Separate names with a comma.
Discussion in 'Miami Dolphins Forum' started by bbqpitlover, Oct 16, 2019.
Are you being intentionally obtuse?
What is the significance of that in terms of the current topic?
5-1 > 0-6 = GOOD!!!
The 2012 Washington Redskins were 6-0 in their final six games of the season with RGIII at quarterback. Is he therefore a good one?
Some of us have, for years, explained WHY Thill has had issues.
Some have countered that with a stats based argument, which always come down to, the issues we claim are statistically not likely.
We've countered that counter a few ways:
- The statistical argument doesn't actually address the issues we raised.
- The purpose of stats are as part of a guide, not THE explanation. Knowledge of the subject and observation should be used in conjunction with stats.
- Unlikely doesn't mean it didn't happen.
The last item is why the "unlikely things happen all the time" argument is significant.
Now, instead of ignoring the questions I've asked like 100 times now, please address the following:
1. What are the ways a QB/team can counter a pass rush?
2. What modern era QB had a terrible line, but was also not allowed to audible and had no commitment to the run game, but was successful?
There's no question Tannehill has played well since he became a starter in Tennessee. The question is whether he can sustain a similar level of play to the degree that it becomes distinctive enough from what he's done previously in his career, because what he's done previously in his career is merely average.
Think about it like this: if Tannehill's first play with Tennessee was a very difficult 99-yard touchdown pass thrown 60 yards in the air with pinpoint accuracy, would we exclaim that he was performing meaningfully differently with Tennessee than he was with Miami? Of course not, because despite the brilliance of the play, it's nonetheless a mere single play.
The point however is that he hasn't done enough more than a single play to warrant any different a conclusion at this point. Sure he's played six games, and that's a lot more than a single play, but it isn't yet enough to distinguish what he's done recently from what he's done previously in his career.
The overall difference of opinion in this area stems from a difference of opinion regarding the surroundings Tannehill has experienced. Again, nobody has measured those surroundings and determined how much, if at all, they deviate from the league average.
You apparently believe that Tannehill experienced surroundings that were highly deviant from the league average, in the negative direction. Even if we were to agree that Tannehill had a terrible line, wasn't allowed to audible, and the team had no commitment to the run game (all of which are debatable), we still don't know the degree to which that made his overall surroundings (negative and positive, combined) deviate from the league average, if at all.
You're arguing for placing a greater emphasis on context, while at the same time neglecting the context of the entire rest of the league, whose quarterbacks' surroundings give Tannehill's meaning. If for example Russell Wilson's overall surroundings have been just as bad as Tannehill's (hypothetically), and Wilson has nonetheless played significantly better than Tannehill, what does that tell us?
The point isn't that we know Wilson had such surroundings (again the above was only hypothetical). The point is that we don't know, and that we don't know what any the rest of the league's quarterbacks' surroundings have been, and how Tannehill's compare to them. That entire area of observation is being neglected here. We just assume Tannehill had it really bad compared to the average QB, without really knowing.
The bold is not debatable. It is fact. it has been confirmed numerous times now.
If you keep ignoring my questions, then you prove you are posting in this discussion in bad faith.
You just ignored the 95% of my post that prevents us from ascribing any meaning to the answers to those questions.
Again, to what degree did Tannehill's overall surroundings -- positive and negative combined -- deviate from the league norm?
If we can't answer that -- and we can't -- then we can't determine the meaning of the answers to your questions.
Sure, let's pretend you haven't literally been ignoring those question since yesterday....
My question #2 is asking you exactly that, btw.
This is why things get heated, ftr.
I've ignored the questions because of what I've described above. If we're going to attribute the difference in performance between Tannehill and other quarterbacks to Tannehill's surroundings, how can we do that with any reliability without also knowing the difference between Tannehill's surroundings and those of the other QBs in the comparison?
Where have you seen any post in this forum that attempts to address systematically the degree to which Tannehill's surroundings in Miami -- again both positive and negative combined -- deviated from the league norm?
I could ask a similar question: what QB had the benefit of playing in his college offensive system with his college coach calling his plays during his first two seasons in the NFL?
This is why it becomes important to know the degree to which Tannehill's surroundings -- positive and negative combined -- deviated from the league norm.
You're believing that the answer to your question has all the meaning, whereas what really has the definitive meaning is the degree to which Tannehill's overall surroundings, positive and negative combined, deviated from the league norm.
What if no other QB has experienced playing in his college offensive system with his college coach calling his plays during his first two years in the league? Why does the answer to your question have more meaning than the answer to that one?
wow, did Tannehill piss in your cereal?
There's nothing that says playing for your college coach in the pros means anything.
What are the ways a QB/team can deal with the pass rush?
Answer that question or stop engaging me.
I didn't change the question. The Patriots are 10-1 with the league's 22nd best passer. Do we trust passer rating or win percentage to determine the likelihood of a NE win tomorrow?
Of course you changed the question. The evidence is in aqua and white. Start with post #701 where you said:
I responded to that in post #703 with (quoting just the relevant stuff):
To which you responded in post #705 (quoting the relevant parts):
So.. up till there you are repeatedly arguing that it's better to look at the end result – wins or losses – to estimate QB ability because wins are ultimately all that count. And that's when you changed the question to something I did not ask or imply (same post #705):
When did I ever say passer rating was a better measure for predicting wins? I said it was a better measure for estimating QB ability.
But as I said before, it's fine if you think win% or the "end result" (e.g., SB win) is better for measuring QB ability, just don't change that measure if Tannehill starts losing games, especially since Tennessee controls their own fate: they win out they win the division.
Tannehill had 3 head coaches and just as many or more OCs in 7 years. He had a new offense like every other year. Those are some disadvantages that are commonplace in the NFL?
These are the things we are taking shirt. When you act like Tannehill wasn't in a terrible situation here, or somehow didn't have it worse than average, it makes people upset because there are obvious ways that he was disadvantaged, far beyond what is normal.
We could trust either one, because the Patriots currently have the best passer rating differential in the league, 2.16 standard deviations above the current league average.
The correlation between passer rating differential and win percentage in the league this year? 0.89. Nearly 80% of the variance in win percentage in the league in 2019 is associated with passer rating differential.
The Titans' passer rating surrendered this year is 89.5, which means their pass defense is nothing special, and so that means Tannehill will keep helping the Titans win only to the degree to which he continues posting passer ratings similar to the ones he has recently.
Ironically, you're arguing for the meaninglessness of passer rating at the same time that the team whose winning you're touting as evidence of that is winning by having one of the best offensive passer ratings in the league during that period.
I didn't say Tannehill wasn't disadvantaged beyond what's normal. I said we don't know the degree to which he was, if at all.
If you're that upset about it, then put that energy into studying the league as a whole and coming up with some quantitative representation of how disadvantaged Tannehill was in comparison to the average team. If it's so obvious, it shouldn't be that hard to do.
What you shouldn't do is browbeat people into believing the same thing you do without any empirical evidence of it, with the threat that you'll "get upset" if they don't. Do the work and support your point with something convincing.
It's meaningless as a predictor because IT'S A RESULT. And that result is heavily affected by things outside of the QB. The whole argument is not about passer rating, it's about your application of that stat.
Again, many words to obfuscate your lack of answering a simple statement. I'll ask again:
Is it normal for a QB to have 3 HCs, was it 5 OCs, in 7 years? Surely you don't need a scientific case study to answer that.
Again you're taking my response out of the context in which it was offered. The question was, which should we trust more, passer rating or win percentage, in predicting winning. Surely you'd consider win percentage a "result" as well?
I was responding to the last paragraph.
Win% is at least actually factoring in all the units contributions.
Is it normal to have one's college coach calling the plays within one's college offensive system during one's first two years in the NFL?
Win percentage is what the guy above is trying to predict. The question he asked is whether we should predict win percentage with win percentage itself, or with passer rating.
Ironically, you do about as well using passer rating, despite the sarcasm of his question.
Quick someone do a quantitative study on how many people it takes to complete a pass. Since no quantitative study has been done yet, how can we possibly know the answer!!!!!!!!!!
EVERYTHING IS A MYSTERY!!!!!!!!!
I do recognize the historical significance to determine where Marino ranks against Unitas, etc...and I'm fine with that. But you also made a comment that passer rating had a high correlation to success, and I'd assume success = win. After all, if we're not somehow measuring for the ability to win a football game, then passer rating is essentially meaningless FOR THE CURRENT SEASON.
I think that's where we're having a disconnect- you're trying to solve "how good is a QB" while I'm saying, "if he's good enough, his team will win." Maybe you didn't directly say it, but you alluded to it by saying Tannehill will likely regress later this season. And hey, he might...we've seen it happen enough over the years. But if you're using passer rating from prior seasons as evidence then giving your opinion on a future prediction, those two things don't belong in the same paragraph.
You've said it yourself- what Tannehill has done in the last five games has essentially no bearing on what he'll do tomorrow vs Indy. He could play much better, much worse or have an average game...we have absolutely no idea. And if passer rating can't give us a single clue what to expect, then why are we talking about it mid-season? After all, we have about eight pages here explaining what passer rating is telling us about RT...yet it really doesn't tell us anything at all that we didn't already know- he had a couple of really good games.
But let's take that a step further...there's five games left in the season. Let's pretend RT has a perfect passer rating in all five contests. Then what? Folks will undoubtedly say, "Well, let's see how he does in the playoffs...then we can judge him." And if he's a Super Bowl winner, then we'll wait for next year to form another conclusion. Even if you don't say it, this conversation will essentially come down to wins only...at no point in 2019 will the raw stats matter in this argument.
In other words, we'll never have a complete answer that everyone here agrees to unless RT has season after season of continued growth. That's why I say passer rating doesn't matter at all in 2019 while the season is unfolding...there's simply nothing there to learn from it that's actually meaningful.
See post #680. When his season-by-season passer rating varies at a level that puts him in the group of quarterbacks that’s significantly different from the one he’s in now, that’s when he’ll be a different quarterback.
Until then we won’t know if he’s simply experiencing an Andy Dalton 2015 season. Andy Dalton played at an elite level for one season, but we certainly wouldn’t designate him an elite quarterback.
Titans have a real shot at winning their division.
In my opinion with RT they are the best team in the South anyways, they get division leader Houston twice and the Raiders (tied for WC) at home.
If they win today at IND which isn't easy then watch out....the most polarized QB in TP.Com message board history will be playing January football for the first time in his career.
He's looked really good in that offense. Looks fresh and renewed. They seem to know how to play to his strengths.
So is he just a broken clock that has the right time twice a day, or is he actually a good QB? Was the issue in Miami the coaching staff? Maybe a lack of talent put on the field? Opinions.
As it relates to Miami his injury really derailed his career. However, he's more of a game manager with an arm that dreams are made of.. combine that amazing arm talent with a proven and sustained running game and a half decent team around him and you have his success. he played his best football when Ajayi was dominating.
I don't think he can overcome his pocket presence and quick twitch reaction time which will prevent him from leading a team deep in the playoffs. But overall it's not a black eye on Miami because a change was needed and his injuries piled up.
People always bring up the name Rich Ganon but I wonder how many other fanbases have thought the same thing about middling QBs. Let's see how he finishes the season. If he can shred the Colts today he may be on a upward trend. Or he may go back to being the Ryan Tannehill on the road late against a division rival that we've come to love.
Excellent points. Especially regarding Ganon. He dink-and-dunked his way (with a good surrounding cast) all the way to the Super Bowl. I don't think Tannehill is gonna win games by himself. You definitely don't want to be behind in a game and think he could lead you back to victory. But they seem to have found a way to make him work. Proof of the importance in coaching. His surrounding cast in Tenn is better than what he had in Miami, tho. That sure helps.
This post should be moved to the other NFL forum.
Pointless to continue this debate after 7 years of it when he isn't even on the ****ing team.
Yup, you posted this^ in the other thread about the exact same topic.
Look at you getting your hypocrite all over the goddamned place again.
Wrong again. The post was about Miami's use of him. Unlike you I'm not going to spend the next 3 months discussing if he was allowed to audible.
My judgment on Tannehill will be purely stats based. And there are statistical tests that take into account sample size. I've already shown that up till now (using Tannehill's ratings so far, not just this year but in prior years) what he's done this year is not statistically significant. If and when it becomes statistically significant I'll say so. So Tannehill's passer ratings so far are necessary if you want to determine whether what he's doing so far is statistically significant. They're clearly not meaningless.
And all that's based on performance so far. None of it requires prediction. That's where you're getting confused. Prediction is never part of the equation here. You CAN use the stats to make predictions, and yes it's much more likely Tannehill is NOT doing something special if you just go by stats (which of course remove contextual information), but that kind of prediction is irrelevant for determining whether performance so far is statistically significant.
Finally.. to measure the ability of any player on a team you want to remove as many confounding variables as possible. So obviously passer rating is better than win% for measuring QB ability because the QB is part of both, but less influential on win%.
The fact that you posed the first question as a question is apropos, because statistically speaking we don't yet know.
I think the shape his performance takes over the rest of this season and the next couple will give us insight into what was happening in Miami and whether it was different from what he's experiencing in Tennessee. Unfortunately we have to wait for that to occur in my opinion before we can derive any reliable conclusions about it.
Based on Tannehill's performance in Miami, you can win with his playing that sort of "Alex Smith" game, which consists of comparatively lower numbers of pass attempts and an extremely good INT percentage, if you have one of the best pass defenses in the league and you can therefore help Tannehill keep pace with the league's best quarterbacks he will no doubt face in the playoffs.
Again, passer rating differential. If your QB is an Alex Smith equivalent, then your pass defense needs to bring the opposing QB down to at least the Alex Smith range to give your team a good chance to win. If your pass defense lets the Patrick Mahomeses and Tom Bradys of the world have their typical performances, then your Alex Smith/Ryan Tannehill likely gets beaten handily.
One difference in how he's being used that fits with the above is the number of pass attempts he's made. His average number of pass attempts per game in Miami was 33.4. During the past five games that number is 27.6, 0.73 standard deviations lower.
This Titans team though NEEDS a Qb like Tannehill who can make throws when needed to extend drives. He is also slowly developing some Chemistry with AJ Brown, that will change the coverages they see, and Derrick Henry can stop seeing loaded boxes.
Things like that are not really measurable through numbers, what he brings to that offense is hard to quantify. But being a degenerate who watches literally every game (almost) each week, you can see the difference in Ryan and Marcus.
I don't even know his official statistics, but Ryan went from 2nd and 9, 3rd and 8 to 2nd and 4, 3rd and 2's. Things like that have made all the difference in the world for HIM, and THEM.
Did I miss the % somewhere?
Again as stated before I am looking at the games as their own individual data point. What I'm asking for is the percentage of 1 in 5 above, 2 in 5 above, 3 in 5 above, 4 in 5 above, and 5 in 5 above.
I get that I gave you what amounted to a 50% for an average QB using the means of average. I would assume this would mean the most likely scenarios are going to be 2 and 3 above average games. followed by 1 above and 4 above, and coming into the last portion 0 and 5 above. Standard bell curve stuff I would assume, by that standard deviation 4 above average can't be more then 25%(it actually can't be 25% in standard deviation model).
What I'm saying is that if that 4 above is a 25% likely hood, then you are claiming no significant % difference in a 4 in 5 above average game stretch, with an average passer rating of 93 to the 4 in 5 above average with an average passer rating of 111.
If I was in Vegas and I was told that I could select any Average QB and make a bet that he will have 4 out of the next 5 games above average, and pull a minimum QB average of 111 rating, I would bet against that happening 100% of the time. and take my 25% payout.
IF Tannehill was still here, and some Tannehill fan said " I believe Tannehill is going to be above average at least 4 out of the next 5 games, and post a average rating of 111 rating, I doubt you would be congratulating them on their good bet, claiming that not only is it a pretty good possibility, but it actually has some of the highest possible odds being 25%, leaving room for only 3 other scenarios likely to occur with that high of a probability.