The Anatomy of a Three-and-Out

[ckparrothead]

I did a little research and found the following.

We have had a total of 30 three-and-outs so far, averaging about 3 or 4 of them per game.

Analyzing those three-and-outs leaves us with the following.

1. Only 4 of the 30 were effected by penalties, curiously enough. Logically speaking, our penalties must have hurt us mostly on drives in which we were driving, and stalled because of penalties.

2. We passed the ball 62 times on those three-and-outs, ran the ball 28 times.

3. We ran the ball on first down 14 out of the 30 times. Conversely, of course, we passed 16 out of 30 first downs.

4. We passed attempted passes two out of the three plays on 18 of the 30 three-and-outs, and attempted passes three out of three plays on 8 of the 30 three-and-outs.

5. We ran the ball two out of three times on only 2 of the instances, and ran three out of three times on only 2 instances (both in the 4th Quarter against the Broncos as we run out the clock).

6. We passed the ball on third down 27 out of the 30 instances, with two of those three remaining 3rd down runs being on clock-killing drives against the Broncos.

In other words, the three and out for us has become a predictable run-pass-pass, pass-run-pass, or pass-pass-pass scenario.

That doesn't mean every time we go run-pass-pass or pass-run-pass we go three and out, just that when we do go three and out, that is how we most often did it.

My own personal conclusion is that we would have a few less three-and-outs, and would have a better time of possession, if we got it to the point where at least SOME of our three-and-outs were two runs and a pass rather than it almost always coming down to two errant Frerotte throws.

 

[guido13]

Try telling that to pass happy Linehan.

 

[Schleprock]

Uhmm, there are only 6 different combinations for 3 plays. You stated 3 of them, or 50% of that, as the problem for 3 and outs. Statistically, your research means absolutely nothing.

 

[ckparrothead]

Uhmm, there are only 6 different combinations for 3 plays. You stated 3 of them, or 50% of that, as the problem for 3 and outs. Statistically, your research means absolutely nothing.

Dude, trying to sound smart does you no good if there's no content behind your message.

The main point is out 30 three-and-outs we have passed the ball on AT LEAST two of the three plays 26 of the 30 times. We have passed the ball on third down 27 of the 30 times.

Statistically, it doesn't appear you know what you are talking about.

 

[Superself]

Uhmm, there are only 6 different combinations for 3 plays. You stated 3 of them, or 50% of that, as the problem for 3 and outs. Statistically, your research means absolutely nothing.

:confused:

Not sure what your point was here.

 

[UberCane69]

His point is that most of the time, you can attribute the 3 and out to:

Run - pass - pass

Most teams know we're passing on third down (27 out of 30 times).

And if pass twice, that is not good. lol.

 

[Schleprock]

Dude, trying to sound smart does you no good if there's no content behind your message.

The main point is out 30 three-and-outs we have passed the ball on AT LEAST two of the three plays 26 of the 30 times. We have passed the ball on third down 27 of the 30 times.

Statistically, it doesn't appear you know what you are talking about.

It doesn't matter what the outcomes are, my stats are still correct. With each and every team in the NFL you have 6 different sets of how you are going to run the plays (taking into account you are punting on 4th down or have acquired a new set of downs on your 3rd try).

Your 26/30 and 27/30 mean nothing when you can't show the converse of that. How many times have we been successful on sets that include AT LEAST 2 passes. And how many are successful on sets that include AT LEAST 2 runs.

Sure, at face value your theory seems correct. However, if you were to find out that of the total NON 3 and outs we have gone on, that 50%+ included a majority of passes, then your theory means nothing. Because we are beign successful as much, or more, than we are being unsuccessful. When only looking at one side it may seem like the 2 passes are what are causing the trouble, but if we have 30+ drives of more than 3 plays that include a majority of passing, then we are living by the sword, dying by the sword. In fact, the league as a whole runs the ball about 45% of the time, passes 55% (obviously). Not having the numbers in front of me, I'll wager that we are equal to, or higher than the 55% average passing in the NFL.

Also, I bet if you pool the league info, you will find that an overwhelming majority of 3rd down plays are a pass.

 

[ckparrothead]

You're reading your own conclusions into my data. I presented the data AS IS, while you have attempted to put statistical conclusions in my mouth. I did not say that this was evidence that when we pass the ball two out of the first three plays of a drive we always go three-and-out, you said that.

My own personal conclusions, not based on statistical truths but on a combination of statistical and visual observation, are that if we made a deeper commitment to running the ball to where our three-and-outs consist about equally of 2 run attempts and a pass vs. 2 pass attempts and a run, we would probably see less three and outs.

The statistics posted do not mean nothing, they just don't mean what you are trying to make it sound like I am attempting to say. And no, the statistics are NOT in line with the rest of the league. Right now we're seeing just about the polar opposite of what the same study would have shown under Dave Wannstedt in 2000, 2002, and 2003.

 

[ccphinfan]

You're reading your own conclusions into my data. I presented the data AS IS, while you have attempted to put statistical conclusions in my mouth. I did not say that this was evidence that when we pass the ball two out of the first three plays of a drive we always go three-and-out, you said that.

My own personal conclusions, not based on statistical truths but on a combination of statistical and visual observation, are that if we made a deeper commitment to running the ball to where our three-and-outs consist about equally of 2 run attempts and a pass vs. 2 pass attempts and a run, we would probably see less three and outs.

The statistics posted do not mean nothing, they just don't mean what you are trying to make it sound like I am attempting to say. And no, the statistics are NOT in line with the rest of the league. Right now we're seeing just about the polar opposite of what the same study would have shown under Dave Wannstedt in 2000, 2002, and 2003.

RUN the ball?? Blasphemy...I'm telling linehan.

 

[ckparrothead]

And by the way, year-to-date the Dolphins have dropped back to pass 327 times and run the ball 230 times (which includes 14 runs by Gus Frerotte which were nearly all scrambles, not designed runs).

That's passing the ball 59% of the time if you don't count the Frerotte runs as dropbacks (which they were), and 61% of the time if you count the Frerotte runs as dropbacks.

 

[Schleprock]

You're reading your own conclusions into my data. I presented the data AS IS, while you have attempted to put statistical conclusions in my mouth. I did not say that this was evidence that when we pass the ball two out of the first three plays of a drive we always go three-and-out, you said that.

My own personal conclusions, not based on statistical truths but on a combination of statistical and visual observation, are that if we made a deeper commitment to running the ball to where our three-and-outs consist about equally of 2 run attempts and a pass vs. 2 pass attempts and a run, we would probably see less three and outs.

The statistics posted do not mean nothing, they just don't mean what you are trying to make it sound like I am attempting to say. And no, the statistics are NOT in line with the rest of the league. Right now we're seeing just about the polar opposite of what the same study would have shown under Dave Wannstedt in 2000, 2002, and 2003.

Do you have any relationship to Bill Clinton?

If you just want to believe what you are saying because you see that 26/30 3 & Outs include 2 passes, that is fine. But if you are trying to pose a thought here for anyone to logically consider as holding any water, then you need to find out how many times we have had more than 3 plays and the majority of plays were passes.

I'm sorry to make this so scientific, but after all, we are here to discuss football and reasoning, and although you are trying to find a conclusion to something you are missing the other half of the pie. Perhaps you are right, perhaps you are wrong. But if we have had, say, 50 drives of more than 3 plays and 50% or more of the plays within those drives were passes...then your theory isn't very viable, it's just happenstance the 3 and Outs occur that way. Or it's logical, because on a whole we pass more than we run.

And yes, without looking I'd wager that we ARE near the NFL league average of a 55-45 pass-run playcalling standpoint. Infact, I bet we are on the higher end of the distribution, around 60-40 pass-run. (EDIT: I was right, you posted the #'s while I was typing)

 

[finfan54]

Well, let me say that I am just happy someone is looking at actual figures and turning the subject around. Thanks CK!

For me, its not neccessarily run-pass-run yadda yadda yadda or anything else for that matter. Its more of the selection of plays at certain times. Things like the end around when you just ran the ball for 7 yards on first down. Why? just run it down there throat again! Run it til they stop it! Secondly, the cutsie plays got to go. They did for NE, I hope it stays that way! Third, its not that hard to figure out what someone called a few weeks ago as the "pulse" of the team. When you are contolling them at the line of scrimmage and running on them regardless of what they do, you just keep running it! Why stop doing whats working!? If the pass is working, by all means, we pass!

my contention is that trick plays suck or should be used sparingly in moments when you need a big play and they least expect it. It should not be a part of your forte'. The other thing is that we run good plays on third and long that fall short when it should be the first of second down play to begin with. The mix is just off sometimes.

All that said, Linehan IMO, did much better against NE and I bet it was planned as much against them from the playcalling the week before. But get a clue, when something is working, just keep running it til they stop It!

 

[ckparrothead]

Again, I am just posting the statistics up for consideration. You can read what you want to read into them, and no I don't have to research every down in football to come up with a statistical conclusion. That's like saying that someone who quotes a quarterback's third down completion rate isn't quoting anything that holds any water because he didn't research how many of those incompletions were dropped passes and how many of the completed passes went for first downs.

All the person was quoting was third down completion percentage, and you'd be trying to tell him that he's trying to give a 100% accurate measure of the outcomes of each and every decision made by the QB on third down.

You are very normative in your thinking, which is why I found it a little funny that you used the words "I'm sorry to make this so scientific." You seem to have a lot of trouble evaluating data on an objective basis and separating out your normative conclusions from the data itself, in order to extrapolate what the data means.

How old are you? Do you work in the field of statistics? Because honestly, it does not sound like you do.

 

[Schleprock]

Again, I am just posting the statistics up for consideration. You can read what you want to read into them, and no I don't have to research every down in football to come up with a statistical conclusion. That's like saying that someone who quotes a quarterback's third down completion rate isn't quoting anything that holds any water because he didn't research how many of those incompletions were dropped passes and how many of the completed passes went for first downs.

All the person was quoting was third down completion percentage, and you'd be trying to tell him that he's trying to give a 100% accurate measure of the outcomes of each and every decision made by the QB on third down.

You are very normative in your thinking, which is why I found it a little funny that you used the words "I'm sorry to make this so scientific." You seem to have a lot of trouble evaluating data on an objective basis and separating out your normative conclusions from the data itself, in order to extrapolate what the data means.

How old are you? Do you work in the field of statistics? Because honestly, it does not sound like you do.

YES, YOU ARE POSTING STATS...but not all of them. It's like me saying that 19/20 car accidents happened at 80MPH, therefore it must be that the speed of 80MPH must be the problem. Meanwhile, no one found/stated that the drivers consistently drove 80MPH all the other times (lets just say 99/100 to help to conceptualize).

Normative thinker? Do you even know what you are talking about? A normative thinker is one who sees things only "inside the box". According to this situation, that seems like you, not I, since you are not willing to accept the fact that all I am saying is that you need to evaluate what we do on "successful" situations in addition to "unsuccessful". Just because you start by stating a 26/30 number doesn't mean you are non-Normative. I'm a bit disappointed you didn't cover the "successful" drives right off the bat as a simple control for your theory.

#1 rule in science of someone grasping at straws and getting personal = calling out age, education, occupation, followed by condescending remark when not even knowing those answers. But then again, those type of people are the reason our dissertation commitees grill us and how conferences get entertainment.

This was a good thread with a good theory, still needs more info though. Therefore I will bow out, I've said my piece and those understanding enough will at least accept its legitimacy I believe.

 

[Dol-Fan Dupree]

#1 rule in science of someone grasping at straws and getting personal = calling out age, education, occupation, followed by condescending remark when not even knowing those answers. But then again, those type of people are the reason our dissertation commitees grill us and how conferences get entertainment.

That is something I can agree with