they got the ball back, they very likely were going to face a situation worse than 4th and 1, and the only thing that would be different is that the clock would leave them no time but to try it. I wasn't exactly right, though if you had offered me fourth and one on the 20 when the Patriots were in 3d and 18 and the 30, I would have traded. But even not judging it in hindsight, I don't think it was as much an off-the-table option as I think many fans and coaches would think. Most would simply shoot down the idea of going for it there as crazy out of hand. Maybe that's right.

I dunno, I'm hijacking my own thread, but with field goals and fourth downs especially it seems like there is way too much concern paid to the downsides and not enough weight given to the relative upside of field position -- even with a great punter, which ours is not.
Agree with pretty much all of that, particularly the bolded. A lot of times these are in low equity situations to begin with, but teams give up equity to extend the game on a regular basis. Raiders kicking from the 4 yard line last night to cut the lead to 14 comes to mind (Saints did something similar in their opener too). Many times exactly what you describe plays out, teams are left with a more difficult conversion later in the game than the one they chose to forgo earlier in the game.

Im not sure Id have gone for it on 4th and 1 with 6:00 left, but it certainly would have been defensible IMO. Results wise, the Pats ended up in pretty much the same spot they would have been in had they failed, just three minutes later.
 
Last edited:

PayrodsFirstClutchHit

Bob Kraft's Season Ticket Robin Hoodie
SoSH Member
Jun 29, 2006
8,319
Winterport, ME
This is purely anecdotal, but it certainly seems like the Pats have been much more conservative over the past 5 years or so in these situations once they pass the 50 yard yard line and before they reach a reasonable FG distance.

Back in the day, I recall the Pats going for it on 4th down in these situations as the default option. It may have been more that the Pats were winning but such a large margin in many of these games, but it seems like punting has become the default state versus the long FG or going for it.
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
Anecdotally I agree with you. They've been more conservative at the end of the half too but, in my opinion, Id point to the Eagles game disaster in 2015 as the trigger for that shift.
 

IdiotKicker

Member
SoSH Member
Nov 21, 2005
10,785
Somerville, MA
I dunno, I'm hijacking my own thread, but with field goals and fourth downs especially it seems like there is way too much concern paid to the downsides and not enough weight given to the relative upside of field position -- even with a great punter, which ours is not.
So a few things on this.

There are a couple studies floating around as to the value of field position, but the general rule of thumb is that every 10 yards works out to between 0.6 and 0.8 expected points allowed. So if you miss a 55-yard field goal and give a team the ball on their 45-yard line as opposed to on their 10-yard line, you're talking about giving them somewhere around 2-3 points in expected value. The hit rate of 50+ yard kicks over the last 5 years is 62.7%, with Gostkowski clocking in at 77.8%, but the sample size for him is tiny (18 kicks). But for the league, assuming they can drop a punt around the 10-yard line, you're looking at an expected value around one point on 50+ yard kicks (60% chance of scoring 3 points minus 40% chance of giving an extra 2.5 points).

And that works out with how you think about football as well - you make that kick, you've got 3 points on the board. You miss it, and the opposing team needs a couple first downs to be in FG range where they're looking at a 70-75% chance of hitting a kick, or continuing a successful drive for a TD.

As for the punt in question, Allen did exactly what he needed to do - he dropped it on the 11-yard line for a fair catch. And this gets into how the Patriots build their special teams roster. NE prizes punters who are good directionally and generate average hang, while sacrificing distance. Why? Because the thought is that with NE's offense, the amount of times you need to change field position with a 55-60 yard punt are minimal. So as opposed to paying $3.5M for an upper-echelon punter, they instead pay around $2M, and have the extra $1.5M to spead among the other special teamers they like to target, or to beef up other positions.

In 2015, I graded Allen out as a top 10 punter, but that's a not the norm for him. Last year he was a bottom-10 punter, but his realistic landing spot is right around average - he doesn't do much distance-wise, is average in terms of hang-time, and above-average to excellent directionally. Watch NE punt any time they are on their 45-yard line or further ahead - find the side Slater lines up on (usually the right), and watch Allen drop in right on the numbers on that side of the field for no return. Allen is the kind of punter NE wants to have - left-footed, below-average cost for a veteran, with good directional skills and a pedestrian leg otherwise. They don't need to pay for a booming leg, and so they don't. On the Texans, Allen would be a terrible fit, which is why they have Lechler, who had my third-highest distance control grade last year. But Lechler graded out worse than Allen directionally, and wouldn't be a fit for what NE does, even though he's the same cost.
 

Marceline

Well-Known Member
Lifetime Member
SoSH Member
Sep 9, 2002
6,441
Canton, MA
Anecdotally I agree with you. They've been more conservative at the end of the half too but, in my opinion, Id point to the Eagles game disaster in 2015 as the trigger for that shift.
Were there 4th down attempts that backfired in that game? I've kind of blocked it out, I can't remember the specifics.
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,069
Hingham, MA
Anecdotally I agree with you. They've been more conservative at the end of the half too but, in my opinion, Id point to the Eagles game disaster in 2015 as the trigger for that shift.
I think this actually goes back to the second half of 2014 - they had some disasters there too. Off the top of my head @ GB, home vs. Miami, playoff game vs. Baltimore
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
So a few things on this.

There are a couple studies floating around as to the value of field position, but the general rule of thumb is that every 10 yards works out to between 0.6 and 0.8 expected points allowed. So if you miss a 55-yard field goal and give a team the ball on their 45-yard line as opposed to on their 10-yard line, you're talking about giving them somewhere around 2-3 points in expected value. The hit rate of 50+ yard kicks over the last 5 years is 62.7%, with Gostkowski clocking in at 77.8%, but the sample size for him is tiny (18 kicks). But for the league, assuming they can drop a punt around the 10-yard line, you're looking at an expected value around one point on 50+ yard kicks (60% chance of scoring 3 points minus 40% chance of giving an extra 2.5 points).

And that works out with how you think about football as well - you make that kick, you've got 3 points on the board. You miss it, and the opposing team needs a couple first downs to be in FG range where they're looking at a 70-75% chance of hitting a kick, or continuing a successful drive for a TD.

As for the punt in question, Allen did exactly what he needed to do - he dropped it on the 11-yard line for a fair catch. And this gets into how the Patriots build their special teams roster. NE prizes punters who are good directionally and generate average hang, while sacrificing distance. Why? Because the thought is that with NE's offense, the amount of times you need to change field position with a 55-60 yard punt are minimal. So as opposed to paying $3.5M for an upper-echelon punter, they instead pay around $2M, and have the extra $1.5M to spead among the other special teamers they like to target, or to beef up other positions.

In 2015, I graded Allen out as a top 10 punter, but that's a not the norm for him. Last year he was a bottom-10 punter, but his realistic landing spot is right around average - he doesn't do much distance-wise, is average in terms of hang-time, and above-average to excellent directionally. Watch NE punt any time they are on their 45-yard line or further ahead - find the side Slater lines up on (usually the right), and watch Allen drop in right on the numbers on that side of the field for no return. Allen is the kind of punter NE wants to have - left-footed, below-average cost for a veteran, with good directional skills and a pedestrian leg otherwise. They don't need to pay for a booming leg, and so they don't. On the Texans, Allen would be a terrible fit, which is why they have Lechler, who had my third-highest distance control grade last year. But Lechler graded out worse than Allen directionally, and wouldn't be a fit for what NE does, even though he's the same cost.
Allen was pretty mediocre in touchback percentage, inside the 20 percentage, and fair catch percentage last year (~16th and 18th but possible there's a non qualifier or two above him skewing that slightly downward). Return yards allowed were second best in the league. Is there anywhere to find hang time stats? He seemed to struggle pretty bad on that in the opener, just wondering how that's trending. I buy he's relatively more effective at what the Pats value at punter. Im not sure he's an average punter right now. We'll see, early in the season.

Were there 4th down attempts that backfired in that game? I've kind of blocked it out, I can't remember the specifics.
They had a blocked punt for a TD right before the half and the game sort of unraveled from there. I dont mean they are more conservative on 4th down decisions before half time, just that they run out the clock more.
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
I think this actually goes back to the second half of 2014 - they had some disasters there too. Off the top of my head @ GB, home vs. Miami, playoff game vs. Baltimore
GB wasnt really a clock decision, Pats scored a TD before GB answered, and Miami game they went conservative and it blew up in their face.

But maybe an accumulation of things, who knows, just seems like a marked increase in conservatism since that point but just my anecdotal opinion
 

tims4wins

PN23's replacement
SoSH Member
Jul 15, 2005
37,069
Hingham, MA
GB wasnt really a clock decision, Pats scored a TD before GB answered, and Miami game they went conservative and it blew up in their face.

But maybe an accumulation of things, who knows, just seems like a marked increase in conservatism since that point but just my anecdotal opinion
True. They also threw at pick in the end zone at the end of the first half in the SD the week after the GB game. Then like you said the following week they tried to run out the clock against Miami, had to punt, and gave up a TD after a lousy punt. Then the following week at the Jets they had the ball with 52 seconds left at their own 17, took a sack, then sat on the ball. That was week 16. Week 17 they didn't play to win, then in the first playoff game against Baltimore Brady threw the pick and the Ravens scored a TD. So basically they fucked up the end of the half in five straight games which is what I remembered. Like you said maybe they started adjusting in 2015.
 

IdiotKicker

Member
SoSH Member
Nov 21, 2005
10,785
Somerville, MA
Allen was pretty mediocre in touchback percentage, inside the 20 percentage, and fair catch percentage last year (~16th and 18th but possible there's a non qualifier or two above him skewing that slightly downward). Return yards allowed were second best in the league. Is there anywhere to find hang time stats? He seemed to struggle pretty bad on that in the opener, just wondering how that's trending. I buy he's relatively more effective at what the Pats value at punter. Im not sure he's an average punter right now. We'll see, early in the season.
Couple things on it:

  • On TB%, remember that we have to take into account starting field position, as it's harder to hit a touchback from deeper in your own territory. Allen's average punt last year came from the NE 36-yard line, the average for the rest of the league was their own 34-yard line. So I look at TB% from a team's own 41-yard line and forward, because if you hit a TB from behind your own 41, you're still netting 40 yards at a minimum.
  • Also on TB%, we do have to account for coverage. One of Allen's 5 touchbacks was caused by a gunner bobbling the ball into the end zone on what should have been an easy pin inside the 5-yard line, so this has to be stripped out of the data.
  • Some further talk on the need for booming kicks - Allen had just one punt from inside his own 10-yard line all season while the rest of the league averaged almost 3 such punts.
  • I have hang-time stats in my own charting, as I hand-charted every game last season, though I have not been doing so this year. In 2016, Allen averaged 4.28s compared to the league average of 4.41s, so he graded out slightly below average. However, in 2015, he was at 4.45s with the league average at 4.42s, and with the samples being relatively small, I tend to think the two season aggregate is pretty much where his true talent level lies. Average to slightly below.
  • Allen went through a terrible stretch Weeks 7-11 last year where he seemed to have the yips and had two shanks and two touchbacks. No idea what happened, but it was pretty much awful all-around from him and killed what had been a pretty good season until then.
  • Allen's Week 3 last year was one of the best games I've ever seen from him - 7 punts, 5 fair catches, no return yards, and every punt at or above the target distance. So you see this game and you're like "Fuck, that's the guy" and he just can't do it enough.
So overall, Allen is a pretty pedestrian punter, but he fits what NE needs, and he does so at a cost that works for them. Punter is not a position I want them spending money on, and young punters tend to have a pretty harsh adjustment period, so cheap consistency, even if it's just average, isn't a bad place to be.
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
Right, I dont have more granular stats readily available for touchbacks, but Id think he might be picking up more fair catch percentage based on field position and he didnt, for sure its a rough metric. It also looks like his hang time keeps getting worse, but I dont have a real good way to track it.
 

IdiotKicker

Member
SoSH Member
Nov 21, 2005
10,785
Somerville, MA
FC% is low, but his positive return percentage is really what I look at. 69% of Allen's punts last year went for 0 or negative return yards. The league average is 64%, so what this tells me is that either a) Allen is great at punting OB so as to not allow any return (he's not, Sam Koch is fucking phenomenal at this though), b) Allen generates a ton of hangtime (nope, not happening), or c) Allen generates enough hangtime for the distance travelled, and he has the best coverage unit in the business because BB can spend the savings from Allen's contract on tying up Slater and co. Plus those coverage guys can pay dividends as they grow into other roles on the team. But yeah, I think we're in the same ballpark on Allen's overall talent in that he's pretty meh overall, just some differences around the edges, but he fits well for what NE wants to do even if he's not perfect.
 

Hoodie Sleeves

Member
SoSH Member
Nov 24, 2015
1,204
This is purely anecdotal, but it certainly seems like the Pats have been much more conservative over the past 5 years or so in these situations once they pass the 50 yard yard line and before they reach a reasonable FG distance.

Back in the day, I recall the Pats going for it on 4th down in these situations as the default option. It may have been more that the Pats were winning but such a large margin in many of these games, but it seems like punting has become the default state versus the long FG or going for it.
Your anecdotal observances agree with the stats.

Football outsiders used to do an "Aggressiveness Index" that basically checked the rates teams went for it on 4th down (and a handful of other things) and compared them to league average - and BB used to be way above the rest of the league - he was bouncing back and forth between top 5 and top 10 last I had seen it (and the last one I can find is 2013). Some of it is him being less aggressive, and some of it is coaching in the league getting younger, and more cognizant to modern statistics.

The last couple of years he seems to have become way more conservative about the end of the first half - they used to be hyper aggressive - and picked up a lot of wins with that score at the end of the first, score at the beginning of the second.
 

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
Your anecdotal observances agree with the stats.

Football outsiders used to do an "Aggressiveness Index" that basically checked the rates teams went for it on 4th down (and a handful of other things) and compared them to league average - and BB used to be way above the rest of the league - he was bouncing back and forth between top 5 and top 10 last I had seen it (and the last one I can find is 2013). Some of it is him being less aggressive, and some of it is coaching in the league getting younger, and more cognizant to modern statistics.

The last couple of years he seems to have become way more conservative about the end of the first half - they used to be hyper aggressive - and picked up a lot of wins with that score at the end of the first, score at the beginning of the second.
He was 9th in 2014, too: http://www.footballoutsiders.com/stat-analysis/2015/aggressiveness-index-2014

My impression is that Belichick is neither a slave to convention nor to the stats. There are a lot of other factors that are going to go into these sorts of decisions: the score, whether he thinks the Patriots are better, how tired his defense / the oppoenent's defense is, how he feels about the OL / DL matchup, etc. The OL / DL matchup, I imagine, factored heavily into some of the decision-making Sunday. Fourth-and-two might as well have been fourth-and-seven because they had no chance of picking up two yards on a conventional running play on fourth down against that DL. I think Belichick sees these factors as more important than the math in making these kinds of decisions.
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
For sure. I think his weighting of those factors has become more conservative than optimal over the last two years.
 

bankshot1

Member
SoSH Member
Feb 12, 2003
24,661
where I was last at
I'm not sure if a 48-yarder qualifies, but I will never understand BB's decision in SB42 to by-pass a long but makeable FG, to extend the Pats lead to 7 points (10-3) and instead go for it on 4th and 13 from the Giants 31, with what looked like a sand-lot play.

Without doing any homework or calcualtions, I would guess the probablitiy of Ghost hitting a 48-yarder was much higher than that stupid fucking Hail Mary to Gaffney in the end-zone.

The loss of an additional 7 yards in field position seems an almost irrelevant risk.

One day, woluld love to hear Ernie Adams explain the reasoning.
 

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
For sure. I think his weighting of those factors has become more conservative than optimal over the last two years.
To be totally honest, I don't think there's a large enough sample to make any conclusions one way or the other. The Patriots have faced 34 4th-and-1 or -2 from their own 40 or closer since the start of 2015 and gone for it 19 times, which is slightly above average but not materially so. Moreover, many of those decisions were not really decisions. Of course they kicked the game-winning FG against the NYG and ARI. Of course they went for it down 7 against Seattle. Of course they went for it down 16 with 3:25 left against Buffalo last year. We're really talking about a very small sample of actual decisions.

FWIW, the Pats went for it 10 times of 17 with a lead, much more than the typical ~33%. This appears to be the highest rate in the NFL.

Similarly, aggressiveness in the 31-to-37 range is slightly above average (6 of 17 versus typical ~27%), but again in there we have a bunch of decisions-that-aren't-really-decisions (4th-and-25 with big lead vs Saints a couple weeks ago; down 7 late versus Jets in 2015.
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
I certainly wouldnt argue its a provable fact given the sample sizes.

Where are you getting these sort of 4th down splits?
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
What was aggressiveness from 2001-2014? Digging through the 4th down stuff, but cant find aggressiveness.
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
Ok, so aggressiveness is just percentage they went for it. Got it, didnt know if that was a metric of some sort. I'll have to find time to play around with it and see if I can find a metric for end of half aggressiveness.

I suspect a combination of cognitive biases on my end and more conservatism in specific situations (i.e. 4th and 1 as the Pats power running game has gotten worse, but the break even still argues strongly to go for it).

EDIT: OK, aggressiveness isnt at the playfinder, its an FO metric, got it.
 
Last edited:

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
Ok, so aggressiveness is just percentage they went for it. Got it, didnt know if that was a metric of some sort.
I think FO's criteria is a little more elaborate, like stripping out situations that are obvious based on being behind (which would probably make the Patriots look more aggressive, both pre- and post- 2014/2015).
 

Stitch01

Member
SoSH Member
Jul 15, 2005
18,155
Boston
One interesting split to run....2001-2009 (Indy 4th and 2) 2010-2017 (after 4th and 2)

Cherry picking to some degree, but if Im running this right I show going for it 4th 1 or 2 plus territory

'01-'09 86/129 (66.7%)
'10-17 43/94 (45.7%)
 

Hoodie Sleeves

Member
SoSH Member
Nov 24, 2015
1,204
I think FO's criteria is a little more elaborate, like stripping out situations that are obvious based on being behind (which would probably make the Patriots look more aggressive, both pre- and post- 2014/2015).
There's a bit of an explanation on what they do here:

http://www.footballoutsiders.com/stat-analysis/2013/aggressiveness-index-revisited

It seems that he dropped significantly in the ratings after 2010:
As a result, Belichick ranks a shocking 23rd out of 34 head coaches in Aggressiveness Index for 2012. This is a massive change from most of Belichick's career. Belichick has the fifth-highest career AI of any head coach with at least three full seasons between 1991 and 2012, and he ranked in the top six for AI every year between 2004 and 2010 before falling to 11th in 2011 (although he was sixth in 2011 in the older version of AI). 2012 was only the third season out of 18 when Belichick ranked in the bottom half of the league; the others were 1994 (0.76 AI, 21st) and 2003 (.84 AI, 23rd).

He doesn't seem like the kind of guy to ever react to the pressure of conventional wisdom, but is it possible that a few high-profile fourth-down failures have actually made Belichick more risk averse? That certainly seems to be the case with Mike Smith. Smith has a career AI of 1.31 but ranked dead last in 2012 with 0.62 AI. The Falcons only went for it once in 91 qualifying opportunities in 2012, and that opportunity itself barely qualifies. It was a handoff to Michael Turner on fourth-and-1, up 13 points on Philadelphia in Week 8 with only 20 seconds left in the game. They didn't convert.
 

Hoodie Sleeves

Member
SoSH Member
Nov 24, 2015
1,204
He was back to 9th in 2014 per link above. At the end of the day the sample sizes here are pretty tiny year-on-year.
Of course they're small, and there's going to be lots of noise.

But the pattern is pretty strong with BB - pretty much his entire career up through 2010, he ranked in the top handful of coaches in the league according to this metric. In 2011, he ranked in the middle, and then in 2012, he ranked at the bottom.

In 2014, he was back up in the middle (9), rankings wise, but that 1.13 was the same distance from 6th ranked Gruden (1.28), as it was from 17th ranked Ryan (.98), and was significantly below the career average he had entering 2010 (1.59)

Its also important to consider that the baseline for the index has gotten lower as the NFL as a whole has become less aggressive on 4th down, so the 1.60 for Caldwell at 3rd in 2014 was significantly less aggressive than the 1.59 that BB was holding for his career entering 2010.

pre2011: 1.59
2010: 1.75
2011: 1.08
2012: 0.81
2013:
2014: 1.13

I can't find data for 2013 (its probably just in their book). But that's a pretty large effect (and even more-so knowing the baseline has decreased - and BB becoming so much less aggressive has probably affected the baseline in a non-trivial way).
 

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
Of course they're small, and there's going to be lots of noise.
I feel like you are not keeping this in mind in the rest of your post.

But the pattern is pretty strong with BB - pretty much his entire career up through 2010, he ranked in the top handful of coaches in the league according to this metric. In 2011, he ranked in the middle, and then in 2012, he ranked at the bottom.
Not at the bottom. 23rd of 34. He's as close to 13th as he is to 34th, assuming AI is a linear scale. And again, small sample - it possible that one or two decisions is the difference between being 23rd and, say, 5th (in classic FO fashion, AI is a black box, so who knows, but based on the sample sizes a handful of decisions would throw it off). Belichick was 23rd in 2003 as well, FWIW.

To the extent there is a real effect here, it seems like beginning in 2009 the Patriots started treating 3rd-and-2 and 4th-and-2 as passing downs. They ran a little more than they passed on 3rd-and-2 from 2001 to 2008 but are about 2/3 passing since. They haven't run the ball on 4th-and-2 since 2009. But again, we are talking very small samples here.
 

Hoodie Sleeves

Member
SoSH Member
Nov 24, 2015
1,204
Not at the bottom. 23rd of 34. He's as close to 13th as he is to 34th, assuming AI is a linear scale..
The distribution isn't linear at all - its a big bell curve. Click on the links. The majority of coaches are bouncing around between 1.2 and .8. He was way out on one tail most of his career, and is now moving around on the other side of the fat part of the curve.

And yes, small samples are a problem - but we have a strong signal here (and 30 years of data on BB).

In most years, the highest AI is about 1.6, with an occasional Payton (or some bad coach that gets fired for being aggressive and failing) at about 2. BB averaged 1.59 up until 2011, and had several years above 1.9.
 

Super Nomario

Member
SoSH Member
Nov 5, 2000
14,012
Mansfield MA
The distribution isn't linear at all - its a big bell curve. Click on the links. The majority of coaches are bouncing around between 1.2 and .8. He was way out on one tail most of his career, and is now moving around on the other side of the fat part of the curve.
I mean linear in the sense that I don't know if 0.15 AI = 0.15 AI. Is the difference between 1.02 AI and 0.92 AI and 0.82 AI the same? I don't know. We don't have a lot of insight into how it is calculated.

And yes, small samples are a problem - but we have a strong signal here (and 30 years of data on BB).
We have a reasonable amount of data to conclude that Belichick is historically aggressive. We have a tiny sample to suggest he is now conservative.
 

Infield Infidel

teaching korea american
SoSH Member
Jul 15, 2005
11,463
Meeting Place, Canada
One interesting split to run....2001-2009 (Indy 4th and 2) 2010-2017 (after 4th and 2)

Cherry picking to some degree, but if Im running this right I show going for it 4th 1 or 2 plus territory

'01-'09 86/129 (66.7%)
'10-17 43/94 (45.7%)
Forgive me for asking, but does anyone remember what year they moved the opponent's starting position after failed FGs from the LOS to the FG spot? I think it was in the 90s but it may have been later.