You're down by 15 with 7:00 minutes left in the game (1 Viewer)

[rick6668]

But it's a one possession game. If you go for 2 the first time and fail, it's a 2 possession game. That's HUGE.

No it isn't. If you read through the entire thread, you'll see that there are subjective reasons why some could argue in favor of kicking the PAT first. I still think they're wrong, but those are more subtle and debatable points.

But the post VaTerp made is objectively, demonstratively incorrect. It's exactly the kind of logic failure that has been exposed countless times already, yet new people keep showing up and making this same flawed argument.

How is it not a one possession game? How is it objectively and demonstrativelty incorrect? If you go for 2 and miss it the first time, you are down 9. You MUST have two possessions to have a chance to extend the game with offensive scores. If you are down 8, you only need 1 possession to have a chance to extend the game with an offensive score. This is correct, objectively and demonstratively.

If you go for 2 down 15 and don't get it then you have put yourself in a situation where you MUST (w/ the exception of a defensive score) have 2 offensive possesions to extend the game. So if you only get 1 more offensive possession the rest of the game you are giving your team no chance to win.

You can disagree all you want and talk about a false concept of delayed knowledge but as a coach I will always minimize the number of possessions needed to extend a game AT THAT POINT in the game.

Whenever you miss the conversion you've added one more score. It doesn't matter if you miss it at the 7 minute mark or at the 2 minute mark.Wouldn't you rather know you need that additional score earlier to more effectively manage the time you have remaining? If you are driving for the potential tie anytime near the end of the game, any coach worth their salt will be trying to tie it with as little time remaining as possible - to prevent the other team from having the opportunity to drive for the win before overtime.

In doing so, however, they are operating counter to the strategy they need to use if they miss the conversion - which would be to conserve time.

You see, the strategy for the trailing team is the EXACT OPPOSITE if they miss the conversion or make the conversion. In going for it early, they avoid the catastrophic possibility of actually working against themselves for up to the seven minutes remaining in the game.

Exactly. Watching the game monday night with the Texans driving down by 8 with just under a minute left, I heard one of the announcers make this comment after a pass completion in bounds, "Schaub is getting everone to the line for a spike, no need to hurry, there's still plenty of time. Make sure everyone is set." In fact, they do not know if there is plenty of time because even if they score the TD they do not know yet whether they will get the 2 PTs. If they get it, they want as little left on the clock as possible, but if they don't, they need as much time as possible. By taking their time and running the clock down, if they miss the 2pt, they are surely finished. If they knew (kicking it before) then they'd know whether they needed to use up or to conserve the clock.

 

[footballnerd]

over/under that this will be settled by OTAs?

 

[VaTerp]

How is it not a one possession game? How is it objectively and demonstrativelty incorrect?

If you go for 2 and miss it the first time, you are down 9. You MUST have two possessions to have a chance to extend the game with offensive scores. If you are down 8, you only need 1 possession to have a chance to extend the game with an offensive score. This is correct, objectively and demonstratively.

If you go for 2 down 15 and don't get it then you have put yourself in a situation where you MUST (w/ the exception of a defensive score) have 2 offensive possesions to extend the game. So if you only get 1 more offensive possession the rest of the game you are giving your team no chance to win.

You can disagree all you want and talk about a false concept of delayed knowledge but as a coach I will always minimize the number of possessions needed to extend a game AT THAT POINT in the game.

There are literally dozens of posts in this thread that explain in great detail what's wrong with this post. It might benefit you to read through them all. In short, the number of possessions needed is the same whether you go for 2 on the first TD or on the second TD. The only real question is when you'd rather find out what that number is - with 7:00 left, or at some point later in the game.

No, it really wouldn't benefit me and I'm not going to go back through 14 pages of this thread. And if you don't care to explain in further detail again that's fine by me. I think I get the crux of the argument and completely disagree. There is nothing in my post that is incorrect or a "logical failure."The number of possessions is not the same if you are down 8 or 9. Down 8 you have a CHANCE to tie with one possession. Down 9 you have NO CHANCE to tie with one possession. PERIOD. That's really all I consider as a coach because I may only get 1 more offensive possession and I am going to take my chances on going for 2 then in case I do in fact, only get 1 more offensive possession.

So despite this logically flawed concept of "delayed knowledge" I am going to, to the greatest extent possible, minizmize the number of possessions needed to extend the game AT THAT POINT in the game.

So... just to make sure we're on the same page:You're saying if one goes for the conversion early, they're automatically down 9. Whereas, if they go for the conversion after the 2nd TD, they'll make it.

Riiiight...

Not saying that at all and a logical reading of my post does not lend itself to that conclusion.Like CalBear said the chances of converting the 2 pt conversion is the same at 7:00 as it is at any other point in the game. As a coach, I am not going to take the RISK of being down 2 possessions when I can keep it a one possession game. I don't know what will happen in the next 7 minutes of the game.

As I said before, it doenst matter if I'm down 7, 8, or 9 in the 4th quarter my strategy is to score as quickly as possible and get stops as quickly as possible. Ideally, I will get two stops and two scores and win the game. But there is the very real possibility that I will only get one more offensive possession and one more chance to tie the game. I am going to make the decision that gives my team a chance to tie the game if in fact I do only get one more shot on offense. It's really that simple.

I don't buy into this concept of delayed knowledge at all because it's not as if I am going to just play for one score down 8, which seems to be a necessary part of the argument on the other side. Again, I am trying to score as quickly as possible and get stops as quickly as possible. But if I do end up with only one more possession, I want to know that I will have a shot at a tie. I am not going to make a decison that greatly increases the chances that we MUST have an extra possession to win the game.

If it so happens to come down to the end of the game and we need 2 to get it and we don't get convert then so be it. We had our shot and didnt convert. But going for 2 early greatly increases the chances that you MUST have an extra possession and that's not a chance that I, or the vast majority of coaches, want to or shoud take. Maximize the opportunity to make it a one possession game AT THAT POINT in the game and see how the next 7:00 play out.

 

[Ignoratio Elenchi]

But going for 2 early greatly increases the chances that you MUST have an extra possession

No it doesn't.

 

[humpback]

You should read the thread. It is mathematically proven that it is correct to go for the 2pt conversion to pull within 7. If only one coach in the whole NFL would learn a little math, they could have a big advantage over the rest of the league.

Basically, there are a couple scenarios that can play out. One gives you an extra chance that the other does not.

If you kick the XP after the first TD, the following scenarios are left:

- Score one more TD, make the 2-pointer, tie game

- Score one more TD, miss the 2-pointer, you lose

If you go for two after the first TD

- Make the 2, score one more TD, tie game

- Miss the 2, score twice more, win the game

- Miss the 2, score only one TD, you lose

The bolded scenario is the only one that's different. The others are just the same thing, but in a different order. So going for 2 first allots you one more opportunity to win that kicking the XP does not. It is likely a low percentage shot, but isn't 5% better than 0%?

Mathematically (all emotion/momentum taken out) there is no argument to kick the XP first. Clearly, math favors going for two first. So, the question becomes whether you think the probability that you're more likely to convert the 2-pointer after the second TD than after the first TD due to a little bit more momentum is greater than the chance that you score twice more after the first TD.

Personally, I don't see how that somewhat meager change in momentum is going to have a huge affect on whether or not you convert the 2 pointer.

I really don't want to rehash the last 15 or whatever pages, but do you guys really think this is the case? No, it has not been proven mathematically that it's correct to go for 2 first, not even close. There are so many possible scenario's left with 7 minutes still to play that you realistically can't prove it (or disprove it for that matter). Another point- do you really think that the math is so clear, yet not one team has ever followed the "mathematically correct" approach? Besides just the HC, NFL teams literally have staff members dedicated to running the numbers. Don't you think if the numbers said clear as day that the "correct" choice was to go for 2 first that at least one team would have done so? What is the explanation for teams making the "absolutely incorrect" choice all 66 times they've been faced with it?

 

[IvanKaramazov]

c'mon IE, everybody know that if you go for two then you automatically are going to miss and make it a two possession game, whereas, if you kick, you can take the two point converison as granted and deem it a one possession game. 1<2. QED.

 

[IvanKaramazov]

over/under that this will be settled by OTAs?

This topic was settled a long time ago. It's just that one side refuses to acknowledge it. This is worse than walking people through the Monty Hall paradox.

 

[VaTerp]

You should read the thread. It is mathematically proven that it is correct to go for the 2pt conversion to pull within 7. If only one coach in the whole NFL would learn a little math, they could have a big advantage over the rest of the league.

Basically, there are a couple scenarios that can play out. One gives you an extra chance that the other does not.

If you kick the XP after the first TD, the following scenarios are left:

- Score one more TD, make the 2-pointer, tie game

- Score one more TD, miss the 2-pointer, you lose

If you go for two after the first TD

- Make the 2, score one more TD, tie game

- Miss the 2, score twice more, win the game

- Miss the 2, score only one TD, you lose

The bolded scenario is the only one that's different. The others are just the same thing, but in a different order. So going for 2 first allots you one more opportunity to win that kicking the XP does not. It is likely a low percentage shot, but isn't 5% better than 0%?

Mathematically (all emotion/momentum taken out) there is no argument to kick the XP first. Clearly, math favors going for two first. So, the question becomes whether you think the probability that you're more likely to convert the 2-pointer after the second TD than after the first TD due to a little bit more momentum is greater than the chance that you score twice more after the first TD.

Personally, I don't see how that somewhat meager change in momentum is going to have a huge affect on whether or not you convert the 2 pointer.

I really don't want to rehash the last 15 or whatever pages, but do you guys really think this is the case? No, it has not been proven mathematically that it's correct to go for 2 first, not even close. There are so many possible scenario's left with 7 minutes still to play that you realistically can't prove it (or disprove it for that matter). Another point- do you really think that the math is so clear, yet not one team has ever followed the "mathematically correct" approach? Besides just the HC, NFL teams literally have staff members dedicated to running the numbers. Don't you think if the numbers said clear as day that the "correct" choice was to go for 2 first that at least one team would have done so? What is the explanation for teams making the "absolutely incorrect" choice all 66 times they've been faced with it?

Exactly. Those advocating going for 2 first are so convinced that they are right. So convinced that math is on their side, which it's not for the reason stated above. So convinced that they are smarter than all of the NFL coaches in this regard.

I would love for one of you to be a coach, at any level, and makes these decisions. Then report back here how it plays out.

 

[Ignoratio Elenchi]

This is worse than walking people through the Monty Hall paradox.

Someone should bump an old Monty Hall thread or something and invite a few of these guys over.

 

[Jersey35]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.

How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.

Just because it isn't being done, doesn't mean it isn't the right thing to do.

Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

 

[TheChairman]

How much time would have to be left on the clock for it to be ok to kick the XP when down by 9?

 

[IvanKaramazov]

This is worse than walking people through the Monty Hall paradox.

Someone should bump an old Monty Hall thread or something and invite a few of these guys over.

:roflrobably there's nothing more that can be said about this that hasn't already been said a bunch of times already earlier in the thread.

 

[CalBear]

How much time would have to be left on the clock for it to be ok to kick the XP when down by 9?

I don't have a specific number in mind, but my answer is, "enough time that both teams will be playing something close to their base offense."

 

[shader]

How did I miss this thread?

And how are there so many people who say "don't go for 2"?????????????

I go for two IMMEDIATELY and don't think twice about it. Why? Because there's a very real possibility that I'm going to fail to gain 2 points. I'd rather know that with 7 minutes to go than with 2 or 3 minutes to go, so that I can plan accordingly.

How can it even be argued otherwise, I have no idea.

 

[rick6668]

Found this in another forum.

Link

 

[Maurile Tremblay]

But it's a one possession game. If you go for 2 the first time and fail, it's a 2 possession game. That's HUGE.

No it isn't. If you read through the entire thread, you'll see that there are subjective reasons why some could argue in favor of kicking the PAT first. I still think they're wrong, but those are more subtle and debatable points.

But the post VaTerp made is objectively, demonstratively incorrect. It's exactly the kind of logic failure that has been exposed countless times already, yet new people keep showing up and making this same flawed argument.

How is it not a one possession game? How is it objectively and demonstrativelty incorrect?

I've put the answer in red.If you're going to assume that you fail when you go for two after the first TD, to be fair you must also assume that you'll fail when you go for two after the second TD.

Since you're going to fail when you go for two after the scond TD, it's not a one-possession game.

(If you don't assume that you'll fail when you go for two after the second TD, then you're not allowed to assume that you'll fail when you go for two after the first TD. The order doesn't change the number of possessions that are needed. Only converting vs. failing changes the number of possessions needed. It is therefore objectively false that you'll need more possessions if you go for two first than if you go for one first.)

 

[humpback]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.Just because it isn't being done, doesn't mean it isn't the right thing to do.Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

I won't bother arguing that point, and I'm not saying that obviously the "correct" decision is to kick it 1st because every coach has done so. However, I can't buy the argument that it's mathematically proven that the correct choice is to go for 2 first. Head coaches may not be math guys, but NFL teams absolutely do have math guys in their organization, and I guarantee you they've run a lot more numbers than we have. How do you think they come up with their play call charts? They don't just throw things at a dart board, they run numbers. If they were so clearly favoring go for 2 first in this scenario, then that's what their charts would say to do. Sure, they can deviate from their charts, but not 66 out of 66 times. I'm pretty sure they have this scenario on their chart, and I highly doubt it says go for 2 first.I assume you're saying that a HC is more likely to keep his job by "going by the book", correct? Then wouldn't you be saying that the book says to kick it first? How could that be the case if the mathematically correct thing to do is clearly go for 2 first? It doesn't add up, unless you think every NFL organization (not just HC) has it wrong, but a couple of guys on the internet have it right.

 

[TheChairman]

Found this in another forum.

Link

While funny, and I'm a big believer in go for 2 now, it also makes the same crappy assumption that the losing team will only score the 2nd TD with no time remaining if they kick the XP first.

 

[Maurile Tremblay]

Found this in another forum.

Link

I was just going to suggest that continuing the discussion in English prose is probably a waste of time. Somebody needs to build a model to demonstrate some simulated results.That flow chart is a step in the right direction, although it doesn't attempt to quantify any of the probabilities. (Actually, that may be a benefit because nobody will trust anybody else's estimated probabilities. By leaving the estimates up to each individual reader, people can see for themselves — based on their own probability estimates! — why going for two first is the better strategy. That is, if people are willing to go through the exercise of doing the work . . .)

 

[Jersey35]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.Just because it isn't being done, doesn't mean it isn't the right thing to do.Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

I won't bother arguing that point, and I'm not saying that obviously the "correct" decision is to kick it 1st because every coach has done so. However, I can't buy the argument that it's mathematically proven that the correct choice is to go for 2 first. Head coaches may not be math guys, but NFL teams absolutely do have math guys in their organization, and I guarantee you they've run a lot more numbers than we have. How do you think they come up with their play call charts? They don't just throw things at a dart board, they run numbers. If they were so clearly favoring go for 2 first in this scenario, then that's what their charts would say to do. Sure, they can deviate from their charts, but not 66 out of 66 times. I'm pretty sure they have this scenario on their chart, and I highly doubt it says go for 2 first.I assume you're saying that a HC is more likely to keep his job by "going by the book", correct? Then wouldn't you be saying that the book says to kick it first? How could that be the case if the mathematically correct thing to do is clearly go for 2 first? It doesn't add up, unless you think every NFL organization (not just HC) has it wrong, but a couple of guys on the internet have it right.

I'll level with you here. I'm not really interested in debating this side of the argument. I just wouldn't assume that NFL teams know better. They don't hire 'challenge guys' to scrutinize replays and radio the coach to tell him to throw the flag or not - which seems like the moat obvious thing in the world. Therefore, to me its just as likely they don't have numbers guys telling the coach what to do with 7 minutes remaining in the game. I betcha most coaches don't want numbers guys telling them what to do.The word "genius" is thrown about far too often with coaches. "smarter than the others" would be more accurate.

 

[cvnpoka]

LOL at the all the condescending tools in this thread that are convinced they are 100% correct but no one that has ever been paid to make these decisions agrees with them.

oklahoma went for 2 on the first td earlier this year. they missed. and were lambasted by the media the next day.http://espn.go.com/ncf/boxscore?gameId=302960142

 

[VaTerp]

But it's a one possession game. If you go for 2 the first time and fail, it's a 2 possession game. That's HUGE.

No it isn't. If you read through the entire thread, you'll see that there are subjective reasons why some could argue in favor of kicking the PAT first. I still think they're wrong, but those are more subtle and debatable points.

But the post VaTerp made is objectively, demonstratively incorrect. It's exactly the kind of logic failure that has been exposed countless times already, yet new people keep showing up and making this same flawed argument.

How is it not a one possession game? How is it objectively and demonstrativelty incorrect?

I've put the answer in red.If you're going to assume that you fail when you go for two after the first TD, to be fair you must also assume that you'll fail when you go for two after the second TD.

Since you're going to fail when you go for two after the scond TD, it's not a one-possession game.

(If you don't assume that you'll fail when you go for two after the second TD, then you're not allowed to assume that you'll fail when you go for two after the first TD. The order doesn't change the number of possessions that are needed. Only converting vs. failing changes the number of possessions needed. It is therefore objectively false that you'll need more possessions if you go for two first than if you go for one first.)

No, that is not the answer. People keep suggesting that I'm assuming that if you go for 2 the first time you will FAIL. That's false.What I have said repeatedly is that by kicking the xtra point you are reducing risk and giving your team a CHANCE to tie the game with only one possession. Of course, if you convert the 2 pt attempt the first time then great. But given that there is a less than 50% success rate I am going to make the higher percentage decision to minimize the number of possessions needed to give my team a CHANCE to tie the game AT THAT POINT in the game.

 

[ROBOPUNTER]

How do so many people here have this wrong?

 

[Maurile Tremblay]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.Just because it isn't being done, doesn't mean it isn't the right thing to do.Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

The guy I'd expect to get it right is Belichick. (With maybe Jim Schwartz second.)Does anyone know whether Belichick has faced this decision before, and what he did?

 

[Ignoratio Elenchi]

I am going to make the higher percentage decision to minimize the number of possessions needed

Nope.

 

[CalBear]

Head coaches may not be math guys, but NFL teams absolutely do have math guys in their organization, and I guarantee you they've run a lot more numbers than we have. How do you think they come up with their play call charts? They don't just throw things at a dart board, they run numbers. If they were so clearly favoring go for 2 first in this scenario, then that's what their charts would say to do. Sure, they can deviate from their charts, but not 66 out of 66 times. I'm pretty sure they have this scenario on their chart, and I highly doubt it says go for 2 first.

I assume you're saying that a HC is more likely to keep his job by "going by the book", correct? Then wouldn't you be saying that the book says to kick it first? How could that be the case if the mathematically correct thing to do is clearly go for 2 first?

Because "the book" is demonstrably wrong in a lot of different situations. Baseball is getting smarter about this, but football's "book" still is based almost entirely on anecdote and existing practice and not on analysis. (See Romer's study of fourth downs).Here's one reason to believe the book is wrong: in 66 attempts, it has never worked.

 

[Jersey35]

No, that is not the answer. People keep suggesting that I'm assuming that if you go for 2 the first time you will FAIL. That's false.What I have said repeatedly is that by kicking the xtra point you are reducing risk and giving your team a CHANCE to tie the game with only one possession. Of course, if you convert the 2 pt attempt the first time then great. But given that there is a less than 50% success rate I am going to make the higher percentage decision to minimize the number of possessions needed to give my team a CHANCE to tie the game AT THAT POINT in the game.

So you'd prefer to miss the conversion with less time remaining?

 

[Maurile Tremblay]

What I have said repeatedly is that by kicking the xtra point you are reducing risk and giving your team a CHANCE to tie the game with only one possession. Of course, if you convert the 2 pt attempt the first time then great. But given that there is a less than 50% success rate I am going to make the higher percentage decision to minimize the number of possessions needed to give my team a CHANCE to tie the game AT THAT POINT in the game.

If you go for two after the first TD and make it, it's a one-possession game and you know it after the first TD.If you go for two after the first TD and miss it, it's a two-possession game and you know it after the first TD.

If you delay going for two until the second TD and you're going to make it, it's a one-possession game but you don't know it yet after the first TD.

If you delay going for two until the second TD and you're going to miss it, it's a two-possession game but you don't know it yet after the first TD.

The timing of when to go for two doesn't affect how many possessions you'll need. (That's decided only by whether you end up making or missing it whenever you do go for it.)

The timing of when to go for two does affect when you'll find out how many possessions you'll need. The sooner you go for two, the sooner you'll get your answer.

 

[Ramblin Wreck]

Whole bunch of you assuming that the leading team doesn't score again.

 

[cvnpoka]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.Just because it isn't being done, doesn't mean it isn't the right thing to do.Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

The guy I'd expect to get it right is Belichick. (With maybe Jim Schwartz second.)Does anyone know whether Belichick has faced this decision before, and what he did?

i think haley would be another. hes been more cavalier than even bb on 4th and short this year.

 

[Maurile Tremblay]

Whole bunch of you assuming that the leading team doesn't score again.

If the leading team scores again, it doesn't really matter what you do. The decision is unimportant in that case, and not really worth discussing.

 

[DBIsports]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.Just because it isn't being done, doesn't mean it isn't the right thing to do.Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

The guy I'd expect to get it right is Belichick. (With maybe Jim Schwartz second.)Does anyone know whether Belichick has faced this decision before, and what he did?

Why Belichick? Are you talking about Belichick before he became a genius with a bad team like the Cleveland Browns or when he suddenly became ultra-intelligent with a good team like the Patriots?

 

[Ignoratio Elenchi]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.Just because it isn't being done, doesn't mean it isn't the right thing to do.Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

The guy I'd expect to get it right is Belichick. (With maybe Jim Schwartz second.)Does anyone know whether Belichick has faced this decision before, and what he did?

Why Belichick? Are you talking about Belichick before he became a genius with a bad team like the Cleveland Browns or when he suddenly became ultra-intelligent with a good team like the Patriots?

Belichick has shown a willingness in the past to opt for the correct play rather than the "by-the-book" play (see 4th and 2 vs the Colts).

 

[Maurile Tremblay]

Why Belichick?

His degree is in economics (like Schwartz), and he does pay attention to math stuff. For example, Belichick read the Roemer paper about 4th downs as soon as it came out. His biographer (I forget which one; I've read two biographies of him) suggested that he's probably the only head coach in the NFL to have read it.

Are you talking about Belichick before he became a genius with a bad team like the Cleveland Browns or when he suddenly became ultra-intelligent with a good team like the Patriots?

The more recent the better since he's constantly learning.

 

[Ignoratio Elenchi]

Whole bunch of you assuming that the leading team doesn't score again.

Whole bunch of us are assuming that the trailing team does score again, too.

 

[DBIsports]

Why Belichick?

His degree is in economics (like Schwartz), and he does pay attention to math stuff. For example, Belichick read the Roemer paper about 4th downs as soon as it came out. His biographer (I forget which one; I've read two biographies of him) suggested that he's probably the only head coach in the NFL to have read it.

Are you talking about Belichick before he became a genius with a bad team like the Cleveland Browns or when he suddenly became ultra-intelligent with a good team like the Patriots?

The more recent the better since he's constantly learning.

I had no idea about the economics and or the Roemer paper...I will go somewhere and shut up now, carry on.

 

[Jersey35]

Pretty sure the coaching profession self-selects individuals for NOT being math guys. It takes huge shifts in group thinking to effect a change in paradigm.How many coaches are ex players? How many current or former players would you bet on in a mensa test? Or a spelling bee for that matter.Just because it isn't being done, doesn't mean it isn't the right thing to do.Heck, I guarantee you if given a choice between "put yourself in the best position to win" or "keep your job another week" nearly all coaches would choose to simply keep their job.

The guy I'd expect to get it right is Belichick. (With maybe Jim Schwartz second.)Does anyone know whether Belichick has faced this decision before, and what he did?

THANK YOU.Belichek was about the only one that immediately came to mind that has an understanding of the math involved here, and with the job security to buck the norm.

 

[humpback]

Head coaches may not be math guys, but NFL teams absolutely do have math guys in their organization, and I guarantee you they've run a lot more numbers than we have. How do you think they come up with their play call charts? They don't just throw things at a dart board, they run numbers. If they were so clearly favoring go for 2 first in this scenario, then that's what their charts would say to do. Sure, they can deviate from their charts, but not 66 out of 66 times. I'm pretty sure they have this scenario on their chart, and I highly doubt it says go for 2 first.

I assume you're saying that a HC is more likely to keep his job by "going by the book", correct? Then wouldn't you be saying that the book says to kick it first? How could that be the case if the mathematically correct thing to do is clearly go for 2 first?

Because "the book" is demonstrably wrong in a lot of different situations. Baseball is getting smarter about this, but football's "book" still is based almost entirely on anecdote and existing practice and not on analysis. (See Romer's study of fourth downs).Here's one reason to believe the book is wrong: in 66 attempts, it has never worked.

Not important, but it has actually "worked". The percentage is still ridiculously low.I guess if you're going to go with you're right and everyone in the NFL is wrong, then we'll just agree to disagree.

I'd still be interested in seeing some actual evidence that it's been "mathematically proven to be correct" to go for 2 first, but I don't think it can be done.

 

[Ramblin Wreck]

Whole bunch of you assuming that the leading team doesn't score again.

If the leading team scores again, it doesn't really matter what you do. The decision is unimportant in that case, and not really worth discussing.

Not necessarily. Say you go for two and miss and are down 9. Leading team runs the kickoff back and goes up 16. Now you have to score twice and make both two point conversions. What are the math odds of making both two point plays instead of just making one to tie the game at the last opportunity to do so?Say you go for two and miss and are down 9. Leading team kicks a field goal to go up 12. Now you are forced to score two touchdowns instead of one touchdown (with 2 point conversion) and a field goal as an option to force overtime. Is it more likely to make a two point conversion and convert a field goal or more likely to score two touchdowns?Either scenario is unlikely to occur but it's just as likely as those arguing that being down 9 instead of 8 is better because you know you have to score twice. My scenarios force you to have to score twice too.

 

[Ramblin Wreck]

Whole bunch of you assuming that the leading team doesn't score again.

Whole bunch of us are assuming that the trailing team does score again, too.

If the trailing team doesn't score again the answer to the original question is who cares since it doesn't matter if you lose by 7, 8, or 9.

 

[CalBear]

Whole bunch of you assuming that the leading team doesn't score again.

If the leading team scores again, it doesn't really matter what you do. The decision is unimportant in that case, and not really worth discussing.

Not necessarily. Say you go for two and miss and are down 9. Leading team runs the kickoff back and goes up 16. Now you have to score twice and make both two point conversions. What are the math odds of making both two point plays instead of just making one to tie the game at the last opportunity to do so?Say you go for two and miss and are down 9. Leading team kicks a field goal to go up 12. Now you are forced to score two touchdowns instead of one touchdown (with 2 point conversion) and a field goal as an option to force overtime. Is it more likely to make a two point conversion and convert a field goal or more likely to score two touchdowns?Either scenario is unlikely to occur but it's just as likely as those arguing that being down 9 instead of 8 is better because you know you have to score twice. My scenarios force you to have to score twice too.

You are still falling into the same trap. You are assuming that the first 2-point conversion will be missed, and that the last-second 2-point conversion will be a "chance." Mathematically, if you are going to assume the 2-point conversion is missed, you have to assume it in both cases, or your analysis will be wrong.Or to put it another way:Say you go for two and make it. Leading team runs the kickoff back and is up 14. Now you don't need a 2-point conversion to tie.

 

[humpback]

I'll level with you here. I'm not really interested in debating this side of the argument. I just wouldn't assume that NFL teams know better. They don't hire 'challenge guys' to scrutinize replays and radio the coach to tell him to throw the flag or not - which seems like the moat obvious thing in the world. Therefore, to me its just as likely they don't have numbers guys telling the coach what to do with 7 minutes remaining in the game. I betcha most coaches don't want numbers guys telling them what to do.

The word "genius" is thrown about far too often with coaches. "smarter than the others" would be more accurate.

Apples and oranges- challenges are very spur of the moment and you have to make a quick decision right away, and actually they do have people scrutinizing the plays and telling the coach to throw the flag or not. Running numbers for scenario's like we're describing aren't done spur of the moment- coaches have a chart telling them whether they should go for 2 or not under dozens of possible scenario's. That's based on tons of data, not just some one's quick opinion if a toe was on the line or not. It's not like some geek is in the coaches ear telling him whether he should go for 2, his number crunching work was done way in advance, the coach decides how to apply it.

 

[CalBear]

Running numbers for scenario's like we're describing aren't done spur of the moment- coaches have a chart telling them whether they should go for 2 or not under dozens of possible scenario's. That's based on tons of data, not just some one's quick opinion if a toe was on the line or not. It's not like some geek is in the coaches ear telling him whether he should go for 2, his number crunching work was done way in advance, the coach decides how to apply it.

You are assuming this is true, without any evidence to support it. Yes, there is a chart, but there's no evidence that that chart is the result of rigorous mathematical analysis. In fact, there's a good bit of evidence that it's not.

 

[Jersey35]

I'll level with you here. I'm not really interested in debating this side of the argument. I just wouldn't assume that NFL teams know better. They don't hire 'challenge guys' to scrutinize replays and radio the coach to tell him to throw the flag or not - which seems like the moat obvious thing in the world. Therefore, to me its just as likely they don't have numbers guys telling the coach what to do with 7 minutes remaining in the game. I betcha most coaches don't want numbers guys telling them what to do.

The word "genius" is thrown about far too often with coaches. "smarter than the others" would be more accurate.

Apples and oranges- challenges are very spur of the moment and you have to make a quick decision right away, and actually they do have people scrutinizing the plays and telling the coach to throw the flag or not. Running numbers for scenario's like we're describing aren't done spur of the moment- coaches have a chart telling them whether they should go for 2 or not under dozens of possible scenario's. That's based on tons of data, not just some one's quick opinion if a toe was on the line or not. It's not like some geek is in the coaches ear telling him whether he should go for 2, his number crunching work was done way in advance, the coach decides how to apply it.

Spur of the moment or not, most coaches aren't "numbers" guys. Those that are do strange things like go for it on 4th down more frequently and other non-standard approaches.I stand by my argument that the fact that it isn't the norm in the NFL does not in any way indicate whether it is the right thing to do.

 

[cvnpoka]

Not important, but it has actually "worked". The percentage is still ridiculously low.I guess if you're going to go with you're right and everyone in the NFL is wrong, then we'll just agree to disagree.I'd still be interested in seeing some actual evidence that it's been "mathematically proven to be correct" to go for 2 first, but I don't think it can be done.

ya its not really a math problem. the math is always the same. its about information and flexibility.

 

[CalBear]

In the Super Bowl era, 130 teams entered the fourth quarter up by 15. Two of those have lost (1.5%). One of those (ATL@GB, 10/1/89) came from two early TDs and a FG, so it's not relevant to our discussion. The other one (BALvJAX, 11/24/96) they went for one early, and tied it with a 2-point conversion to send it into OT. (Unfortunately PFR doesn't have times on its box scores, so I don't know exactly when the two TDs happened).

Now here's the interesting part:

609 teams entered the fourth quarter up by 14. 34 of those have lost (5.6%), with four ties.

446 teams entered the fourth quarter up by 17. 12 of those have lost (2.7%), with one tie.

The difference between being down by 15 and down by 14 is a two-point conversion; if all else is equal, the likelihood that you can come back from a 15-point deficit should be equivalent to:

( probability of coming back from 14 down ) * (( probability of making 2-point conversion) / ( probability of making PAT ))

That is, if a 2-point conversion is a 40% probability, and a PAT is a 98% probability, we would expect our win percentage when down by 15 to be about 41% of our win percentage when down by 14. In reality, the win percentage when down by 15 has been 27% of the win percentage when down by 14.

In addition, teams down by 17 (clearly three scores) have won nearly twice as often as teams down 15. Obviously this is a counter-intuitive result, and it strongly suggests that teams down 15 are making sub-optimal decisions.

 

[Maurile Tremblay]

^ That's pretty interesting.

 

[Ignoratio Elenchi]

In the Super Bowl era, 130 teams entered the fourth quarter up by 15...

2-point conversions weren't used in the NFL until the mid-90s, so you'd probably want to restrict this data to years since then.

 

[Jersey35]

In the Super Bowl era, 130 teams entered the fourth quarter up by 15...

2-point conversions weren't used in the NFL until the mid-90s, so you'd probably want to restrict this data to years since then.

Nonetheless, that's really interesting stuff!

 

[TobiasFunke]

Head coaches may not be math guys, but NFL teams absolutely do have math guys in their organization, and I guarantee you they've run a lot more numbers than we have. How do you think they come up with their play call charts? They don't just throw things at a dart board, they run numbers. If they were so clearly favoring go for 2 first in this scenario, then that's what their charts would say to do. Sure, they can deviate from their charts, but not 66 out of 66 times. I'm pretty sure they have this scenario on their chart, and I highly doubt it says go for 2 first.

I assume you're saying that a HC is more likely to keep his job by "going by the book", correct? Then wouldn't you be saying that the book says to kick it first? How could that be the case if the mathematically correct thing to do is clearly go for 2 first?

Because "the book" is demonstrably wrong in a lot of different situations. Baseball is getting smarter about this, but football's "book" still is based almost entirely on anecdote and existing practice and not on analysis. (See Romer's study of fourth downs).Here's one reason to believe the book is wrong: in 66 attempts, it has never worked.

Not important, but it has actually "worked". The percentage is still ridiculously low.I guess if you're going to go with you're right and everyone in the NFL is wrong, then we'll just agree to disagree.

I'd still be interested in seeing some actual evidence that it's been "mathematically proven to be correct" to go for 2 first, but I don't think it can be done.

The decision maker- in this case, the coach- has motivations factored into his decision other than giving his team the best chance to win the game. Another motivation- 'm guessing his primary motivation- is his job and his wealth. Going for the XP to keep it a "one possession" game in the parlance of sportscasters is the safer play. If you go for the XP, all you have to say if you're asked about it at the press conference is that you wanted to keep it a "one possession game" and most media members and fans won't press the matter further. On the other hand, if you go for two, you have to explain your decision. And as this thread demonstrates, explaining the rationality of this decision to fans is no easy task.Paradoxically, the more people fail to understand the reasons for going for two early and the more the sportscasters refer to an 8 point game by the meaningless phrase "one possession game," the stronger the coach's motivation to make what I and many others believe is the "wrong" decision in terms of giving you the best chance to win. The worse decision as far as winning the game actually becomes the better decision once factors like keeping your job and getting a raise come into play.

In other words, the wrong decision for the team is turned into the right decision for the coach by the fans and the media.