Jesus..... you don't even have that right....... in those games, the lack of effectiveness HAPPENED EARLY IN THE GAME before anyone knew if it was going to be a high volume or low volume passing game. There is no evidence that Tannehill's performance drops as the number of passes in a game grows. In the three losses, Tannehill was MORE effective in the 2nd half (56% success and 8.6 YPA) than the first half (35% success and 8.5 YPA). You have the relationship between passing volume and effectiveness REVERSED.
BTW, one of the articles I linked earlier today detailed a statistical analysis of this very topic for Tannehill from 2012 to 2016. Here are the conclusions: Conclusions. - Tannehill as every other QB in NFL benefits from balance. When Tannehill is playing well he does not need to throw as much. - In the great majority of situations where Tannehill has thrown a lot it has been because the team has fallen behind early. Way before Tannehill threw the ball 30 times. - Tannehill's performance over his five-year career has not been consistent. While there has been an upward trend there is still a questions as to which Tannehill will show up on any Sunday. - But there is no evidence that Tannehill's performance is related to the number of throws he makes. The same is still true.
This is why you cannot be trusted...... Here are the numbers for Brees and Tannehill, down by 9 or more in the 4th quarter: Brees - 44% success 8.0 YPA Tannehill - 62% success 8.7 YPA In both cases I set the filters to the regular season games that each played (7-16 for Tannehill) and (1, 8-16 for Brees). I left off week 2 for Brees because he got hurt and Bridgewater finished the game. I knew you were hiding something when you posted "wasn't significantly different". Whenever the numbers are against Tannehill, you post them. When the are for Tannehill, you use weasel words......
I'll let you attempt to establish the statistical significance, if any, of that difference in performance, given the low number of pass attempts in that highly circumscribed situation (down 9+ points in the fourth quarter in weeks 7 through 17 in 2019). I've done enough mathematical work here, and it's about time I stopped doing math to refute others' positions. They can do some math to support their own positions for once.
That conclusion was based on a correlation between passer rating and pass attempts. I've already gone over at length why a correlation is the wrong choice for measuring that relationship.
What you don't seem to get is that the Titans' overemphasis on the run game (in comparison to other teams) permeated their season during Tannehill's starts. So whatever the situation -- first half, second half, leading, trailing, regular season, playoffs, etc. -- Tennessee nonetheless maintained an overemphasis on the run game in comparison to other teams in those situations, despite that they had the league leader in passer rating at quarterback. So whatever Tannehill did in 2019, he did within that context -- the context of having his team distribute the offensive load well beyond him. You cannot simply divorce his performance from that context. It happened squarely within it. Take the comparison with Drew Brees down 9+ points in the fourth quarter for example. In terms of YPA, Brees's performance wasn't significantly different from Tannehill's in that situation (unless you can establish otherwise mathematically), yet Brees functioned within that situation with 93% of the offensive load on him, whereas Tannehill did so with only 71% of it. This is the key finding here. Tannehill was able to perform well because his load was limited, even in situations in which teams characteristically dial up the loads on their quarterbacks. He played an entirely different season than the average QB in that regard.
Correlation is absolutely the right measure. You're trying to separate "low volume" from "high volume" as if those are separate categories. But where you put the threshold that separates those categories changes the results of the analysis, and that placement is totally subjective. Correlation takes away the need to create such artificial categories. I already showed you that Tannehill was not significantly different from expected in terms of the correlation between passing attempts and adjusted passer rating. You even admitted that you couldn't make your case statistically and that your "low volume" theory is an opinion. Let's not backtrack on that. It's fine to have an opinion, but from a purely statistical point of view you haven't made the case that Tannehill performs worse than expected when he has to pass more.
We did this in post #6197: So if it's true as you said above that a correlation doesn't tell us which of the two kinds of quarterbacks we're dealing with, what measure do you suggest we use to accomplish that?
That post of mine you quoted gives you a hint: you need to show the residual standard errors are different at different points on the scale. For that particular epidemiological study I was referring to, I used sliding windows and weighted regression (weighted regression is regression where you can weight the importance of fitting different data points — in this case the weight was the number of data points in each window) to get mean and standard deviation functions for every data point in the two datasets, thus representing each point as a mean plus some multiple of a standard deviation (eerily similar to the idea of using z-scores to adjust lol). That standard deviation function would be the functional representation of the residual standard errors, and it has a 95% confidence interval. That 95% CI is what one could use to answer your question. Of course, like I said I had thousands of data points to work with. The 95% CI in this case is probably so large you wouldn't get statistical significance anyway.
I don't give a **** about Tannehill at this point. The "crusade" I'm on at this point is all about having a forum in which objective information is actually given the weight it deserves over and above personal opinions. This is simply one such instance in which that is at issue. Watch when cbrad debates someone about whether "defense wins championships." Do you think he really gives a **** about whether defense wins championships? What he cares about, in my view at least, is whether the forum operates on the basis of objective information and not old wives' tales.
So we have your review, an another review and simple logical analysis of the Titans play calling. They all point to the guy having no idea what he is talking about.
Tannehill's passing volume was 2.14 standard deviations below the league norm in 2019, and that overemphasis on the run game permeated their season, across situations in which teams typically turn the ball over to their quarterbacks. You continue to ignore that Tannehill's performance occurred within that context, and act as though he performed as well as he did with a typical offensive load on his shoulders.
Says the guy who's operating within a delusional world in which the offensive load placed on Tannehill can be ignored when determining the degree of difficulty of his performance.
This thread needs to be closed... arguing over the merits of a mediocre ex Dolphin QB for 1000's of posts is folly.
If this were true you'd spend as much time attempting to discredit other QBs and not focus just one Tannehill. You also wouldn't keep changing your arguments as they get shot down. You wouldn't post misleading information. Nobody is discounting your "objective information" (the volume passing stats being the most recent). People take issue with your interpretation of that objective information. You are on a never ending search for proof that Tannehill is not a good QB. That biases everything you do. You never consider alternative explanations for that objective information.
Start a thread on Lamar Jackson's performance in 2019 and the likelihood that it'll be replicated and we'll go another 7,500+ posts if there's a blind Lamar Jackson adherent in the forum who ignores or disagrees with basic facts and perpetuates the discussion far beyond the point that it should end. As I've said previously, when a QB plays at an average level for six seasons and there is no irrefutable objective information that indicates it was due to anything other than him, then the sensible thing to do when he plays well for just one season is to explore the situational advantages he may have been experiencing in that one season, and then determine the likelihood that they will recur. Why do you think Tennessee was able to hang on to Tannehill by paying him something less than the very highest amount a QB could be paid in the league? Why weren't teams banging down his door, offering to make him the highest-paid QB in the league? If 2019 is truly representative of his ability, there should've been teams offering to make him the highest-paid player in the league for even just one or two years, in the hopes of getting him and winning a Super Bowl or two with him. The answer is that the guy had six years in Miami that were mediocre, and even the guys who do this for a living aren't sold that his 2019 performance is replicable. You're operating within the world of "FinFaninBuffalo loves Ryan Tannehill" and not seeing the big picture here.
There was a **** ton of evidence that it was due to other things. You simply chose to ignore it all. Tannehill wanted to stay in Tenn. You have no idea if there were other teams contacting Tannehill. Objective people consider Tannehill a top 10 QB. That is how he got paid. A 117.5 passer rating with a 9.6 YPA isn't replicable. Everyone agrees on that. Only you think that opinion is some kind of smoking gun. If Tenn thinks they are getting the guy you blame for Miami's failures, why offer him a contract at all? They already had that guy under contract (Mariota). Certainly that guy isn't worth a top 10 contract, yet here we are. And they did it in an offseason where Brady, Winston, Newton, Brees, Bridgewater, Dalton, and Rivers were available. You interpret everything related to Tannehill wrong.
Again... How do you factor in that when he threw passes, they traveled farther than other QBs passes? A guy throwing long passes is usually not as successful as on shorter passes. Throwing longer passes successfully will reduce the volume of passes... ESPECIALLY if you then pound the rock.
Team 1 takes 3 throws to travel 30 yards, team 2 takes 5. Mixed in the middle is 2 runs for 10 yards. They both travel 40 yards. Team A uses 3/2. Team B uses 5/2. Don't you know team B has the better QB...... Because passing volume!!!! Effectiveness be damned!!!! LOL.
What you're implying above is that there was a strong, negative correlation between Tannehill's YPA and his passing volume, game-by-game, in 2019 -- i.e., the greater his YPA, the lower his percentage of pass dropbacks. Here are those data for the 15 games in which he played: % Pass Dropbacks ; YPA 28.07 ; 4.8 28.3 ; 6.3 33.33 ; 9.9 36.54 ; 14.39 45.76 ; 14.48 46.94 ; 9.53 48.28 ; 8.27 51.67 ; 10.76 55.17 ; 10.07 56.72 ; 7.75 58.62 ; 6.74 63.16 ; 5.85 66.67 ; 9 67.19 ; 8.49 The above correlation is an extremely weak -0.06, whether we include the Denver game or omit it. The same is true for pass attempts and YPA: ATT ; YPA 15 ; 4.8 14 ; 6.3 20 ; 9.9 18 ; 14.39 27 ; 14.48 19 ; 9.53 22 ; 8.27 29 ; 10.76 27 ; 10.07 36 ; 7.75 31 ; 6.74 33 ; 5.85 39 ; 8.49 16 ; 9 The correlation above is an extremely weak -0.04, whether we include the Denver game or omit it. This is yet another instance, as has happened throughout the thread, of when someone makes an assertion supported by no objective evidence, and then I refute it with objective evidence. Yet I am supposedly "not to be trusted." I would encourage you to consider that the people offering up opinions or conclusions without any evidence, only to have them refuted objectively, are the ones "not to be trusted."
So when you imply that Tannehill's "effectiveness" is responsible for his passing volume, and then the correlation between his effectiveness and his passing volume is shown not to exist, we should somehow then back off of the idea that Tannehill is responsible for the (absence of a) relationship? So if the correlation actually exists, then it's Tannehill and his "effectiveness." If it doesn't exist, then it's something else, not Tannehill. LOL -- again, you got yourself a nice little world there. Tannehill can do no wrong. He's a player without fault and without weakness. Everything wonderful is attributable to him. Everything questionable is attributable to something else.
The "low volume theory" is not merely an opinion, when Tannehill's percentage of pass dropbacks in 2019 was 2.14 standard deviations below the league norm, and that pattern permeated the Titans' season, regardless of situation. You're explaining that in terms of Tannehill's "effectiveness," though that explanation was just thoroughly disposed of.
You’re getting frantic. there are far too many variables for your simplistic approach to be meaningful. You need a far better grasp on the nuances of the game.
1st down and 61 to go, with 1:21 left in the game...Tennessee trailing by 5. 1st down; 18 yard scramble by Tannehill 1st down; incomplete pass 2nd down; 21 yard pass by Tannehill 1st down; 23 yard pass from Tannehill...TD 2pt conversion; 2 yard run by Tannehill...conversion is good With the game on the line, against the BEST team in the league, it was Ryan Tannehill and HIS play that led Tennessee to win, moving the ball 61 yards in :58 seconds. Anyone who says Tannehill can't win a game is blindly obtuse and completely nonobjective.
First of all, my pulse is about 50. Second, the "approach" you're talking about is yours. You're saying Tannehill's "effectiveness" was the cause of his low passing volume. However, there wasn't even a correlation between those two variables.
The point is that you hypothesized that Tannehill's "effectiveness" (YPA) was the explanation for his low passing volume in 2019. Why was there no need for any other variables in that hypothesis when you mentioned it? It's only when it's shown that there is no correlation between those two variables (YPA and passing volume) that we need to consider "dozens of other variables."
I've been telling you for the whole thread that your approaches are too simplistic. You don't understand the game well enough.
That could conceivably be the retort in any instance in which there is credible information presented that refutes what you believe, and there would be no way of determining whether you're correct. In other words, you're repeatedly falling back on an unfalsifiable position. If my "simplistic understanding of the game" is what every point in the debate hinges on, then you've essentially won the debate before we've started. Every point that I make, regardless of how well it's supported, can simply be responded to with "you don't understand the game well enough," and there would be no way of controverting that claim. Again, that's unfalsifiable, and when you're debating from an unfalsifiable position, you're debating from an inherently weak one.
It is not the information that is the problem. It is the conclusions you draw from it. You've been trying to demonstrate that Tannehill's performance is tied to the number of pass attempts. When cbrad showed you that there was no correlation between the two, you claimed that correlation wasn't the correct way to measure the relationship. Just nonsense. I also linked you to a separate study of Tannehill and it also showed no correlation between pass attempts and passer rating. That study came to the same conclusions that I did and the opposite of what you concluded and still you persist. Here are the conclusions by that study. They should look familiar to you. - Tannehill as every other QB in NFL benefits from balance. When Tannehill is playing well he does not need to throw as much. - In the great majority of situations where Tannehill has thrown a lot it has been because the team has fallen behind early. Way before Tannehill threw the ball 30 times. - But there is no evidence that Tannehill's performance is related to the number of throws he makes. From the same article: 1. Attempts per game vs Passer Rating. Correlation was -0.161. That shows a weak negative correlation. You got three people refuting your conclusions and you still refuse to acknowledge that you are wrong.
The quote below is from your post: So if that's true, then how about if the "balance:" 1) consists of a percentage of pass dropbacks 2.14 standard deviations below the league average in the regular season, 2) consists of a percentage of pass dropbacks six standard deviations (that's still almost funny to see in print...) below the league average in the first two games of the playoffs, and 3) permeates the team's offensive strategy throughout the season such that it exists at greater than league normal quantities even at times when teams typically put games largely on their quarterbacks' shoulders -- for example, trailing in games, trailing by more than one score in the second half, and trailing by more than one score in the fourth quarter? What "benefit" (i.e., the quote above) would the quarterback then be expected to experience?
LOL. That offensive strategy PREDATES Tannehill joining the team. 2018: 52% pass 48% run. I think you should focus on Mahomes. In the KC wins, they used a 58/42 pass run. In the losses it was 73/27. Clearly KC needs to limit Mahomes to a below average ratio in order to win. Right?