Jump to content
Indianapolis Colts
Indianapolis Colts Fan Forum

Frank Reich and 4th down..


csmopar
 Share

Recommended Posts

"The numbers strongly favored going for it here -- by a factor of 6.6 percentage points in expected win probability value -- in what was one of just two recommended go-for-it fourth-down scenarios on the night for Indy. (The other was a fourth-and-4 call later in this drive, when the Colts appeared to take a delay-of-game penalty in an effort to bait the defense into going offsides before punting.) The offense had a 68 percent chance of converting in what was, according to NGS ball-tracking technology, a true 1-yard to go situation, and the expected win probability with a conversion was set at 60 percent. If the Colts punted, their expected win probability would have dropped to 47 percent."

 

This is the kind of stuff I have a problem with, because its pretty misleading. Its a lot of statistical mumbo jumbo.

 

We didnt do it, and we still won. The key is knowing when to do it and the sheet cant tell you that. 

 

And the variance of why its been effective in other situations probably depends far more on the players and teams involved and is totally not transferrable to a blanket view of what the statistics actually mean. 

 

Its like this......You technically have a 49% chance to win a hand of blackjack. But the house wins a hell of a lot more than 51% of the money.

 

How is that possible? Because decisions have to be made and knowing when to take a chance and when to play it safe is still the key to winning.

 

At a certain point a statistic is only a record of what already happened, not a predictor of what WILL happen.

 

  • Like 7
Link to comment
Share on other sites

4 minutes ago, GoatBeard said:

Its like this......You technically have a 49% chance to win a hand of blackjack. But the house wins a hell of a lot more than 51% of the money.

 

How is that possible?

...because of the single and double green zero. 

  • Haha 1
Link to comment
Share on other sites

9 minutes ago, Four2itus said:

Not much sleep last night. :funny:

Its all good. I didnt even know what that was because I dont gamble. Had to look it up. It happens to us all my friend lol

 

Thats why I used blackjack because its the simplest example of the flaws in the idea of "probability".

 

Of course if you make a big 4th down conversion its gonna give you a better chance of winning than if you punt......but is that because you simply kept the ball away from the other team and they cant score without the ball, driving down thier probability of winning the game by simply lowering their odds of scoring more points than you thru a diminished TOP?

 

Or did it simply increase your odds of winning because it gave you more opportunities to score more points?

 

I suspect there are a lot of people analyzing these numbers that arent truly qualified to do so.

 

 

 

Link to comment
Share on other sites

39 minutes ago, GoatBeard said:

Because decisions have to be made and knowing when to take a chance and when to play it safe is still the key to winning.

 

At a certain point a statistic is only a record of what already happened, not a predictor of what WILL happen.

Actually, nearly 100% of the time stats are merely a record of what already happened in the past.

 

And the stats were formed by some coach making a decision based upon a situation that could be significantly different.  Down and distance are not the only factors in the decision, but that's the simple (because its the only) way all different situations can be reduced to a data point.

 

  • Like 3
  • Thanks 1
Link to comment
Share on other sites

1 hour ago, DougDew said:

Actually, nearly 100% of the time stats are merely a record of what already happened in the past.

 

And the stats were formed by some coach making a decision based upon a situation that could be significantly different.  Down and distance are not the only factors in the decision, but that's the simple (because its the only) way all different situations can be reduced to a data point.

 

Right. Im just responding to the guy who wrote the article because he represents how a lot of people look at probability, and I disagree with their perspective. 

 

 

 

  • Like 1
Link to comment
Share on other sites

10 hours ago, GoatBeard said:

Right. Im just responding to the guy who wrote the article because he represents how a lot of people look at probability, and I disagree with their perspective. 

 

 

 

I was just reinforcing what you were saying.

 

Stats are simply just a mathematical computation of values.  And that's a huge weakness.  Formulas require consistency in the values.  Yards.  Number of Carries.  Interceptions.  Down and Distance, etc.

 

In the NFL, each one of those situations is unique from team to team, even situation to situation.  Its called idiosyncrasies.  Mathematical formulas hate idiosyncrasies....but the NFL is hardly anything else but a collection of idiosyncrasies.

 

NFL stats are pretty basic stuff because all of the idiosyncrasies have to be stripped away in order for the information to fit into the simple formulas. 

 

 

  • Like 1
Link to comment
Share on other sites

I coached football for a long time and punting a football is one of the worst decisions a coach can make. A punt is a turnover no different than a fumble or an interception. You are giving the ball to the other team. On any 4th and 1 you should have at worst a 90@ coversion rate. Why give the ball to the other team when they only have a 10% chance of sucess. What you have to combine is the statistic for making a first down wiht the statistic of the other team scoring once you punt. Lets say you are on your own 20 4 and incheds with a 95% chance of converting (the only thing that would stop the conversion would be basically a bad snap). If you put the ball on a net 40 yard put they get the ball aroung their 40 yard line where that have a least a 50% chance of getting a field goal.   You are much better off going for the first down. 

 

A coach should never punt unless the odds of not making a first down are under 50% assuming normal game conditions. (i.e. up 4 points and have 4and 5 on your 40 1:00 to play you punt and play the odds that they cannot score a td).   

 

Never voluntarily give the ball back to the other team. 

  • Like 1
  • Haha 2
  • Confused 1
Link to comment
Share on other sites

59 minutes ago, JMichael557 said:

I coached football for a long time and punting a football is one of the worst decisions a coach can make. A punt is a turnover no different than a fumble or an interception. You are giving the ball to the other team. On any 4th and 1 you should have at worst a 90@ coversion rate. Why give the ball to the other team when they only have a 10% chance of sucess. What you have to combine is the statistic for making a first down wiht the statistic of the other team scoring once you punt. Lets say you are on your own 20 4 and incheds with a 95% chance of converting (the only thing that would stop the conversion would be basically a bad snap). If you put the ball on a net 40 yard put they get the ball aroung their 40 yard line where that have a least a 50% chance of getting a field goal.   You are much better off going for the first down. 

 

A coach should never punt unless the odds of not making a first down are under 50% assuming normal game conditions. (i.e. up 4 points and have 4and 5 on your 40 1:00 to play you punt and play the odds that they cannot score a td).   

 

Never voluntarily give the ball back to the other team. 

This is completely nonsensical.

 

So if the other team has Vita Vea you think you still have a 95% chance of converting on 4th and inches? What if your team cant run block in short yardage very well? Still 95%? Where are you getting that number anyways? 

 

You wouldnt last a year coaching in the NFL with that mentality.

 

 

"A punt is a turnover no different than a fumble or a interception?"

 

You mean other than the 50-60 yards of difference in field position that drastically increases your opponents odds of scoring points? Cmon.

 

  • Like 5
Link to comment
Share on other sites

1 minute ago, Myles said:

You can always run Hines up the gut.

Well according to JMichael we would have a 95% chance of converting, no matter who we were facing, so why not?

 

Hey JMichael...... going for it and not making it is far more like a turnover than a punt my friend!

  • Like 1
Link to comment
Share on other sites

Analytics can't be the only thing coaches base decision on. (or really anyone in real life decisions) 

 

Have to look at how the team has been playing against said opponent (what has been working/what hasn't) in that particular game. 

  • Like 3
Link to comment
Share on other sites

Analytics are leading to some seriously dumb situational football and honestly its kind of hilarious. If Staley wants to go for it on his own 30 in the second quarter and look like a dufus I am here for  it.

  • Like 2
Link to comment
Share on other sites

28 minutes ago, Nesjan3 said:

Analytics are leading to some seriously dumb situational football and honestly its kind of hilarious. If Staley wants to go for it on his own 30 in the second quarter and look like a dufus I am here for  it.

I cant believe some people put so much stock in this stuff honestly. Its just a tool  some of it is valuable and some of it is worthless.

 

I get it for identifying trends and using it in gameplanning......What your opponent excels at, their tendencies, etc. Or what your team needs to do better, etc. 

 

But some people truly dont understand the consequences of bad decisions like going for it on 4th down and not making it and flipping the field position battle in a highly contested football game. 

 

Stats dont factor in game dynamics like confidence, momentum, home or away, etc.

 

People are not robots they are humans and can be very unpredictable, inconsistent and prone to ups and downs based on the mental aspects involved with playing competitive sports. Carson Wentz is proof of this. His mental state obviously played a factor in his performance last year. He looks completely different this year facing similar adversity. Why? Who really knows other than him? But environment, confidence and comfort certainly play a role there.

 

A lot of the advocates dont want to admit that a lot of it is related to the explosion of fantasy football and the industry that surrounds that,  and companies marketing these stats as giving you an edge in that hobby. The more stats you can offer in that industry, the better you look to consumers. Whether or not they are truly valuable isnt very relevent.

 

And in the NFL these teams are just as thirsty for that same edge. So there are people who analyze the game statistically and provide that service for them. And if used correctly, it can help improve the preperation for games. But misinterpreted it can also lead you astray. And I absolutely feel like the professionals are just as prone to misinterpreting some of them as a novice like myself.

  • Like 2
Link to comment
Share on other sites

I sure as hell am glad this forum or the general public are not deciding when we are going for it and when we are not. Frank seems to have some of the best people behind him in this domain and he has been one of the better coaches making decisions on 4th down, 2pt conversions, etc. If fans were deciding how sports are played we would still be watching post-ups galore and long-2s in the NBA, we'd probably be going 65% rush in the NFL, etc. The general public's intuitions are incredibly bad about what is a good/efficient play and what is not. Those intuitions in huge majority of cases are very much influenced by what those people have gotten used to watching throughout their lives, what they've seen succeed in the past(without ever having seen an alternative in most cases), etc. It's incredibly hard to tell people something they've thought and accepted as the indisputable truth for their whole lives is wrong. Eh... I can't wait until those type of decisions become so mainstream, that we won't have a thread about it every week just like nobody talks about why you shouldn't one dribble into a long 2 instead of shooting the catch and shoot 3 anymore. 

 

  • Like 2
Link to comment
Share on other sites

42 minutes ago, GoatBeard said:

And I absolutely feel like the professionals are just as prone to misinterpreting some of them as a novice like myself.

 

How could you possibly know? If you haven't taken the time to absorb and analyze this data, then how can you know that it's being misinterpreted?

 

What you're saying is that the interpretations you're seeing don't agree with your long held viewpoint, which means those interpretations must be wrong, even though you don't know the information upon which those interpretations are based.

  • Like 1
Link to comment
Share on other sites

16 minutes ago, stitches said:

I sure as hell am glad this forum or the general public are not deciding when we are going for it and when we are not. Frank seems to have some of the best people behind him in this domain and he has been one of the better coaches making decisions on 4th down, 2pt conversions, etc. If fans were deciding how sports are played we would still be watching post-ups galore and long-2s in the NBA, we'd probably be going 65% rush in the NFL, etc. The general public's intuitions are incredibly bad about what is a good/efficient play and what is not. Those intuitions in huge majority of cases are very much influenced by what those people have gotten used to watching throughout their lives, what they've seen succeed in the past(without ever having seen an alternative in most cases), etc. It's incredibly hard to tell people something they've thought and accepted as the indisputable truth for their whole lives is wrong. Eh... I can't wait until those type of decisions become so mainstream, that we won't have a thread about it every week just like nobody talks about why you shouldn't one dribble into a long 2 instead of shooting the catch and shoot 3 anymore. 

 

 

 

lol 

you-need-to-relax-will-ferrell.gif

Link to comment
Share on other sites

1 hour ago, lollygagger8 said:

Analytics can't be the only thing coaches base decision on. (or really anyone in real life decisions) 

 

Have to look at how the team has been playing against said opponent (what has been working/what hasn't) in that particular game. 

 

Very true. It can't just be down and distance. The probabilities can never be evaluated in a pure vacuum, IMO. For example, the Seahawks, for some reason, they are very good on 3rd and short against the run and similarly on 4th and short, just based on my eye test and their stellar LB play in reading gaps over the years. It is not just based on last night's game vs the Saints. Stopping Cam Newton and the Patriots in 2020 on 4th down, stopping Brady and the Patriots in Foxboro in 2016 with another goal line stand, are just a few examples off the top of my head.

 

So, opponents plus run or pass probabilities based on opponents should also be factored into the analytics, as opposed to purely down and distance, IMO. 

  • Like 2
Link to comment
Share on other sites

18 minutes ago, chad72 said:

 

Very true. It can't just be down and distance. The probabilities can never be evaluated in a pure vacuum, IMO. For example, the Seahawks, for some reason, they are very good on 3rd and short against the run and similarly on 4th and short, just based on my eye test and their stellar LB play in reading gaps over the years. 

 

So, opponents plus run or pass probabilities based on opponents should also be factored into the analytics, as opposed to purely down and distance, IMO. 

The things that trickle down to the general public are probably very surface level. I personally would be shocked if the models teams use (at least the ones invested in it) are not much more complex and all encompassing. Just an example of things I can think of and are probably relatively easy to implement into a model - down and distance, field position, opponent stats, stats vs one front or another front, stats vs specific type of coverage, score, game clock, possession time, flow of the current game, recent form, the specific REFS officiating the game ... and so on and so forth. I doubt the information that comes to Frank on game day is product of simple stat like what % of 4th and 1 get converted in the NFL, regardless of opponents, regardless of score and time, regardless of all other things you can think of. 

Link to comment
Share on other sites

18 minutes ago, Superman said:

 

How could you possibly know? If you haven't taken the time to absorb and analyze this data, then how can you know that it's being misinterpreted?

 

What you're saying is that the interpretations you're seeing don't agree with your long held viewpoint, which means those interpretations must be wrong, even though you don't know the information upon which those interpretations are based.

Because people make mistakes Superman. The data is in on that fact and it has been confirmed.

 

Or are you one of these people that confuse evolution and improvement? They arent always the same thing.

 

 

Link to comment
Share on other sites

13 minutes ago, stitches said:

The things that trickle down to the general public are probably very surface level. I personally would be shocked if the models teams use (at least the ones invested in it) are not much more complex and all encompassing. Just an example of things I can think of and are probably relatively easy to implement into a model - down and distance, field position, opponent stats, stats vs one front or another front, stats vs specific type of coverage, game clock, possession time, flow of the current game, recent form, the specific REFS officiating the game ... and so on and so forth. I doubt the information that comes to Frank on game day is product of simple stat like what % of 4th and 1 get converted in the NFL, regardless of opponents, regardless of score and time, regardless of all other things you can think of. 

 

The more variables you introduce, the less predictable and useful it gets, IMO. Plus, there needs to be some stability on the other side w.r.t opposing coaches (useful with teams with less coaching turnover), coverages (like Eberflus and some teams that typically don't do something post snap that they don't show pre-snap) etc. that are less variable for the information to become more useful.

 

Then, if Edelman is beating his guy like a drum most of the time, I don't care what the analytics say, I am going to him because it is the same guy he is lined up against and the QB chemistry (in this case Brady) with him. Insert any go to guy on a key 3rd or 4th down that a QB has confidence in similarly. Like the sideline throw with a 10-15% chance of completion etc., that is why the commercial "STAT THAT". :) 

 

Link to comment
Share on other sites

38 minutes ago, stitches said:

I sure as hell am glad this forum or the general public are not deciding when we are going for it and when we are not. Frank seems to have some of the best people behind him in this domain and he has been one of the better coaches making decisions on 4th down, 2pt conversions, etc. If fans were deciding how sports are played we would still be watching post-ups galore and long-2s in the NBA, we'd probably be going 65% rush in the NFL, etc. The general public's intuitions are incredibly bad about what is a good/efficient play and what is not. Those intuitions in huge majority of cases are very much influenced by what those people have gotten used to watching throughout their lives, what they've seen succeed in the past(without ever having seen an alternative in most cases), etc. It's incredibly hard to tell people something they've thought and accepted as the indisputable truth for their whole lives is wrong. Eh... I can't wait until those type of decisions become so mainstream, that we won't have a thread about it every week just like nobody talks about why you shouldn't one dribble into a long 2 instead of shooting the catch and shoot 3 anymore. 

 

The NBA has nothing to do with the NFL. The two sports couldnt be more different. 

 

Field position is not a factor in the game of basketball like it is in football. And this is a huge difference. 

 

In football, field position is a very important factor and the data backs that up. Its directly linked with your ability to actually score points. Its so important that every decision needs to consider it. Its very hard to score. In fact its so hard to score it is possible you dont score at all. None of this has changed. Its a fundamental aspect of the sport.

 

In basketball there has never been a shutout at any meaningful level of the game. It is designed to have a lot of scoring,  especially at the NBA level. So of course optimizing that scoring is important. Every single play is designed to produce points. So if you can maximize the amount of points you score thru shot selection, it matters. And that has always been a part of the game, its certainly not new. Its how they came up woth the 3 point shot in the first place. My fathers generation invented it, not yours lol

 

In football a single play can be designed to gain a single yard. Thats a stark contrast.

 

Like really......What does Bball have to do with football?

Link to comment
Share on other sites

15 minutes ago, GoatBeard said:

Because people make mistakes Superman. The data is in on that fact and it has been confirmed.

 

Or are you one of these people that confuse evolution and improvement? They arent always the same thing.

 

How can you tell someone they're making a mistake when you don't know the information they were using to make their decision?

 

The bottom line is you seem to value your gut instincts more than you value the data, and your bias is confirmed when you see a fail on a decision you don't agree with. 

 

Someone mentioned the Chargers earlier. Going into that Ravens game, they were 7/8 on fourth down. How many points did they gain on those 7 conversions, and how much did that impact their ability to win games? That gets thrown out of the window when you criticize their decisions in the Ravens game, and of course that criticism only comes because of the outcome. They're still 7/10 on fourth down, still a net positive overall.

 

End of the day, each decision has to stand on its own. The model may say you should go for it on 4th and short, but if it's late in the fourth quarter and you're up 9 points, and you know your LG isn't 100% and the QB is kind of flustered, you might make a different decision. That's fine. It doesn't mean the model is wrong. Humans still have to make the decisions, and humans still have to execute the play on the field. But the data, and the interpretation of it, is still legitimate. 

 

It's just pretty arrogant to say 'I don't know what information you're using, but you're using it wrong.' 

  • Like 2
Link to comment
Share on other sites

11 minutes ago, chad72 said:

 

The more variables you introduce, the less predictable and useful it gets, IMO. Plus, there needs to be some stability on the other side w.r.t opposing coaches, coverages (like Eberflus and some teams that typically don't do something post snap that they don't show pre-snap) etc. that are less variable for the information to become more useful.

 

Then, if Edelman is beating his guy like a drum most of the time, I don't care what the analytics say, I am going to him because it is the same guy he is lined up against. Insert any go to guy on a key 3rd or 4th down that a QB has confidence in similarly. Like the sideline throw with a 10-15% chance of completion etc., that is why the commercial "STAT THAT". :) 

 

Right. You might not want to target Jalen Ramsey in coverage. The analytics certainly suggest that.

 

But if he sprained his ankle last week amd you have a great receiver, you absolutely would.

 

I love analytics guys attitude......"if you havent analyzed all the data, you dont know what youre talking about"........while they themselves havent analyzed all the data either but are clearly offended that i dont believe in the data I havent analyzed because they obviously believe in the data they havent analyzed.

 

Geez. Its my opinion. Go ahead and argue against my opinion. Tell me why you think you should never punt. Ive already stated why you should sometimes. 

 

You dont know anymore than I do about what they use this data for unless youve been there with them. And we all know you havent.

Link to comment
Share on other sites

12 minutes ago, GoatBeard said:

The NBA has nothing to do with the NFL. The two sports couldnt be more different. 

 

Field position is not a factor in the game of basketball like it is in football. And this is a huge difference. 

 

In football, field position is a very important factor and the data backs that up. Its directly linked with your ability to actually score points. Its so important that every decision needs to consider it. Its very hard to score. In fact its so hard to score it is possible you dont score at all. None of this has changed. Its a fundamental aspect of the sport.

 

In basketball there has never been a shutout at any meaningful level of the game. It is designed to have a lot of scoring,  especially at the NBA level. So of course optimizing that scoring is important. Every single play is designed to produce points. So if you can maximize the amount of points you score thru shot selection, it matters. And that has always been a part of the game, its certainly not new. Its how they came up woth the 3 point shot in the first place. My fathers generation invented it, not yours lol

 

In football a single play can be designed to gain a single yard. Thats a stark contrast.

 

Like really......What does Bball have to do with football?

I don't know why you think they don't consider field position, down and distance and a host of other factors. There is a reason Frank goes for it very regularly in the opponent side of the field on 4th and short and almost never on our own 20. It's precisely because those things are being taken into account. BTW getting more points is NOT always the best thing you can do to win a game. The goal of most of those decisions is to maximize win probability, not points. There are situations when draining the clock is more valuable and when the only way you could lose a game is by scoring(rather than letting the clock expire). All those are part of the things that come into consideration when coaches make their decisions and IMO the more and the better information(advanced analytics included) the coach has the more likely it is he will make good decisions for his team.

 

 

Basketball has very little to do with football but the fans, pundits and old guard of the two sports have a lot in common when it comes to innovations and new ways the games get played. Just 5-10 years ago you could hear NBA legends, national commentators and pundits on national TV/radio spout takes about how teams should establish low post presence (that is the equivalent of establishing the run in the NFL, IMO), how jump shooting teams can't win in the league, how getting shots 2 steps into the paint was better shot than the 3 because it's closer to the basket, etc. 

Link to comment
Share on other sites

1 hour ago, GoatBeard said:

This is completely nonsensical.

 

So if the other team has Vita Vea you think you still have a 95% chance of converting on 4th and inches? What if your team cant run block in short yardage very well? Still 95%? Where are you getting that number anyways? 

 

You wouldnt last a year coaching in the NFL with that mentality.

 

 

"A punt is a turnover no different than a fumble or a interception?"

 

You mean other than the 50-60 yards of difference in field position that drastically increases your opponents odds of scoring points? Cmon.

 

how long does a coach last in the NFL that cannot design and offensive set that can't at least get a few inches. That should always be sucessful. The average net punt in the NFL is 45 yards. I would agree that the closer you are to the end zone the greater the chance of scoring but even from your own 20 yard line a team should score 30% of the time.  So this is how the game should be viewed.  I have two sacks of Marbles. I give you one sack which has 100 balls 90 are black and 10 are white. I take a sack in which 30 are black and 70 are white.  You get to make the decision out of which bag a ball will be selected. If the ball selected is black you win.  Which bag would you choose to draw from?

Link to comment
Share on other sites

10 minutes ago, GoatBeard said:

Right. You might not want to target Jalen Ramsey in coverage. The analytics certainly suggest that.

 

But if he sprained his ankle last week amd you have a great receiver, you absolutely would.

 

I love analytics guys attitude......"if you havent analyzed all the data, you dont know what youre talking about"........while they themselves havent analyzed all the data either but are clearly offended that i dont believe in the data I havent analyzed because they obviously believe in the data they havent analyzed.

 

Geez. Its my opinion. Go ahead and argue against my opinion. Tell me why you think you should never punt. I've already stated why you should sometimes. 

 

You dont know anymore than I do about what they use this data for unless youve been there with them. And we all know you havent.

 

Sometimes, we have a QB (has happened with Peyton, Big Ben etc.) taking a shot deep on 3rd down into someone else's territory by the end zone with a risky throw knowing that even if gets intercepted as a turnover, the downside is typically giving it to the opposition within their 20 yard line, which typically ends up about the same as a punt on 4th down. You don't try to throw the INT but you feel that kind of risk is acceptable while giving your WR a chance to go up and get it. Like Pittman???

 

Link to comment
Share on other sites

2 minutes ago, Superman said:

 

How can you tell someone they're making a mistake when you don't know the information they were using to make their decision?

 

The bottom line is you seem to value your gut instincts more than you value the data, and your bias is confirmed when you see a fail on a decision you don't agree with. 

 

Someone mentioned the Chargers earlier. Going into that Ravens game, they were 7/8 on fourth down. How many points did they gain on those 7 conversions, and how much did that impact their ability to win games? That gets thrown out of the window when you criticize their decisions in the Ravens game, and of course that criticism only comes because of the outcome. They're still 7/10 on fourth down, still a net positive overall.

 

End of the day, each decision has to stand on its own. The model may say you should go for it on 4th and short, but if it's late in the fourth quarter and you're up 9 points, and you know your LG isn't 100% and the QB is kind of flustered, you might make a different decision. That's fine. It doesn't mean the model is wrong. Humans still have to make the decisions, and humans still have to execute the play on the field. But the data, and the interpretation of it, is still legitimate. 

 

It's just pretty arrogant to say 'I don't know what information you're using, but you're using it wrong.' 

Thats not what I said. Its pretty arrogant to suggest I said something I never said when its all right here in print for everyone to see, and then have the nerve to suggest Im arrogant. Because I do know what information they are using. They are using probability of making a 4th down. This isnt rocket science.

 

And Im confident in my ability to see the flaws in another persons decision making. That has nothing to do with football.

 

"Like when you say they are still 7/10 on fourth down, so its still a positive overall"

 

Which isnt true if the 3 missed opportunities led to the other team scoring points and the 7 made opportunities didnt lead to your team scoring points. 

 

The data isnt the problem its your interpretation and assumptions made that are the problem.

 

 I complained in the Ravens game about chasing points when they had a big lead and compounding the previous mistake of missing an XP. Thats just a fact. It likely cost us that game. If we kick that 2nd XP and make itz we win. And it was probable according to the data. More probable than making the 2 point conversion in fact. Which is a proper use of data to make a solid decision, IMO.

 

 

Link to comment
Share on other sites

6 minutes ago, stitches said:

I don't know why you think they don't consider field position, down and distance and a host of other factors. There is a reason Frank goes for it very regularly in the opponent side of the field on 4th and short and almost never on our own 20. 

 

Basketball has very little to do with football but the fans, pundits and old guard of the two sports have a lot in common when it comes to innovations and new ways the games get played. Just 5-10 years ago you could hear NBA legends, national commentators and pundits on national TV/radio spout takes about how teams should establish low post presence (that is the equivalent of establishing the run in the NFL, IMO), how jump shooting teams can't win in the league, how getting shots 2 steps into the paint was better shot than the 3 because it's closer to the basket, etc. 

I never said that. Are you even reading my posts?

 

I was oroginally talking about the article posted.

 

And then I responded to JMichaels post suggesting the exact opposite of what you just said.

 

He said "You should never punt", which scoffs at field position and its importance.

Link to comment
Share on other sites

5 minutes ago, GoatBeard said:

Thats not what I said. Its pretty arrogant to suggest I said something I never said when its all right here in print for everyone to see, and then have the nerve to suggest Im arrogant. Because I do know what information they are using. They are using probability of making a 4th down. This isnt rocket science.

 

And Im confident in my ability to see the flaws in another persons decision making. That has nothing to do with football.

 

"Like when you say they are still 7/10 on fourth down, so its still a positive overall"

 

Which isnt true if the 3 missed opportunities led to the other team scoring points and the 7 made opportunities didnt lead to your team scoring points. 

 

The data isnt the problem its your interpretation and assumptions made that are the problem.

 

 I complained in the Ravens game about chasing points when they had a big lead and compounding the previous mistake of missing an XP. Thats just a fact. It likely cost us that game. If we kick that 2nd XP and make itz we win. And it was probable according to the data. More probable than making the 2 point conversion in fact. Which is a proper use of data to make a solid decision, IMO.

 

Right on.

Link to comment
Share on other sites

5 minutes ago, GoatBeard said:

I never said that. Are you even reading my posts?

 

I was oroginally talking about the article posted.

 

And then I responded to JMichaels post suggesting the exact opposite of what you just said.

 

He said "You should never punt", which scoffs at field position and its importance.

I replied to your messages responding to my posts. Of course I read those.

 

I'm not on the side of never punting, I think the punt is a very fine choice in a lot of situations, but I also think people would be shocked to learn in how many cases when they automatically think "punt", going for it is actually a solid choice.

Link to comment
Share on other sites

6 minutes ago, JMichael557 said:

how long does a coach last in the NFL that cannot design and offensive set that can't at least get a few inches. That should always be sucessful. The average net punt in the NFL is 45 yards. I would agree that the closer you are to the end zone the greater the chance of scoring but even from your own 20 yard line a team should score 30% of the time.  So this is how the game should be viewed.  I have two sacks of Marbles. I give you one sack which has 100 balls 90 are black and 10 are white. I take a sack in which 30 are black and 70 are white.  You get to make the decision out of which bag a ball will be selected. If the ball selected is black you win.  Which bag would you choose to draw from?

But football isnt marbles.

 

Punting is a strategic move. There is a reason for it. It is not a turnover which is what you originally said.

 

The average NFL drive is 6 plays. 

 

The average NFL play is around 5 yards.

 

Then its safe to say the average drive is likely around 30 yards. Most drives dont end in points, solely because of field position. And long scoring drives skew these numbers a little more. Most drives are short. And lreventing points is almost as good as scoring them.

 

30 yards is not enough to be in scoring position if you start at your own 20-25 yard line. 

 

But is is enough to score if you start somewhere around midfield.

 

This is why field position is so important. 

 

Its not hard to figure out why teams will only take the risk if they are on the outer edge of that equation, or the opponents 40 yard line and in. 

Link to comment
Share on other sites

8 minutes ago, GoatBeard said:

 

"Like when you say they are still 7/10 on fourth down, so its still a positive overall"

 

Which isnt true if the 3 missed opportunities led to the other team scoring points and the 7 made opportunities didnt lead to your team scoring points. 

 

 

We are getting into IFs and BUTs here, based on revisionist tendencies though. IF Blankenship makes the FG, we win, we overcome the low probability "gut feeling" decisions the coach made based on his confidence with his players to execute. Teams overcome situations and the impact on the score. That day vs the Ravens, we did not. 

 

The thing is analytics is not black and white once variables are introduced for gray areas, what you do with the probability can backfire on you just as much as it can help you. Ultimately, the team has to find a way to win, whether they were put in lower probability situations or not. Maybe it is the KISS principle or like Al Davis said, "Just Win Baby". 

Link to comment
Share on other sites

15 hours ago, GoatBeard said:

"The numbers strongly favored going for it here -- by a factor of 6.6 percentage points in expected win probability value -- in what was one of just two recommended go-for-it fourth-down scenarios on the night for Indy. (The other was a fourth-and-4 call later in this drive, when the Colts appeared to take a delay-of-game penalty in an effort to bait the defense into going offsides before punting.) The offense had a 68 percent chance of converting in what was, according to NGS ball-tracking technology, a true 1-yard to go situation, and the expected win probability with a conversion was set at 60 percent. If the Colts punted, their expected win probability would have dropped to 47 percent."

 

This is the kind of stuff I have a problem with, because its pretty misleading. Its a lot of statistical mumbo jumbo.

 

We didnt do it, and we still won. The key is knowing when to do it and the sheet cant tell you that. 

 

And the variance of why its been effective in other situations probably depends far more on the players and teams involved and is totally not transferrable to a blanket view of what the statistics actually mean. 

 

Its like this......You technically have a 49% chance to win a hand of blackjack. But the house wins a hell of a lot more than 51% of the money.

 

How is that possible? Because decisions have to be made and knowing when to take a chance and when to play it safe is still the key to winning.

 

At a certain point a statistic is only a record of what already happened, not a predictor of what WILL happen.

 

 

That's not quite how it works regarding you blackjack take.

 

It's a 1% edge to the house of Basic Strategy is followed. So yes, if the 'wrong' choice is made is skews the odds. There isn't knowing when to take a chance or not as it  is the best way to play the game. Why you can quickly anger a table if you don't know how to play.

 

However, BJ is somewhat unique (or used to be) in that it has a 'memory' in that hands are impacted by previous hands. That's where card counting comes in. BUT you don't go away from playing nonsense, you just bet more on hands where the count is good. But hugely varying your bet makes you stand out like a sore thumb. Especially when dealers likely know the count too. 

 

*Source - I used to be a croupier.

 

Football is a much more complex multifactorial system. But the same premise applies. If you have enough data to give you a positive expected value (EV) for a choice in a certain situation, over the long term you win if you stick to it. But in football terms you're applying a overall probability that won't be fully tuned the actual down's circumstances. Match ups, weather, what the QB had for dinner. 

 

Not so much statistical mumbo jumbo, more it's often badly explained/understood. 

 

 

  • Like 1
Link to comment
Share on other sites

2 minutes ago, stitches said:

I replied to your messages responding to my posts. Of course I read those.

 

I'm not on the side of never punting, I think the punt is a very fine choice in a lot of situations, but I also think people would be shocked to learn in how many cases when they automatically think "punt", going for it is actually a solid choice.

Well i felt like you were indirectly talking to me because Ive posted a lot in this thread and suggesting Im some old fashioned goof that scoffs at the progress of analytics, which is only partly true. I personally think it has ruined the intrigue in the NBA, but thats abother story altogether. But I do understand WHY they are using it. They see it as a cheat code. A way to produxe more points on much less work. I was just a big fan of that work so I dont care for it.

 

So maybe I over reacted a little lol

 

Anyways, i dont mind the data. The data isnt complicated at all. Applying it to the game is very complicated. Using it effectively is far from easy. And I am only speaking to people who misuse it.

 

 

  • Like 1
Link to comment
Share on other sites

19 minutes ago, GoatBeard said:

Thats not what I said. Its pretty arrogant to suggest I said something I never said when its all right here in print for everyone to see, and then have the nerve to suggest Im arrogant. Because I do know what information they are using. They are using probability of making a 4th down. This isnt rocket science.

 

And Im confident in my ability to see the flaws in another persons decision making. That has nothing to do with football.

 

"Like when you say they are still 7/10 on fourth down, so its still a positive overall"

 

Which isnt true if the 3 missed opportunities led to the other team scoring points and the 7 made opportunities didnt lead to your team scoring points. 

 

The data isnt the problem its your interpretation and assumptions made that are the problem.

 

 I complained in the Ravens game about chasing points when they had a big lead and compounding the previous mistake of missing an XP. Thats just a fact. It likely cost us that game. If we kick that 2nd XP and make itz we win. And it was probable according to the data. More probable than making the 2 point conversion in fact. Which is a proper use of data to make a solid decision, IMO.

 

 

That's why you calculate expected value (points) not just a win/lose succeed/fail for the trials. 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...