The Dolphins Make Me Cry.com - Forums

This isn't a criticism of Gase specifically or even the Dolphins, but of the NFL in general.  The coaches in the NFL (all of them) don't seem to understand analytics.  Coach after coach is chasing this old, antiquated mindset to extend the game and keep it close.  We punted on 4th and 4 with mere minutes remaining, giving the ball to Tom Brady with the lead. 

Under almost no circumstances does that win the game.  We needed one of the craziest plays in franchise history with no timeouts and no time left to pull it off.

I just can't wrap my brain around this kind of thinking -- the number of things needed to win the game when you punt away is huge.  You need to stop the other team and hope they punt you the ball back at the same place, but they're going to burn time and all your timeouts.  ...THAT....vs getting 4 yards.

The numbers just don't make sense.  Firstly, I think that all of these coaches come from generations of coaches in these coaching trees that all learn the same way -- extend the game, keep it close, win at the end.  I don't think that was ever smart and correct by the math of it, but it's even less true now as the rules change and offenses have so much more freedom.  Your NFL games aren't 14-9 grind out field position battles anymore.

Of course I'm super glad we won the game, but we essentially gave it away.  People may call out Gase for his conservatism, but it isn't Gase; it's the entire league.  The next guy will likely come in and do the same thing.

I agree. We play not to loose rather than playing to win. There is no way we shouldn't have tried to get a first down in that situation.

It's not just us.  That's what makes me so crazy.  It'd be way easier to just the play calling of one person, but it's every team.  We just notice it more because we seem to be in this situation a lot where we need it.  They all do it.  It's nuts.

Hear hear!

I haven't seen the numbers, but I have a hard time imagining that punting would be the better option.

I think the only thing that changes this trend is if the talking heads on TV call out the coaches as it happens. Just give the numbers and show how bad some of these decisions are.

I'm sorry but this is all just hindsight. The coaches don't have the benefit of hindsight when they make they decision. The reason that the last play was even necessary to win the game for Miami is because of what happened AFTER that punt and Gase had no way of knowing what was going to happen and you didn't either. Only AFTER it happens are you bitching about it. I'm not saying that you weren't saying that before the punt, perhaps you were, but you don't KNOW what's going to happen. If they go for it and don't get the first down it's a HORRIBLE decision, that is WITHOUT A DOUBT. Now maybe they do get the 1st down but that doesn't guarantee a win. Not even close.

The situation is that Miami is trailing 28-30. It's 4th and 4 from the 40 yard line with 4:38 left in the game and you have 2 timeouts left. Going for it is NOT the right decision. It's just not. There's A LOT of game left to play and there's a LOT of things that can happen. You can still win this game punting. If you punt you pin them back in their end. Maybe they muff the punt and you get the ball back and you kick a FG to take the lead or better yet score a TD to take the lead? Maybe your defense can force a turnover in their end and you drive for the winning FG or TD? Maybe your defense can hold them to a punt and your team still drives for the winning FG or TD? These are all possibilities that are at LEAST as likely as you making 4 yards on that 1 play that merely gives you a CHANCE to win the game. Now maybe if you are within Pats territory and a 1st down gives you an immediate chance at a FG and the lead then you go for it, but it doesn't. Not even close. Even if you GET the 1st down there's absolutely no guarantee you don't have to punt 2 minutes later with NOTHING gained. The risk is NOT worth the reward as you still may not even get into FG range with a 1st down.

Sorry Dave but you would make an even worse head coach then the head coaches in the NFL because you wouldn't have hindsight on your side. Second guessing is your right, but it doesn't make you right. The decision was the right one. Not because we won the game but because the circumstances dictated that decison at that time. Gase had no way of knowing that the decision would mean they would have 7 seconds to try to win the game on a 69 yard multi lateral impossible play THAT WOULD WORK!!!!! That's just how it turned out.

I don't think that can happen either, because the way you get to become a talking head is by being a player or a commentator that grew up under this same style of thinking.

I think you have to have an owner with foresight to understand business, and that person needs to employ a statistician.  I'm not expecting Gase to all of a sudden understand math -- it's not his fault.  You need an outsider to shake it up.

Just FYI the win probability indicator (which is mostly BS anyway but there's at least some math behind it I assume) I don't think really supports your notion either. Before that 4th and 4 yard play from the 40 it indicated NE had about a 70% chance of winning. After the punt NE had about a 74% chance of winning meaning that punting only lowered Miami's odds of winning by about 4%. I feel VERY strongly that if Miami had gone for it and not made it their odds of winning would have been lowered by MUCH more. NE probably would have had above an 80% chance of winning at that point. Granted that their odds of winning had they made it would have improved, but I don't think Miami's odds would have improved more then NE's would have improved had they failed and your odds of making a 4th and 4 are NOT very good. No where close to 50/50. Maybe 20% or 25%? So you are banking on a play with about 20 to 25% chance of making it to increase your odds of winning say 5% whereas if you don't make probably decrease your odds of winning by an even larger %? I don't believe that's statistically smart football and the win probability indicator doesn't think so either.

If you want to look at the tipping point in the game look to earlier in the game when Miami was winning 28-27 and had a 3rd and 1 from their own 44. At that point they had a better then 50% chance of winning, one of the few points in the game where they had that. Miami lost 8 yards on the ensuing play when Tannehill was sacked and the win probability flipped back to NE's favor that they would never relinquish till the last play of the game. They couldn't make a 3rd and 1 from the 44 to take command of the game but you think going for it on 4th and 4 from their own 40 is smart football? I don't.

Once you go for it on 4th and 4 at your own 40, down by 2, you're not going to punt the next 4th down.   You're all in at that point, as well you should be.

Considering you're maybe 20-25 yards away from a potential game winning FG, I can't see how you care anymore about field position.  You have control of the game, and have been moving the ball well, so 2 minutes and 2 TO to gain 20 yards should be plenty.

A more aggressive coach rams it down your throat and takes the win.  To punt in that scenario is conservative, and you saw the result.

Or they fail to make the 4th down and give the ball to the Patriots at their own 40 yard line and practically guarantees that you will be down by 5 (at the very least) when you get the ball back and have to go the length of the field to take the lead or maybe even get you within 2. Conservative doesn't always lose (as you saw the result yesterday) nor does aggressive always win. I'd say by and large they are pretty equal in terms of winning percentage. The problem is that hindsight is 20/20 and it's ALWAYS the OTHER thing that would have won the game for you when you lose. Sorry but that's just not the case.

Going for it on 4th down there makes sense only if you are the better team and you are playing at home or if there are under 2 minutes to play. In that circumstance I would agree with you. Be aggressive and let the chips fall where they may, you probably won't hurt your chances of winning too much even if you don't get it but that's not the situation that Miami was in. Their chances of making the 1st down are low and their chances of winning even if they make it are still low. Their chances of winning by punting are low too, but better then if they don't make it. In that case punt the ball and hope for a turnover or something that gives you a better chance at winning the game, don't lose it right there with 4 and half minutes to go in the game. That would be second guessed even more I guarantee it because the odds are not with you that this will win the game for you even if you make it.

Pappy, I sincerely appreciate you giving me a well-reasoned answer that you attempt to support with data.  That's more than anything else I've heard.

That said, I don't agree with your numbers.

By the numbers, the Dolphins average 5.6 yards per offensive play (6.2 at home).  I think that the odds of converting are higher than 20-25%.  But also, even if you don't convert, you're not really in much different shape.  You still have to hold NE to a FG and get the ball back to score, you will just have more time to do it.

Or they will score a TD and you'll lose, which you would have, regardless.

I have a hard time believing that punting that ball leads to Miami winning that game 1 out of 4 times, which your numbers suggest...there's just no way.  And what Brian says is also true, once you make that 1st time, you're all in to win the game, so your traditional 3 down success rates go out the window.  You're in 4 down territory for the remainder of the game.

When I compare what is said today by talking heads (and coaches) to what was said 10-20-30 years ago, I think there has definitely been progress.

Average per play doesn't really tell you much. You could get 0 yards on 3 of 4 plays and then get 20 on 1 play and you have an average of 5 yards per play but only a 25% success rate of gaining 4 yards on a play. That plus 1st and 2nd down are obviously different from 3rd or 4th down so you can't treat all downs the same. Just look at conversion rates on 3 or 4th down for teams. Let's assume for arguments sake that converting a 1st down on 3rd and 4 yards is the exact same as converting a 1st down on 4th and 4. I don't really have numbers for that for Miami but Miami's 3rd down conversion rate for all distances is around 32% for the year. I think my numbers are pretty close since they are probably better at 3rd and 1-3 then they are at 3rd and 4 or more so it's probably a bit under that 32% or about what I said.

I thought I already wrote this, but it's not here, so I either deleted it by accident or posted in another thread.

Anyway, I looked and found 2017 NFL league-wide statistics.  The rate of conversion on 3rd and 4 is 50%.

Lol ... you posted it in another thread. I was a bit confused.

I went ahead and deleted it for you.

Hah...thanks.  I thought I was going crazy.

So then 2 questions.
1) What is the league wide rate of conversion of 4th and 4?
2) What is Miami's rate of conversion for 4th and 4?

The 1st question would be relevant to the league in general whereas the 2nd question is relevant just to Miami.

But just for the sake of argument let's say that Miami's rate of conversion for 4th and 4 is 50%. Miami would have basically been doing a coin flip on the outcome of the game rather then giving their defense a chance to do it's job. Not sure I agree with you that is the preferred approach no matter how bad your defense is and I can certainly understand why coaches are reluctant to risk it all on a coinflip. They would prefer to trust their punt team and defense.

If it'd been 4th and 1 or they would have been on New England's side of the field I think Gase would have went for it.

I looked up the 3rd down rates, because there's a lot of data on it and it's effectively 4th down, because you punt on 4th down anyway.  I don't have the 4th down data, but even if I did, I think it's less valuable, because it'll be a super-small sample size and most 4th down attempts are under weird circumstances, with many coming at the end of the half or the game and not indicative of a normal play.

This site is interesting: http://www.advancedfootballanalytics.com/index.php/home/research/game-strategy/120-4th-down-study

And specifically this picture: (http://farm3.static.flickr.com/2618/3688516023_07450826e5_o.png)

This basically shows how just the math works out.   Go for it on 4th and 1 anywhere on the field.  But at the 40 yard line, you should even be going for it on 4th and 11.  Of course, this chart isn't time-specific.

According to that chart Miami did what it should have which is punt since they were 60 yards away and it was 4th and 4.

^ This chart doesn't apply to the Miami situation, since the argument to go for it was time/score based.  The article specifically uses this chart to define "normal" situations. 

This chart is saying that you should go for it on 4th and 3 from 60 yards out, in the middle of the first quarter -- time and score not withstanding.

Ah so you're not saying that Gase did the wrong thing because of analytics, he did the wrong thing because it was the 4th quarter with 4 and half minutes to go in the game analytics be damned! I misunderstood. In my opinion the analytics are your guide and you only deviate from the guide when you have no other choice. Clearly there was another choice.

I feel that Gase made the wrong choice in this situation because there wasn't time to correct.  I wouldn't be bitching about punting from 70 yards out if it were the middle of the 2nd quarter.  I also wouldn't be bitching about it if we'd had the lead.

In this case, it was circumstantial.  But while searching for conversion rates to help illuminate my position, I ran across this chart.  ...which is also something I think has merit.

I think it was circumstantial in the fact our defense sucks. The odds of us getting a first down were better than our defense stopping New England in time for us to score again. Outside of a less than 1% play working ... I'd have been right.

But then again there was a questionable PI call that got NE inside the 10 as well. If that doesn't happen maybe Miami forces a punt and they are only down 2 points and just need to drive for a winning FG. The odds of that happening are MUCH better. Maybe they still lose, but maybe not. My point is that it's not right to say that we needed a 1% chance play of working because at the time they punted they didn't know that. The game was still very much up for grabs by either team. What happened AFTER the punt made a HUGE difference in needing that 1 play and a BIG part of that was the PI call and that could have just as easily been allowed. That play changes the whole dynamics of the last 2 minutes of the game, not the punt at 4:30 left.

I saw the Seattle Seahawks continue to do this and lose to the Niners.  It's just crazy.  They kept pinning them at the 1 over and over, rather than getting 4 yards and some points.  Lost to a turd team in OT because they kept trying to extend the game.

And then there's the Cowboys that kept going for it on 4th down and not making it and lost. The Dolphins went for it on 4th down and were sacked which made the final score look even more lopsided. Giants went for it on 4th and 4 from the 4 yard line and failed to make a TD and lost.  Every week you can go through the box scores and find games where teams went for it on 4th down and lost. It works both ways, it's just that people always think when they lose the other thing would have made it a win. That's a complete fallacy.

This statement sums up what I'm thinking about.  Who cares if a loss is lopsided?  This isn't college.  Play to give yourself a chance to win.  If you don't win, OK.

Bookies care! Bookies lives matter too!

Sometimes punting GIVES you the best chance to win. Doesn't mean you will win or that if you don't you will lose, but it may give you the best chance. Your statement "Play to give yourself a chance to win" is inconsistent with your desire to go for it on 4th down more often in my opinion and in the opinion of many coaches. You're absolutely right that it's not college and it's also not high school and it's not Madden either and going for it on 4th down is not a sure fire way of winning games in the NFL. The odds are against you. The only time the odds are with you is if it's 4th and short (like 2 yards or less) or you have a better then average offensive team. Your statement should be more like "Be aggressive regardless of whether it's prudent or not".

I don't agree that the odds are against me.  The odds are very much for me, as evidenced by the chart I showed, if you believe that the data is legit.

That's not what the chart indicates. The chart compares the relative advantages/disadvantages of attempting a 1st down on 4th down as opposed to kicking a FG or punting. Obviously since getting a 1st down and keeping possession is better then attempting a FG or punting then you are willing to risk lower odds to gain that advantage. That doesn't mean the odds are with you it simply means the odds of going for it on 4th down may not be much worse then the odds of converting a FG especially the further you are out, the risk is worth the reward. That's why it spikes up in the middle because it means that punting offers very little advantage to simply going for it on 4th down once you are out of FG range. If you don't convert the 4th down all you have lost is a little field position. But the further back you go then punting becomes more advantageous because of the difference in field position. That doesn't mean the chances of getting a 1st down on 4th down are better based on the field position, the chances of going for it on 4th down and converting it are essentially the same EVERYWHERE on the field. I believe you said the league average for 2017 was around 50% for 4th and 4 which means that approximately half the teams have a better chance of failing to convert a 4th and 4 then converting it. The chart shows the relative advantage/disadvantage that is gained/lost by either attempting the 4th down, attempting the FG or punting. The odds play a factor, but it doesn't mean the odds are with you. As I already said for the most part the odds are only with you if it's 4th and short or you are a better then average offensive team. Obviously things like the defense you are playing are also a factor in your odds of making it so you would have to adjust this chart based on your own team's relative success of making a 4th down and the team you are playing against. I believe that is exactly what most coaches do.

Below is the chart that shows your chances of making the 1st down on 4th down which is about 50% league wide at 4th and 4. Under 4 yards to go it's a little better than 50% but not much, it's not until you get to around 2 yards that your chances of making it are about 60% but remember this is league wide so you have to determine if your team is above or below league average and if your opponent is above or below league average to determine if your chances are better then 50/50 over about 2 yards. Now these stats are from 2000-2008 so perhaps it's gotten a little better in the last 10 years as offenses generally speaking have more advantage, but probably not by a large amount. Also note that since this analysis is from 2000-2008 when the starting position after a kickoff was probably about 5 yards less then it is today since the touchback now moves the ball to the 25 that would also change the go for it/fg/punt chart as starting position for your opponent would change and thus changes the expected points calculation downward making going for it less beneficial. Also the expected points calculation is gonna be largely effected by the extra point being moved back to where it is as it was pretty much a foregone conclusion during the time of this study and it's a lot different now. All things considered the chart above and below are probably still pretty valid.

(http://farm3.static.flickr.com/2612/3696016235_5df243c99b_o.png)

Agree if option A has a 99% chance you will lose by 13 and a 1% you will win by 1 vs B which give you a zero % chance of winning but you will only lose by 6.  Take option A.

This is silly and meaningless. Whatever gives you the best chance to win is what you take regardless. If option A gives you a 1% chance to win and option B gives you a 2% chance to win you take option B. I never said that the Dolphins shouldn't have gone for it on 4th down because they ended up losing by more points, I simply said that going for it on 4th down didn't help them win. It didn't help the Giants win and it didn't help the Cowboys win either.