1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Ryan Tannehill

Discussion in 'Other NFL' started by bbqpitlover, Oct 16, 2019.

Ryan Tannehill is...

  1. A terrible QB

    0 vote(s)
    0.0%
  2. A below average QB

    4 vote(s)
    5.7%
  3. An average QB

    7 vote(s)
    10.0%
  4. An above average QB

    39 vote(s)
    55.7%
  5. An elite QB

    16 vote(s)
    22.9%
  6. The GOAT.

    4 vote(s)
    5.7%
  1. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    I love rhetorical questions.....
     
    resnor likes this.
  2. Galant

    Galant Love - Unity - Sacrifice - Eternity

    19,127
    11,058
    113
    Apr 22, 2014
    I was writing a longer response but I don't think it's helpful. Let's just say I disagree that medical research and the sports world are the same when it comes to drawing conclusions. It's irrelevant though.

    As I said, we're talking about a sport forum here and my point was not specifically about statistical analysis which I've said can have value. You in particular have more or less wanted to stick to the stats and refrain from drawing conclusions until the data are in. My point was, and is, that the conclusions people want to draw here, or statement they want to make, are generally untenable. That's why I asked for people to make some statements about Tannehill. The only way to make the sort of statement people make is to get very theoretical or very vague about things. Tannehill is elite. Tannehill is poor. What's elite? What's poor? Tannehill couldn't play in the majority of systems. It's all beyond verification or so theoretical as to be tantamount to fantasy.

    I challenge everyone to ask themselves what can definitely be state about Ryan Tannehill, or any other player, or team. It's a good way to get some perspective or a reality check. Lock down what is certain first, then expand from there. Consider the statements one makes and ask what you're really trying to say, and ask how important those things are. Try to take a wide perspective. Those things at least well help ground a discussion or a perspective. It's easy to get tunnel vision or double down on something.

    In general, don't go beyond what can truly be known, or else, the further one wanders from that territory realise that less and less can be said with any certainty.

    As for this:
    @The Guy
    "What if the HoF QBs' career passer ratings (for example) were significantly higher than those of other QBs? What then does career passer rating mean in terms of QBs' ability?"

    I don't think you get my point. Are you trying to suggest that passer rating is an accurate way of assessing QB ability?
     
    resnor likes this.
  3. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    So what do you think after today? Brady was off a LOT and totally failed in an attempted comeback even when he had ample time to do so (due to the Bears not trying to run out the clock!). VERY different from the Brady we used to see.

    Brady ended with a 86.7 rating for the game and now has a 96.8 rating for the year (on 196 attempts). The league average is 96.5 so Brady is almost exactly statistically average at this point. Anyway, this thread looks like it's large enough to accommodate tracking Brady in addition to Tannehill so we'll see how he ends the year.

    Oh, also interesting is that all this is with Gronk. Somehow TB doesn't know how to leverage that.
     
    resnor, The Guy and KeyFin like this.
  4. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    The whole quarterback rating system is subjective and many times blown out of proportion.

    Would you rather have a quarterback with a 125 seasonal average passer rating and the team with a 9-7 record, or a quarterback with a 90 seasonal passer rating and the team has a 12-2 record?.

    At the end of the day, the ONLY stat that matters is wins and losses. Thinking back to our last glory days with Marino, I would have traded so many of those passing stats and records for WINS...and so would have Marino.
     
  5. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Yep, Brady may in fact be done. If he maintains this level of performance throughout the season, what do you think it says about the attribution of his previous success to him versus Belichick/the system, given that he's been at this level for a while now, even with Belichick?
     
  6. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    What you're saying would be relevant if there were no correlation between quarterback performance and winning in the NFL. That relationship is fairly strong, however, and so quarterback performance certainly isn't "blown out of proportion." Otherwise quarterbacks wouldn't be far and away the highest-paid players in the game. So, obviously the guys doling out the money here don't share your opinion.

    Take for example the passer rating you nominated above -- a season passer rating of 125. There has been only one such passer rating in the vicinity of that in the past 10 seasons -- Aaron Rodgers in 2011, who had a season passer rating of 122.6, well above that year's league average of 84.3.

    His team's record that year? 15-1. And entirely predicted by the relationship between passer rating and winning in the NFL. Certainly you don't think the team achieved a 15-1 record with smoke and mirrors and it had nothing to do with Rodgers's performance.
     
  7. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    It doesn't mean a thing. Would anyone here try to argue that Brett Favre was a better quarterback that Johnny Unitas? Would anyone here try to argue that Sammy Baugh was a better quarterback than Kurt Warner? Would anyone try to argue that Warren Moon was a better quarterback than Terry Bradshaw? Passer rating is irrelevant in trying to argue one quarterback is "better" than another, particularly if you're trying to make that comparison with HoF quarterbacks.

    I have often made the argument, although I can't PROVE it, that today's (or VERY recent quarterbacks) wouldn't be able to compete at the same statistical level as yester-year's quarterbacks due to the utter brutality of defenses of yester-year. I think though that anyone with ANY common sense though would be able to see this being a reasonable conclusion. And if you are able to see this as a reasonable, to try and use passer rating as a gauge to determine one HoF quarterback being better than another is quite nonsensical.
     
  8. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Why are you comparing HoF QBs to each other? The question was what it would mean in terms of the validity of passer rating as a measure of QBs' performance if career passer ratings distinguished HoF QBs from non-HoF QBs. So compare Brett Favre for example to Jim Harbaugh and see what happens.
     
  9. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    Brady IS done!!! I said that last season, I said that when he signed with Tampa Bay and said he was going to get his d1ck knocked in the dirt. The 3 sacks on Brady last night all came in the 2nd half and Brady was frazzled all night.

    He should have hung up his cleats and called it a career.
     
  10. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    Because you brought it up...

     
  11. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Yeah well, readily apparent in that in my opinion is a distinction between HoF and non-HoF QBs.
     
  12. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    If @cbrad is willing, I'd like for him to list the top 200 QBs in history, ordered by adjusted career passer rating (with adequate sample size of career passing of course). Let's see if there's a strong correlation between career passer rating and people's perceptions of the ability of those players. If there is, then passer rating is certainly an accurate way of assessing QBs' ability, at least in terms of the opinions of people here.
     
  13. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    It's not subjective. It's 100% objective: there's a formula.

    I think that's very likely true. But I also think the same is true of the reverse: yesterday's great QB's wouldn't have similar stats in today's game, once adjusted for era.

    You might argue that because defenses were allowed to be more brutal on QB's and WR's in the past that anyone who succeeds in that more brutal environment would just have it easier in a less brutal environment, but what you see in other sports is that crossing over from one set of rules to another is difficult regardless of how permissive or not those rules are.

    Example: the best kickboxers in the world can't beat the best boxers under boxing rules (analogous to what you're describing: going from more permissive rules to less permissive), while the best boxers in the world can't beat the best kickboxers in the rules under kickboxing rules (i.e., the reverse). So there's evidence that claim is true in BOTH directions.
     
  14. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    Sure there's correlation! Don't be so lazy and actually LOOK!. I picked a year...ANY year...2015 for example.

    Here are the 1st place teams in each division, their record and the passer rating of their starting quarterbacks.
    AFC
    NE: 12-4; 102.2
    CIN: 12-4; 106.2
    HOU: 9-7; 91.4
    DEN: 12-4; 86.4
    NFC
    WAS: 9-7; 101.6
    MIN: 11-5; 88.7
    CAR: 15-1; 99.4
    ARI: 13-3; 104.6

    Now this is just picking any season indiscriminately and looking at this one particular season, you can see that Denver and Minnesota had double digit win seasons, 12-4 and 11-5 with quarterback apsser ratings in the 80's. You also see the Redskins had a quarterback passer rating 101.6 with a dismal 9-7 record.

    Oh, Did I fail to mention the season leading passer rating quarterback, Russell Wilson with a 110.1 resulted in a 10-6 season, 2nd place in the NFC West?

    Pittsburgh attained a 10-6 record with a 94.5 rating...which ranked 11th.

    So how did all of this play out in the end? Well, I'm glad you asked!!!

    Green Bay (92.7) defeated Washington (101.6)
    Seattle (110.6) defeated Minnesota (88.7)
    Kansas City (95.4) defeated Houston (91.4)
    Pittsburgh (94.5) defeated Cincinnati (106.2)

    Green Bay and Pittsburgh, with significantly lower passer ratings defeated Washington and Cincinnati

    Arizona (104.6) defeated Green Bay (92.7)
    Carolina (99.4) defeated Seattle (110.1)
    New England (102.2) defeated Kansas City (95.4)
    Denver (86.4) defeated Pittsburgh (94.5)

    Carolina and Denver, with significant lower passer ratings defeated Seattle (#1) and Pittsburgh

    Carolina (99.4) defeated Arizona (104.6)
    Denver (86.4) defeated New England (102.2)

    Carolina and Denver, with lower passer ratings defeated Arizona and New England

    Denver (86.4) defeated Carolina (99.4)

    Denver with a significantly lower passer rating wins the Super Bowl

    So please, don't go touting that higher passer ratings result in higher wins. I just picked this one season out of thin air and did the actual leg work. This one season alone proved lower passer ratings resulted in double digit wins. It also proved that higher passer ratings resulted in single digit wins. Furthermore, it also proved that once the playoffs hit, those passer ratings don't mean squat. of the final four, the two teams with the LOWER passer rating advanced to the Super Bowl and the team with the LOWEST passer rating of all the playoff teams, the team with a 12-4 regular season record and a passer rating of 86.4 won the Super Bowl.

    There is correlation. Take the time and actually look.
     
  15. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Good question. First we have to see how Belichick ends the season. If NE has its first 9-7 season since 2002, that alone is evidence that Brady was not easily replaceable in Belichick's system. Right now, the evidence is pointing in that direction because Newton hasn't shown a Belichick bump yet, and you just had a game where 2 backup QB's were atrocious.

    So even if Brady ends around average, this would suggest that HC + QB was the "right fit" and that HC alone wasn't sufficient. The other thing is that if Brady ends around average in Tampa, it's evidence that QB ability is relatively independent of this particular system change. Same is being shown with Newton right now. In any case, those are the two things I'm looking for: 1) how irreplaceable/replaceable was Brady, and 2) does QB ability transfer in these cases.

    As far as who is more responsible for that dynasty's success, if we find that Brady was not easily replaceable, then that's evidence that both were similarly important. I'm just happy they parted so we have this "natural experiment".

    I did that once, from 1-100. Not sure if you can find that post, but this exact issue came up and the result was as expected. With a few exceptions the ordering fits intuition, though of course people might disagree in the specific ordering, and it's really hard for people to distinguish among more average QB's. I'd do it again except it takes time to type!! Anyway, if this becomes important enough a question I'll do it again, but I know there's a post out there with that 1-100 ranking, I think from last year.
     
    The Guy likes this.
  16. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    I get your boxing analogy and to a degree I would agree with THAT but as far as yester-year's quarterbacks not being able to statistically compete with today's quarterbacks, I don't necessarily agree with that though.

    Would Sammy Baugh and YA Little put up the same kind of numbers that Manning, Brees and Brady have? No, I doubt that as the game was more of a running game than passing game however...

    Marino? Elway? Moon? Kelly? Fouts? Tarkington? Hell, had THOSE quarterbacks had the exact same rules that today's quarterbacks have, Manning, Brees and Brady would STILL be chasing records that would be unobtainable! I can't PROVE it but...those were ***PASSING*** quarterbacks! Those teams THREW the ball and set unheard of records for their time...under brutal defensive rules.

    With today's patty-cake defensive rules the NFL has instituted? Geesh, forget about!!!!
     
  17. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    You know what's so funny about this? 2015 was an anomalous year lol. Here's a graph of the correlation between passer rating and win% for every year in the SB era (wait till the image loads.. somehow this site is slow now):
    [​IMG]

    So 2015 was particularly low.

    Either way, please note that correlation isn't 1 = perfect. So you shouldn't expect to always see higher passer rating for greater win%. The average across NFL history is about 0.634, which translates to about 40% of the variance in win% being explained by passer rating. In other words, 40% of win% being explained is what your expectation should be when someone says correlation is high. I mean.. that percent explained can't be TOO high like well over 50% because there's this thing called defense you have to take into account.
     
    The Guy likes this.
  18. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    And now do the correlation for passer rating differential, where the defense is considered but the focus is just as strongly on quarterbacking (either doing it or stopping it).
     
  19. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Find me any correlation that isn't 1.0 and I'll find you exceptions to the rule with regard to it.

    Again, either you are right and the entire league is wrong for clamoring for QBs and paying them what they do, or you are wrong and the entire league is right. Which do you think it is?
     
  20. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    I know, but the same argument should apply to the kickboxer right? They learn the same techniques as the boxer, yet it doesn't happen. For one the stance is different when you have to worry about kicks so you get used to something different.

    I think the same is true of these passing QB's. They probably would succeed really well in today's game IF they only played in today's game, but not if you transplanted them from the past to today. It's not the skillset, it's what you adapt to. Same I think is true in reverse. Aaron Rodgers and Drew Brees would do well 30 years ago IF that's what they grew up with.
     
  21. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    The average correlation between passer rating differential and win% in the SB era is 0.7972, which means that about 63.56% of the variation in win% can be explained by just the passing game (as captured in passer rating).
     
    The Guy likes this.
  22. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    OK I can certainly find that. I thought it might be a copy and paste from another program for you. Certainly didn't want you typing that!
     
    cbrad likes this.
  23. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    But the point I was getting at was don’t speak in absolutes and do the actual leg work I just picked 2015 out of thin air and worked that season for about an hour. I could pick another season out of thin air and do the same thing but I have to get to work on the honey-do list
     
  24. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,321
    113
    Nov 24, 2007
    Rockledge, FL
    So now SALARY is an indication of how good a quarterback is? I thought this was already discussed. If salaries an indication, then anoint Tannehill and quit saying he’s average.

    Your back and forth cherry picking positions is tiresome
     
  25. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    So for @The_Dark_Knight, that means these sorts of seasons:

    https://www.pro-football-reference....comp=gte&c2val=0&c5val=1.0&order_by=game_date

    ...occur not because Drew Brees played poorly -- his passer ratings were very high each of those seasons -- but because the Saints were so poor at defending the pass. Note that the passer ratings surrendered by the Saints in those years were horrendous, which is primarily why their season records were below .500, despite Brees's very high-quality play.

    And so again, we're talking about quarterback play here -- either doing it, or defending against it.
     
    Pauly likes this.
  26. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Anoint him what, 11th best in the league in terms of average salary per year, not even a standard deviation above the average of the top 32 QBs in the league?

    https://overthecap.com/position/quarterback/

    Sure, I'll anoint him that.
     
  27. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    We aren't speaking in absolutes. We're speaking about correlations and acknowledging they aren't perfect. You're finding exceptions to the rule as a means of disproving the correlation, whereas exceptions to the rule don't necessarily disprove correlations.
     
  28. Galant

    Galant Love - Unity - Sacrifice - Eternity

    19,127
    11,058
    113
    Apr 22, 2014
    One wonders whether Tannehill feels successful. In terms of money earned, stats earned, wins earned... I'm sure he wants to win a SB ring.
    Is he worthy of respect? I think the majority of players and pundits would say yes.
    Would most teams in need of a QB take a shot at him? I think so.
    Would he be guaranteed a SB anywhere? No.
    Would even the most recognised QB's be guaranteed a SB ring anywhere? No.

    By every practical standard I think Tannehill ranks as good QB.

    If someone wants to say that unless you're somewhere in the upper-echelons of QB history you're a failure, we'll have to agree to disagree. That's nonsense, in my view.

    If someone wants to argue that some metric somewhere argues that Tannehill might not give a chance to win a ring, they can, I think it's irrelevant, since there are no guarantees, guys with the right metrics might never win one, guys with the 'wrong' metrics might. Because rings aren't a QB stat. In my opinion, it's about getting the right combination of team and coach in the right year. Time and chance happen to them all.
     
  29. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    I think a better argument or maybe even perhaps the one he was trying to make, is that QB rating, while tied to winning, is still a reflection of the passing offense as a whole and at least not entirely attributed to the QB himself.
     
  30. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    As sample size increases, passer rating becomes more and more attributable to the individual ability of the QB.
     
  31. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    NE had a bunch of roster turnover other than Brady leaving.
     
  32. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Yes, in fact they were hit a lot more than other teams due to COVID. However.. note that the key players they lost were all on defense except Brady. Van Noy, Chung, Hightower, etc. All key players on defense, yet true to form it's not affecting the NE defense too much as they're still ranked 9th in points allowed.

    But on offense where they did NOT lose many key players (arguably losing Gronk after 2018 was bigger than anything this year except Brady) they are suddenly ranked 22nd in points scored, down from 7th the year before, and NE has never been ranked worse than 8th in points scored since 2006.

    That's pretty strong evidence of the influence of both Brady and the QB in general. Let's also not forget how terrible those two backups were against KC. You can visibly see the difference (and cringe!).
     
  33. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    One note about the above -- last year during Tannehill's starts, the Titans had 0.06 EPA per run play, second in the league behind only Baltimore, whose EPA per run play was fueled in large part by Lamar Jackson's 1,200+ yards and 6.9 yards per carry.

    This year that figure for the Titans has plummeted to -0.062 EPA per run play, 17th in the league. If that continues, running the ball as much as the Titans do will become untenable in terms of winning, and Tannehill will be forced to have a larger role in the offense, without as much of the advantage of exploiting defenses that are defending predominantly against the run game.

    That would make this year for Tannehill fundamentally different from last year, and if that happens we'll get a great deal more insight into his functioning in my opinion.
     
  34. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    How do you come to that conclusion? A staff is going to scout certain types of players for a certain type of offense they run, generally. A guy is likely to play with the same -type- of players his whole career or most of it with one team even if they arent stagnant.
     
  35. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    The variation in those surrounding variables increases as sample size increases.
     
  36. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
     
  37. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Wait, this is not true. It's very important to understand the distinction.

    If you have a random variable (which is not the case for football, but let's start with this), variance stays the same as sample size increases. For example, draw samples of size N=10, N=100 or N=1000 from some random variable and calculate the variance and you'll see they're basically the same.

    In football you often don't have random variables, and what you're referring to with "variation increases with larger sample size" is that it takes a larger sample size for changes in a surrounding variable to approximate "random". But once it starts to approximate "random", variance stays the same regardless of further increases in sample size.

    For example, in most cases variance in opponent strength after 2 seasons is essentially the same as after 5 seasons or after 10 or 20 seasons. For WR's or OL you might need to wait 5-10 seasons for things to approach "random", but once it does the variance stays the same. Of course for some variables, like coaching, it never truly approaches random, but that doesn't invalidate the general principle.

    So why does increasing sample size increase the likelihood that any observed difference in a "QB stat" is due to the QB? Because the effect of the surrounding variable in your sample approaches "average effect for that surrounding variable" with increasing sample size, leaving the difference between two such QB stats more likely due to the QB.

    The general rule is that the variance in the mean of a sample decreases with larger sample size. So as sample size gets larger, the mean effect of the surrounding variable within your sample (e.g., mean opponent strength for the opponents you've faced) has smaller variance => approaches the mean effect of the surrounding variable across the league, thus leaving any observed difference in a "QB stat" more likely due to the QB.
     
    The Guy likes this.
  38. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    That’s exactly what I meant. :salute:;)
     
    cbrad likes this.
  39. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    This is basically why the house always wins in casino gambling, even with games such as baccarat where the house margin is 1% or 2%. Once you have enough players making enough bets the variance in the mean becomes tiny.
     
  40. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    This is the crux of the disagreement over Tannehill in Miami. In general, I agree that this is what normally happens,

    The thought was that the sample size for Tannehill in Miami must have accounted for the variation in the play of the OL. The problem, IMO, is that observation of the play of the OL, the year over year volatility of the OL roster, and the dismal record of the members of the OL that were cut by the Dolphins all point to the increased sample size NOT accounting for the play of the OL. I think it clearly demonstrated that the people selecting and coaching the OL were simply bad at their job.

    Add to that two things. First, Tannehill's play in front of an OL not constructed and coached by the same clowns resulted in stellar play for the QB. Second the terrible play of the OL continued in 2019 and the high turnover on the line continued to this season.

    It does appear that Miami may have finally gotten things right this season.
     

Share This Page