Does any coach punt from the opponent's side more than Garrett?

[DCBoysfan]

What are you talking about? Failing on fourth down gives the opponent the ball, but so does punting. They didn't give the Panthers more opportunities. In fact, you could argue that they got more offensive drives than they would have if they'd punted, since the Panthers weren't running as much time off the clock on their own drives as they might have. The 49ers didn't score because they failed on their 4th down conversions, not because they attempted 4th down conversions. The strategy was correct: it just didn't work out this time. The punting strategy was an even more certain loser. And always remember, the 49ers stink. Most strategies are going to result in them losing. They should still pursue the strategies that give them the best possible chance of winning, though.

You're right about one thing: "Strategy like gets you fired, which is why it's not done." That is exactly correct. Coaches don't do it even though it's good strategy, because they will get punished when it doesn't work. Shanahan knows that, which is why he's backtracking now.

I'm talking about giving your opponent the ball at mid field, that's what I'm taking about, for some reason your crazy logic suggest going for it like you playing a video game because we want to loose "respectfully" instead of playing to win. I'm also taking about not getting blown out 23-3. The reason Kyle is backtracking is because he realize he was wrong.

As I already mentioned your test case was tried Sunday, he coached like you suggested and the result was a 23-3 BLOWOUT, I don't need to see anymore test cases, and if Kyle continues to coach like that he will have plenty time to play Madden and go for it till his hearts content.

 

[JD_KaPow]

I'm talking about giving your opponent the ball at mid field, that's what I'm taking about, for some reason your crazy logic suggest going for it like you playing a video game because we want to loose "respectfully" instead of playing to win. I'm also taking about not getting blown out 23-3. The reason Kyle is backtracking is because he realize he was wrong.

I suggest you go back and read what I wrote again. I said that teams SHOULDN'T play to lose "respectably." Punting is how you lose respectably. Going for it more often is how you play to win, but it comes with a greater chance of getting blown out.

As I already mentioned your test case was tried Sunday, he coached like you suggested and the result was a 23-3 BLOWOUT, I don't need to see anymore test cases, and if Kyle continues to coach like that he will have plenty time to play Madden and go for it till his hearts content.

You're still missing the point. There's no difference in the standings between a 23-3 blowout and, say, a 10-0 loss. I can say with great confidence that if the 49ers never went for it on 4th in that game, they would never have scored. How can I say that? Because they did in fact fail to score on every drive, except one where they went for it on 4th. Sure, if they'd punted, they may have given up less than 23 points. But they certainly would have lost, so who cares? I just can't fathom how you can look at what happened in that game and say, "oh yeah, if they'd punted all those times they might have won." On the other hand, if they'd converted a couple more of those 4th downs, it might have been a completely different game.

 

[DCBoysfan]

I suggest you go back and read what I wrote again. I said that teams SHOULDN'T play to lose "respectably." Punting is how you lose respectably. Going for it more often is how you play to win, but it comes with a greater chance of getting blown out.

You're still missing the point. There's no difference in the standings between a 23-3 blowout and, say, a 10-0 loss. I can say with great confidence that if the 49ers never went for it on 4th in that game, they would never have scored. How can I say that? Because they did in fact fail to score on every drive, except one where they went for it on 4th. Sure, if they'd punted, they may have given up less than 23 points. But they certainly would have lost, so who cares? I just can't fathom how you can look at what happened in that game and say, "oh yeah, if they'd punted all those times they might have won." On the other hand, if they'd converted a couple more of those 4th downs, it might have been a completely different game.

Like I said, they lost 23-3 because they continued to put their defense in bad positions. No need for us to go back and forth, as I have already mentioned, he coach like you suggested and lost and got blown out, PERIOD. The coach has admitted it was mistake, because it was a mistake.

Also you have no idea if they would have won or lost if they would have punted, but we do know what DID happen when they didn't, now don't we. Enjoy your nite.

 

[wick]

Situation is everything. When you are going against an offense that is struggling to get out of it's own way, the field position battle is paramount. Sometimes, all it takes, is a gift of a starting position to get an offense going and the momentum moving in their favor. Jason absolutely made the right decision against the Giants. As for other instances, I'd have to know more about the circumstances in-game to weigh in on it.

How do you say that about the first such decision, which occurred prior to the Giants offense taking a snap?

 

[joseephuss]

How do you say that about the first such decision, which occurred prior to the Giants offense taking a snap?

Dallas held the Giants offense in check the last time they played. Only gave up 10 points in that game and the one TD they gave up was to a guy who wasn't playing this past Sunday night. They wanted to force the Giants to drive the whole field instead of potentially giving them a short field. Dallas knew their opponent well.

 

[xwalker]

Not at all. The third down data isn't skewed and constitutes the vast majority of the data set. And imperfect data isn't "worthless". Imperfect data is better than no data, as long as it is carefully evaluated and applied.

We're not talking here about something that's in the margins. The discrepancy between how coaches approach 4th down and what the analysis says they should be doing on 4th down isn't close. There's a massive difference there such that no data skew would change the result. NFL teams should go for it on 4th down much more than they do, period. They would win more games if they do, period (unless of course everyone adopts the strategy at once). There are marginal cases you can argue one way or the other, but the overall story is very very clear.

And remember, teams have plenty of resources to analyze this in tremendous detail. There's no reason a team can't go into every game with charts similar to the one in that linked article, tailored for that opponent and various game situations. Sure, they'd be working from imperfect data and some assumptions, but they would undoubtedly make better decisions in the aggregate.

Nonsense.

I'm referring to the article in the link at the end of this post.

There is no way to make a chart that is accurate to that level of detail with available data.

His chart indicates teams should go for it even when a fail leaves the opponent with only 9 yards to score a TD.

I do agree that teams are too conservative, but trying to make a chart to quantify every distance to go at every spot on the field is basically impossible with any degree of accuracy considering the available data. Even with available data, all of the outlier stuff would need to be removed. If your using 3rd down data then many late game situations should be temoved. A defense that's getting blown out on the scoreboard won't put up the same resistance on 3rd down as when the score is close. They may even have backups in by then. Third down is not 4th down. Even a 10% variation in the available data versus reality would drastically change the chart.

I worked for a company that designed software used by many sports teams to analyze data. The data and the ability to analyze it is far beyond what a guy on the internet could come up with.

I didn't work on the software. I worked a project to process data with configurable hardware which is infinitely faster than software.

Most teams have looked at the data that is available on this issue and they still don't do anything close to what the chart in the article recommends.

Chip Kelly did't even do what the article recommends and he was considered to be the most aggressive coach on this topic.

https://mobile.nytimes.com/2014/09/...pponents-side-more-than-garrett.385187/page-2

 

[JD_KaPow]

Nonsense.

I'm referring to the article in the link at the end of this post.

There is no way to make a chart that is accurate to that level of detail with available data.

His chart indicates teams should go for it even when a fail leaves the opponent with only 9 yards to score a TD.

I do agree that teams are too conservative, but trying to make a chart to quantify every distance to go at every spot on the field is basically impossible with any degree of accuracy considering the available data. Even with available data, all of the outlier stuff would need to be removed. If your using 3rd down data then many late game situations should be temoved. A defense that's getting blown out on the scoreboard won't put up the same resistance on 3rd down as when the score is close. They may even have backups in by then. Third down is not 4th down. Even a 10% variation in the available data versus reality would drastically change the chart.

I worked for a company that designed software used by many sports teams to analyze data. The data and the ability to analyze it is far beyond what a guy on the internet could come up with.

I didn't work on the software. I worked a project to process data with configurable hardware which is infinitely faster than software.

Most teams have looked at the data that is available on this issue and they still don't do anything close to what the chart in the article recommends.

Chip Kelly did't even do what the article recommends and he was considered to be the most aggressive coach on this topic.

https://mobile.nytimes.com/2014/09/...pponents-side-more-than-garrett.385187/page-2

Not sure what the argument is at this point, or what you're calling "nonsense" about what I said. Of course the article is based on aggregate generic data. I'd agree that the table in the article is extremely generic and should have substantial regions labeled as "too close to call" rather than having a definitive answer for each position and yards-to-go. But as I said above, there's no reason a team can't put a chart together for every game that shows a recommendation for each case based on the best data available to them. But the issue here isn't at the margins. You think a team shouldn't go for it on 4th and 1 at their own 15? That's fine. (I think they should in most cases). But teams are punting, routinely, on 4th-and-2 at the opponent's 39, which is clearly nuts. That tells me very plainly that teams are not making these decisions rationally. Which you agree with: you say that you agree teams are too conservative.

All the concerns you have with relevance of data are pretty easy to deal with for a competent analyst, and nobody would suggest that the chart will be perfect. But making decisions informed by the best data you can gather is way better than making them based solely on your coach's gut feeling, which I'm pretty sure is how most teams still do this.

 

[xwalker]

Even this would not work I don't think.

Every year your offense has different personnel.

You could have a great QB for a couple of years. Then he retires and you have a scrub.

Or in the 3rd year you find a Zeke type RB that changes your equation.

You could have a rash of injuries to your o-line or receivers such that the HC may be more conservative.

I don't think there is a way of ultimately determining using math when you should or should not go on 4th down. The coach has to have a feel for his team and the circumstances they are in and make a final call.

Most Pro teams have the ability to analyze data with detailed criteria. They could analyze all data with a specific QB and RB if those players are veterans that would have been in enough of these situations to get a big enough sample size to be stastically relevant.

The concept gets really complicated. Failing on 4th down in the 4th quarter likely affects winning more than failing in the 1st quarter. The variables are near infinite.

I do think NFL teams could be more aggressive on 4th down but I don't think they should ever go by a static chart.

They could have a guy in the booth run an analysis on their specific situation and call down to the Head Coach with a recommendation; however, considering how often something as simple as the booth telling the HC to challenge or not challenge a call has failed, I wouldn"t hold my breath that something more complicated could be done in that situation just based on the human element.

 

[xwalker]

Not sure what the argument is at this point, or what you're calling "nonsense" about what I said. Of course the article is based on aggregate generic data. I'd agree that the table in the article is extremely generic and should have substantial regions labeled as "too close to call" rather than having a definitive answer for each position and yards-to-go. But as I said above, there's no reason a team can't put a chart together for every game that shows a recommendation for each case based on the best data available to them. But the issue here isn't at the margins. You think a team shouldn't go for it on 4th and 1 at their own 15? That's fine. (I think they should in most cases). But teams are punting, routinely, on 4th-and-2 at the opponent's 39, which is clearly nuts. That tells me very plainly that teams are not making these decisions rationally. Which you agree with: you say that you agree teams are too conservative.

All the concerns you have with relevance of data are pretty easy to deal with for a competent analyst, and nobody would suggest that the chart will be perfect. But making decisions informed by the best data you can gather is way better than making them based solely on your coach's gut feeling, which I'm pretty sure is how most teams still do this.

I think some coaches have a chart. It may be with the coaches in the booth but I think many coaches have a preconceived idea of what they want to do.

Punting from the opponent's 40 does seem odd to me in most situations on 4th and 2. There is definitely a big gap from what coaches do now and the chart. Every coach could be something like 20% more aggressive and still be far from the chart.

 

[FuzzyLumpkins]

You can't take credit for pinning them at the 6 and also for the chance of a penalty. You need to look at the expected field position gained by the punt.

4th and 3 from the opponents' 43. Success rate for that distance and that position on the field is around 55%. (It gets lower down near the opponents' end zone, because they have less field to defend).
If you go for it and succeed, your expected points scored is at least 2.66. (It may be better because you may gain more than 3 yards on the 4th down play).
If you go for it and fail, your expected points scored is -1.63. (Your opponent has the ball near midfield).
If you punt, you expect them to get the ball somewhere around the 12-14 yard line. From there, your expected points scored is right around 0. (It goes positive closer to the end zone, because of safeties and because you're likely to have good field position on your NEXT drive).

Based on this, going for it gives you 0.73 expected points scored, and punting gives you 0 expected points scored. The choice is clear.

Okay, but the Giants have a very good defense, so maybe your chances aren't 55% (although I'll remind you that we were 53% on 3rd-down conversions in this game, and several of those were much longer than 3 yards). Well, it turns out the breakeven point for this decision is at a 36% success rate, and you can't convince me that we didn't have a better than 36% chance of converting that down.

Okay, but we have a super-duper kicker who we know can pin them at the 6. Now the breakeven is 39%. I'm still going for it.

Okay, but the Giants have a really bad offense, so pinning them is even better for us. Well, yeah, but giving them the ball at their 43 is also not as bad for us, so that washes out somewhat. It's certainly not a big enough effect to change this decision, which really isn't close.

You are completely ignoring the other side of the coin and the fact that the game continues.

The Giants were struggling to get first downs. If you pin them and force them to punt then what are you chances to score?

Conversely what are their chances to score at the 45 as opposed to the 10?

Keeping them from scoring is just as important as scoring yourself. Given the physical nature of football emotion and thus momentum is real.

We never gave them a chance to score.

 

[reddyuta]

did anybody actually answer the OPs question? If i was to guess then the answer would be "no".

 

[Nightman]

Not sure what the argument is at this point, or what you're calling "nonsense" about what I said. Of course the article is based on aggregate generic data. I'd agree that the table in the article is extremely generic and should have substantial regions labeled as "too close to call" rather than having a definitive answer for each position and yards-to-go. But as I said above, there's no reason a team can't put a chart together for every game that shows a recommendation for each case based on the best data available to them. But the issue here isn't at the margins. You think a team shouldn't go for it on 4th and 1 at their own 15? That's fine. (I think they should in most cases). But teams are punting, routinely, on 4th-and-2 at the opponent's 39, which is clearly nuts. That tells me very plainly that teams are not making these decisions rationally. Which you agree with: you say that you agree teams are too conservative.

All the concerns you have with relevance of data are pretty easy to deal with for a competent analyst, and nobody would suggest that the chart will be perfect. But making decisions informed by the best data you can gather is way better than making them based solely on your coach's gut feeling, which I'm pretty sure is how most teams still do this.

It is very frustrating because everyone agrees with the data....... NFL coaches are way too conservative on 4th down

And we also agree that coaches ignore the data because of peer pressure

The real, trying to win book says to go for it but the fake media guide/book says to punt( but only when it fails)

I tell everyone this.......how do you feel when the other team has it at our 37 on fourth and 1 and lines up to go it.........you are nervous and anxious..... another score would be crushing..... but instead they try to get us to jump, take the penalty and punt........ we all exhale and say hell yeah we are getting the ball back......we held them........

NO ONE roots for them to go for it because we know the odds are in their favor..... the field position is so much less important than getting possession

 

[jrumann59]

It is very frustrating because everyone agrees with the data....... NFL coaches are way too conservative on 4th down

And we also agree that coaches ignore the data because of peer pressure

The real, trying to win book says to go for it but the fake media guide/book says to punt( but only when it fails)

I tell everyone this.......how do you feel when the other team has it at our 37 on fourth and 1 and lines up to go it.........you are nervous and anxious..... another score would be crushing..... but instead they try to get us to jump, take the penalty and punt........ we all exhale and say hell yeah we are getting the ball back......we held them........

NO ONE roots for them to go for it because we know the odds are in their favor..... the field position is so much less important than getting possession

That is game by game or series by series. With NYG OL issues and lack of a passing game. I would have punted if we were at Dan Bailey's career long. Besides a turnover on downs they could have fumbled, thrown an interception, taken a long sack or long tackle for a loss. The possible negative plays outweighed the possible positive ones which was a1st donw or a TD.

 

[jday]

How do you say that about the first such decision, which occurred prior to the Giants offense taking a snap?

Just a different situation where punting is better.

I think a better way of explaining is to say the only time you go for it is when either the matchup dictates early aggression or a situation such as trying to dig your way out of a two score deficit dictates that as necessity. Otherwise, the field position battle is far more important.

Against a team such as the Giants, with a defensive line like theirs, it is rarely a good idea.

 

[Jake]

It sure feels like Garrett punts a ton from the opposition's side of the field, and Sunday night was certainly no exception. We punted four times from Giants territory out of nine total drives. Those fourth downs:

• 4th and 3 from the Giants 43
• 4th and 2 from the Giants 44
• 4th and 7 from the Giants 42
• 4th and 20 from the Giants 42

The last drive is definitely a punting situation. I'm not sure what the numbers say about the 4th and 7, but I'd go for that with our offense. The first two should be automatic go for it situations unless there is a compelling game situation, which there was not at those points.

 

[JD_KaPow]

You are completely ignoring the other side of the coin and the fact that the game continues.

The Giants were struggling to get first downs. If you pin them and force them to punt then what are you chances to score?

Conversely what are their chances to score at the 45 as opposed to the 10?

Everything you just asked about is discussed in my post and explicitly included in my calculations you quoted. That's all taken into account.

Keeping them from scoring is just as important as scoring yourself. Given the physical nature of football emotion and thus momentum is real.

We never gave them a chance to score.

You know what the very best way to keep them from scoring is? Don't just give them the ball by punting, when you have a great chance to keep the ball away from them by getting a first down.

 

[JD_KaPow]

Just a different situation where punting is better.

I think a better way of explaining is to say the only time you go for it is when either the matchup dictates early aggression or a situation such as trying to dig your way out of a two score deficit dictates that as necessity. Otherwise, the field position battle is far more important.

Against a team such as the Giants, with a defensive line like theirs, it is rarely a good idea.

You're not explaining anything. Your'e making assertions without any support or evidence. The things you're asserting are simply wrong, and there's a ton of data out there to support that.

It's interesting. I watched this process happen 20 years ago in baseball. New stats and analytics were coming along and it became obvious to those looking at the data that teams were making poor strategic decisions about things like stolen bases and hit-and-run plays (not to mention player evaluation). There was massive resistance from fans and the press and the teams themselves to incorporating these ideas. And now, finally, years later, they've completely taken over the game. And now the teams are leading the way on data-driven innovation, with all the shifting being the most obvious example. Football is way ahead in some areas, but in others, like 4th down calls, it's the same pattern. I'll be very surprised if, 10 years from now, 4th down decision-making hasn't drastically changed.

 

[JD_KaPow]

did anybody actually answer the OPs question? If i was to guess then the answer would be "no".

Several people did, with different looks at the data, which shows that Garrett is at least as conservative as other coaches and probably significantly more so.

 

[AdamJT13]

did anybody actually answer the OPs question? If i was to guess then the answer would be "no".

I did earlier. Yes, there are other coaches/teams that punt from the opponent's side more than Garrett -- but not many (we're seventh in punt percentage between midfield and the opponent's 30 since 2011). However, only two teams have gone for it fewer times in those situations, so that adds to the perception that we punt more than other teams do.

 

[AdamJT13]

I think a better way of explaining is to say the only time you go for it is when either the matchup dictates early aggression or a situation such as trying to dig your way out of a two score deficit dictates that as necessity. Otherwise, the field position battle is far more important.

When is field position ever more important than points?

Would you take a knee on fourth-and-1 from the opponent's 9-yard line to guarantee they start at their own 10? Of course not -- because points are far more important than field position.