Dolphins’ QB Josh Rosen solid in first start with Miami

SOS has some thoughts on Rosen:

 

Certainly, the most fair & accurate analysis to date.
Of course, those cult-like believers of the skinny version of Tim Tebow will disavow the analysis.

 

Watching the video, Rosen could work on selling the run a little more in those play actions, because it looked very much like Marino's efforts.

And the phantom offsides which was flagged continues to add to the evidence that games are being influenced to ensure preferred outcomes.

 

If we actually made a concerted effort to run the ball and not give up on it, it would make play action pass more effective.

 

It has been shown that PA is not dependent on actually running the ball.

 

That's correct, though for future reference I think it's important to separate two periods: pre-1978 and post-1978. The rule changes in 1978 dramatically changed the nature of the game, and what you said applies much more to post-1978 than to pre-1978.

From 1966-1977 the correlation between rushing attempts and passing Y/A is 0.3662, which means that 13.4% of the variation in passing Y/A can be explained by rushing attempts. However, from 1978-2018 that correlation drops to 0.1724 which means that only 2.97% of the variation in passing Y/A is explained by rushing attempts.

And both of those numbers are overestimating the actual influence of rushing attempts on passing Y/A because it includes the 4th quarter where the leading team tends to have more rushing attempts AND higher passing Y/A without play action.

So at maximum, since 1978 you can expect to influence about 3% of your ability to pass the ball efficiently through rushing attempts.

Actually, this is a fairly well known result among those who've looked at football statistics. FO probably has the best summary graph but only for the years 2011-2013 (and based on what I just described their results should apply to post-1978):



FO has their own definition of "rush success" which one can quibble with (they set an arbitrary threshold) so if you want you can ignore 3 of those graphs, but the other 3 graphs (top left, top right and bottom middle) should lay to rest the idea that rushing attempts – even rushing attempts designed to decrease predictability – increases passing Y/A in the modern game.

 

In other words, if after reading this you believe that rushing attempts positively influence the passing game in today’s NFL, you are delusional.

 

I still hold, though, that while rushing doesn't really move the needle much, NOT rushing will negatively impact you. Being one dimensional is a killer. So, you're not rushing because you are trying to get 150 yards on the ground, you're simply rushing to keep the defense from just sitting back with 7 or 8 guys in coverage, which makes passing incredibly difficult.

 

Yeah I think I would phrase this differently.

Clearly, "predictability" affects efficiency of any play, rushing or passing:
https://www.thephins.com/threads/wi...-quarterbacks-do-we.94622/page-3#post-3203469

So you have to rush at least a certain amount or you'll become too predictable. In that sense rushing attempts matter. However, what that graph shows is that it is VERY difficult to decrease predictability in the near future by adjusting the number of rushing attempts.

Decreasing predictability is .. predictably .. hard for humans as tons of psychology experiments show (one classic experiment showed that an algorithm implementing Bayesian inference – the mathematically optimal method for adjusting expected probabilities given past data – was capable of predicting the next number in sequence of 1's and 0's "randomly" generated by humans at much greater than chance level).

I've been arguing for a long time that one very obvious application of statistics in football that should improve win% at least by some marginal amount is to use a random number generator to decide which of various plays to call. The coach should decide the set of applicable plays, but the machine should choose the play. I guarantee you that decreases predictability.

 

Absolutely. I mean, I know football isn't Madden...but think about the human aspect of choosing plays. Humans will always develop favorite plays and just naturally gravitate to a certain style or pattern. When I used to play Madden a ton, I realized with all the hundreds of plays available, I would use only maybe 30 consistently. Then Madden bright in the option to allow the cpu to call they plays for you, or maybe just give you the options to choose from, I forget which exactly, but it really improved my offense in the game because I was so much more unpredictable.

 

3 of the final 4 teams in the NFL last season were top 5 in YPA between the tackles (I.E. smash mouth football) no accident.

Rushing matters and matters big time.

 

That may be true, but it's also true that the correlation to win% isn't moved much at all by successful rushing.

Again, this is intuitive. One would assume that 4 teams playing to get into the Super Bowl would be the teams with the best offenses. As such, I expect to see great passing and good running. However, that is far different than asserting that the effective rushing stats are what made them the final 4 teams. Since we can look at how win % is affected by rushing, we should be able to agree that rushing well barely affects win%. Like I said though, in the post you quoted, not rushing most certainly moves the needle towards losing, so not rushing would, I'm guessing as I haven't done the math, maybe cbrad could show the correlation of not rushing to losing%??, correlate to losing. So, what we see is rushing is effective in that it keys you more balanced and prevents you from losing, instead of it driving you to winning, which is what many fans assume, erroneously, that it does.

 

Do you have a link for rushing Y/C "between the tackles"? I've never looked at those data.

If you don't discriminate between types of rushing, the correlation between rushing Y/C and win% across NFL history is 0.1613, which means that only 0.1613^2 = 2.6% of the variation in win% is explained by rushing efficiency. In other words, not much.

The other thing to point out is the relative importance of rushing Y/C and passing Y/A because you have to choose between the two:
https://www.thephins.com/threads/wi...-quarterbacks-do-we.94622/page-3#post-3203648

That equation there summarizes the relative importance: Win% = 13.81*Passing Y/A + 2.53*Rushing Y/C - 55.57

The fact that 2.53 is positive confirms what you're saying that greater rushing efficiency helps, but the ratio 13.81/2.53 = 5.46 confirms that passing efficiency is much more important in the modern game, and by about 5.5 times over rushing efficiency.

If you're talking about how much more you'd expect a team to lose if they went below the "minimum" necessary to keep play calling somewhat unpredictable, then I don't think we have the data to calculate that correlation because there's evidence few if any teams are rushing below that minimum level.

What's the evidence? Remember those graphs I posted that plot rush% as a function of both point differential and quarter? There was very little difference in rush% between winning and losing teams in quarters 1-3. What that tells you (among other things) is that winning and losing teams in quarters 1-3 don't differ much in how predictable their play calling is. And if that's true, we don't have the data to figure out what happens if you go too low on rush%. We also can't use data from the 4th quarter because the cause-effect relationship is pretty clear there: point differential => killing the clock, not the other way around.

 

That's definitely the more precise phrasing to use, but as you said in another post above, we'd struggle to find any teams that weren't running the ball at the minimum ratio necessary.

So anybody who is cautioning teams "not to abandon the run," in terms of the relevant statistical data that actually have meaning, is cautioning them not to do something teams rarely if ever do (run the ball less than the minimum amount necessary). That caution really isn't necessary, then, and if the traditional beliefs about running the ball persist despite the evidence in post #7 above, they become fairly delusional.

 

Growing pains :/

The whole offense, really.

 

Agree that if you don't even attempt to run then you isolate your passing game and that is not good. But running the ball well will always be a good thing for an offense.

 

One could dismiss that due to small sample size (Rosen has only 60 passing attempts so far) except that Rosen topped the list in 2018 too lol:

Well it's well known he has to work on his accuracy. We'll see if he does better in tomorrow's game.

 

Yeah but it wasn't 40+%!! Lol

 

Hard to evaluate those numbers when your OL has been that bad. Being on the run, not being able to step into your throw and just generally being under duress will impact that.

 

Except that the 2018 list is a virtual who’s who of young QBs in the league, Roethlisberger notwithstanding. Those can’t all be bad offensive lines, and if they are, you have to wonder about the quarterback’s effect on them. The relationship between the quarterback and the line is bidirectional.

 

I’m specifically talking about Rosen. He went from the worst pass blocking OL last year to the worst one this year. Didn’t even look at anyone else on the list.

 

Right, but the company he's in there can be informative. Again you have a virtual who's who of young QBs in the league in the 2018 list, Roethlisberger notwithstanding.

When youth and inexperience is such a strong common thread among those QBs, it suggests that youth and inexperience, and not the functioning of their offensive lines, is responsible for their performance in that regard. What are the odds that that list of nine QBs in 2018 is also a list of nine of the worst offensive lines in the league? Slim.

So, it behooves you to look further into the list than just Rosen. Doing so may be informative and may yield a different conclusion. It's entirely possible that Rosen, and not his offensive lines, is the active ingredient in his performance in that regard. His offensive lines could be among the worst in the league, and he could be entirely responsible for his performance in that regard nonetheless.

Just because a QB has a relatively poor offensive line doesn't mean it explains every negative facet of his performance. Some of that could be on him exclusively, and replacing his line with the best one in the league could result in nothing significantly different in that specific regard.

Again, an extreme analogy just to illustrate the point: throw a middle school QB into the NFL with the worst offensive line in the league, and then replace his offensive line with the best one in the league. Will that alone make him play at an NFL level, or will he continue to display inadequacies that have nothing to do with his offensive line?

 

2nd year with a well below subpar team. New team, new coach, new OC, another horrible o-line, etc. Attempting to use analytics to try and grade/analyze Rosen in his NFL career so far is straight up useless. Good luck figuring out if the receiver and QB were on the same page in diagnosing the coverage, if the receiver ran the right route, which lineman missed his blocking assignment, the overall effectiveness of the playcall made, etc. with a mathematical formula. And concerning Rosen and the very poor teams he has been on, such situations occur at a higher frequency than other even average teams, let alone the top consistent ones, so the comparison examples are pale.

 

Analytics are always weaker with small versus large sample sizes.

 

I’m sure your point is valid but this is one of the worst OL’s we’ve ever fielded and last years AZ OL was a sh** show. Tough to evaluate anyone under these conditions. When 7 guys can’t block 4, you’re not gonna win many football games.

 

Dude.. are you watching this game? Our OL is VERY good so far in the first half. I really hope this is a sign we have a good OL coach.

Regardless.. with sufficient sample size (a full season should be enough for the purpose of comparing Rosen in AZ to Rosen with the Dolphins) stats will show a significant difference if there really is one.

 

I’m in Budapest so no. But I saw the Dallas game and it looks like we’ve given up 3-4 sacks this game. I’m assuming they came back to reality but in the 2nd half?

And yeah, I think he’s getting this group to way overperform the talent level of this group because let’s face it, we have maybe 1 starter on most teams.

 

First half was very good. 2nd half playing like crap. The INT you see by Rosen on the stat sheet was just a pure bone headed throw by Rosen, not even forced. Rosen also took sacks he shouldn't have (twice if I recall) including one on the goal line.

It is what it is. He seems to be good in ideal conditions but he can't be counted on to win a game. Still only game 4.

 

So far he’s essentially a quarterback much like Tannehill, who makes his living against poorer defenses, in non-clutch situations.

Someone who, if the Dolphins ever made the playoffs with him, would likely get beaten handily when facing opposing teams with very good quarterbacks and very good defenses.

Of course that can change in the future, but so far he has displayed nothing greater than that.

 

I honestly thought that was a safety he took... got away with it.. he played terrible in the second half, looked like Joey Harrington..

 

Here's a fun and sad little fact: Rosen's passer rating for today's game was 88.9 which was his highest ever in 15 games he's started. Obviously, passer rating has little to no meaning for a single game, but when you can't even ONCE do better than league average (92.9 in 2018) in 15 tries there is something really wrong. And there was no OL excuse today.

Next game is against the first team we might have a chance against: Washington who is 0-4 right now and after playing the Patriots will probably be 0-5 when we meet them. Washington also has the 2nd worst defense by points allowed per game. If Rosen still can't perform at or above average against the Redskins defense, there's something seriously wrong IMO.

 

3 games in - ready to throw in the towel in a rebuild...no - reconstituting .. no... playing for the 1st Draft pick ...year. EGAD!

 

Just for reference, Ryan Tannehill had 7 90+ rated games in his first 15 starts in 2012.

I defended Tannehill because he played well in spots and thought he could extend those spots to his overall game. I’ve learned my mistake and I’m not giving Rosen that luxury. Unless he stands up and delivers consistently and soon then we don’t want to waste more time with him. Let him be another team’s reclamation project.

 

I was talking with a gentleman yesterday who's a huge Dallas fan. Of course I got the sympathetic smile and eye roll when I told him who my team was. We started discussing offensive lines, present as well as past. Now just to be perfectly clear; I hate the Cowboys, despise them but the one thing I'll give the Cowboys credit for...for years and decades even, they always field a powerhouse offensive line. That dedication to a strong foundation led to the success of Elliott, Prescott, Romo (way overrated), Aikmen, Smith...it goes on and on.

If we would have ever had that powerhouse of an offensive line, the running backs like Ricky Williams (who was already a brilliant talent in his own right), Ronnie Brown, Ajayi, Miller and Drake now would flourish. Quarterbacks such as Henne, Tannehill and Rosen would be household names. But do we build the foundation of a house first in Miami? Not at all. We start building on a sloppy foundation putting up fancy windows and door and roof without making sure we're building on a solid foundation...then complain when the whole damned house falls apart.

If we had a solid offensive line that gave Rosen just the flipping opportunity to do his job as a quarterback...for 4 full quarters, then we'd actually be able to determine whether or not he's the guy for the future. But as I said, we don't do that in Miami. Ever since Shula retired, that was no longer a priority and since Marino's retirement, we keep searching for the NEXT Marino. Well if we don't get back to the absolute fundamental BASICS of football, we'll continue to go after skill player this...and skill player that...and change quarterbacks yet again...and again and again and again and the Dolphins will become the Detroit Lions of the south.

And that's just speaking of the offense...don't even get me started on the defense.

 

I hear ya, but I think we need to consider OL play as generally the same for all when considering QB play. These days they are all feeling the pressure and being made uncomfortable when they drop back. No excuses. Rosen is getting it done or he’s not. Right now he has a long way to go and we are drafting a QB next year, IMO.

 

My issue is you referring to any conditions that Rosen has played in as "ideal."

 

Go talk to a Seattle Seahawks fan about how Russell Wilson has overcome their perennially poor offensive line with his individual ability.