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

Is Ryan Tannehill the long term solution at QB?

Discussion in 'Miami Dolphins Forum' started by Chuck Wilson, Nov 1, 2015.

Is Ryan Tannehill the long term answer at QB for us?

  1. Yes

    44 vote(s)
    40.7%
  2. No

    39 vote(s)
    36.1%
  3. Not quite sure, need to see more

    25 vote(s)
    23.1%
  1. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    Didn't Kaep sign a nice contract too? Just sayin ;)

    I believe I read they can cut him after 2016. Not saying they would. Just that they can.
     
    dolphin25 likes this.
  2. keypusher

    keypusher Well-Known Member

    1,351
    448
    83
    Nov 29, 2007
    Yes, I understand that he was not rating QBs by wins. He was trying to come up with a good methodology -- one that was better than passer rating, which basically "proves" that all the greatest quarterbacks in the history of the game are playing right now.

    Here, from a much earlier post, is his response to attacks on "adjusted yards per attempt," which is the bedrock of his methodology.

    "Adjusted Yards per Attempt isn't a good stat, doesn't correlate well to wins, or is biased against players in the West Coast Offense."

    This criticism isn't a good one. I'll put this on my to-do list for the blog, but I know I've heard that Yards/Attempt For and Yards/Attempt Allowed correlates very, very well with team winning percentage. I'm confident that Adjusted Yards per Attempt is an upgrade on yards per attempt, so I strongly believe that this is the right metric.

    As for favoring one system or another, I'd also disagree with that. Completion percentage favors QBs in the WCO. Yards per completion favors QBs in vertical offenses. Yards per attempt is the best of both worlds, and it shouldn't really favor either.

    Adjusted yards per attempt probably hurts the players in vertical offenses more, because of the high penalty for interceptions. But INTs are very costly to a team (check out the winning percentages of teams that lose the turnover battle), so I'm ok with this. In short, a good QB will have a good adjusted Y/A, regardless of system. For what it's worth, I think quarterback Rating cuts way too much the other way; it disproportionately rewards QBs with very high completion percentages.


    Obviously his methodology isn't perfect, but it's sensible. What struck me was how highly his results correlated with W/L.

    If anyone cares, I'm agnostic on Tannehill. I don't know enough.
     
  3. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    But that correlation is meaningless without looking at the teams that surround those QBs. I don't think most QBs with good win/loss records were on teams that were full of garbage.
     
    Fin D likes this.
  4. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    This thread has displayed one truth....facts are irrelevant to ANY discussion.


    Sent from my iPhone using Tapatalk
     
  5. keypusher

    keypusher Well-Known Member

    1,351
    448
    83
    Nov 29, 2007
    We don't know if it's meaningless or not. Correlation doesn't prove causation, but it implies it -- it suggests that there's something worth investigating. The one thing we know (without checking) about the teams that surrounded great quarterbacks is that they had great quarterbacks.

    But, eyeballing Oremland's top 10, here are my thoughts:

    P. Manning -- I think Manning does it himself to a huge degree. He's basically the O/C for any team he plays for. And we all saw what happened to the Colts when he was out. A poster child for the importance of the QB position.
    Otto Graham -- Played in the championship game every year of his career. Great teammates and the greatest coach of his day, Paul Brown. But, it's striking how much the Browns fell off when he retired, even though they got Jim Brown in 1957.
    Johnny Unitas -- Had very good teammates for most of his career, and was fortunate to be coached by Weeb Ewbank and Don Shula.
    Marino -- See Payton Manning. As Oremland says, had just two years on a Super Bowl caliber team ('84 and '85), and that was solely because he was on the team (the defense fell off a lot between '82 and '84). Even Shula was in a slow decline for most of his career.
    Joe Montana -- I think his first Super Bowl win was with one of the least talented teams (apart from him) to win a title. But, that changed later, and anyway he was playing for one of the best coaches ever.
    Fran Tarkenton -- a really long career, most of it with crappy teammates. Paul Zimmerman wrote that a game where he "bled first downs" for a no-talent Giants team was the greatest quarterbacking performance he ever saw. On the other hand, I think the only reason he winds up with a winning record overall is that he played for those good 70s Vikings teams.
    T. Brady -- I don't think he's been particularly blessed with great teammates over the years, but he's got the best coach.
    Steve Young -- great teammates, no question.
    Roger Staubach -- great teammates and a great coach.

    So, kind of a mixed bag. Looking over the list, I'm struck by what a difference coaches make.
     
  6. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Brady has had teammates, who while not necessarily the greatest (i.e., Edelman is not the greatest WR in the league), they excel in the system they are in. Of course, Brady has possibly the best tight end in the league.
     
  7. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Whether a criticism against ANY/A is justified or not depends on the goal. If the goal is to just come up with something that has a high correlation with how we would intuitively rank QB's then the ONLY measure of how good his methodology is (including the use of ANY/A) is how good it fits to intuition. I think his approach works fairly well.

    But if the goal is to develop a measure of QB ability independent of testing it against intuition, then the critique of ANY/A is that the weights (like the 20 for TD and 45 for INT) are subjective. The way people usually get rid of "subjective" weights in mathematical modeling is either to: 1) let a computer find the optimal weights by going through a ton of possible ones (this is called Maximum Likelihood Estimation, or MLE), or 2) use a separate theory/model to justify the weights. For ANY/A as an independent measure, MLE won't work because you don't know the "ground truth" (e.g. an intuitive ranking). And he shows no other theory/model that says TD:INT should be weighted by 20:45.

    So I'm not critiquing what he did as long as his goal is to produce math that leads to a ranking similar to our intuition (and that's his goal), but ANY/A as a measure on its own isn't justifiable.
     
    keypusher likes this.
  8. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Gotta be careful about saying when correlation is meaningless. If you just want a measure of something, then correlation can be sufficient.

    For example, let's say the number of Chinese hacking attempts on the DoD for whatever unknown reason correlates perfectly with the number of ice cream cones sold in Florida. Even if there's no causal relationship, the number of ice cream cones sold in Florida is still a perfect measure of Chinese hacking attempts on the DoD, in that absurd hypothetical case.

    Same thing here. This is all post-hoc reasoning we're talking about. We're asking how well W/L record for QB's correlates with what we think is a good ranking of QB's. However well it does is however good a measure it is of how good QB's are (after the fact). And that's true even if there was NO relationship between W/L and QB ability, which we know isn't true.

    It's only when you're trying to argue that the correlation tells you how much of the W/L the QB was responsible for that you get into trouble. Even then, statistics can help (though not solve the problem) by estimating the "variance explained" by the correlation.
     
    keypusher likes this.
  9. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,650
    67,542
    113
    Dec 20, 2007
    Gronk should of been a dolphin..we needed him, his back was fine, and his film was ridiculous..
     
    resnor likes this.
  10. bakedmatt

    bakedmatt Well-Known Member

    2,129
    909
    113
    Mar 29, 2008
    Orlando, FL
    He IS the best TE in the league. He's one of the leagues best/most dominant players of any position.
     
  11. Phins Up Wins Up

    Phins Up Wins Up Banned

    1,471
    269
    0
    Nov 27, 2014
    Brady like any other good QB makes them great. They excel in the system because Brady orchestrates it. He knows how to use his guys and get them the ball. Edelman here would be nothing with Tannehill throwing the ball. Your downplaying what great QBs do for their teams is a joke. Brandon Lafell before his injuries had almost 1,000 yards last year playing with Brady. Before that he was trash on the bench in Carolina. His pass blocking is 21st. Doesn't have an elite WR but he gets the job done. Tannehill on that team they wouldn't even make the playoffs. We have a great tight end here in Jordan Cameron but he is being wasted here due to Tannehill not hitting him in the seams down the field. But yeah great QBs don't make their teams better. You're right.
     
    gunn34 likes this.
  12. Phins Up Wins Up

    Phins Up Wins Up Banned

    1,471
    269
    0
    Nov 27, 2014
    Brady would love to have Lamar Miller and the receivers Tannehill has here. And Jordan Cameron would be Aaron Hernandez in New England. Brady does more with less. That's what great quarterbacks do.
     
  13. keypusher

    keypusher Well-Known Member

    1,351
    448
    83
    Nov 29, 2007
    Thanks for this instructive reply. I thought the impact of the adjustment is relatively modest, but maybe it's not. Marino has 7.3 career y/a but 7.0 ay/a.

    [Sorry, I was responding to cbrad, but I must have screwed up the post somehow for that not to show.]
     
    cbrad likes this.
  14. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    You bring up a good question.. how does Y/A compare with AY/A on average (I wouldn't just go by a difference for one QB). Not sure if you want to gather the stats on this (I can't do that now, but I can in the evening) but I'd be very interested in the distributions of those two. Maybe someone has already done the analysis and we can just look it up.. if not I'd just pick the top 20-30 QB's (from some reasonable list.. can be your own) and see what the distributions look like.
     
  15. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Oh yeah? I disagree. What makes these receivers "great" is the system, and how often they throw the routes that the receivers are good at. Like Welker. Welker did nothing different in NE, other than get thrown 30-40% more balls, and he went over 1,000 yards a couple times, and went to multiple Pro Bowls. Brandon LaFell is EXACTLY like Welker. His ypc was actually higher almost every year in Carolina. Know why he almost got 1000 yards in NE? They threw him the ball almost twice as much as Carolina did.

    But yeah, it's all Brady.
     
  16. roy_miami

    roy_miami Well-Known Member

    1,385
    560
    113
    Oct 11, 2013
    Yeah well Tannehill is running the same system that made Foles look like a semi-god on a weekly basis.
     
  17. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Oh, Chip Kelly is our OC? I had no idea.
     
    Aqua4Ever04 likes this.
  18. keypusher

    keypusher Well-Known Member

    1,351
    448
    83
    Nov 29, 2007
    I definitely can't do it any time soon. I'm pretty sure a query can be configured at Pro Football Reference to give you the answer, but damned if I know how.
     
  19. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    No problem.. had some time to do it just now. Using the top 20 QB's ranked by total career wins, average Y/A was 7.34 and average AY/A was 6.82, so on average Y/A and AY/A differ by about 0.5 Y/A. The correlation between the two was 0.7 and the biggest differences were for Bradshaw (Y/A=7.2 vs. AY/A=5.8) and Stabler (Y/A=7.4 vs. AY/A=5.8).
     
    keypusher likes this.
  20. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,650
    67,542
    113
    Dec 20, 2007
    closest thing to me getting tail is watching three seasons of new girl with my 12 year old...Ive been there lil bro, one day we shall share some stories..maybe teach ya a few things about those cretures, and maybe if your lucky, tell you about my secret move.
     
  21. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,650
    67,542
    113
    Dec 20, 2007
    lol.some things never change...
     
  22. Phins Up Wins Up

    Phins Up Wins Up Banned

    1,471
    269
    0
    Nov 27, 2014
    Lafell didn't get thrown to a lot because he was on the bench. Brady took him from a bench player to nearly a 1,000 yard receiver. Welker wasn't good in New England? Lmao geez kid you do way too much reaching.
     
  23. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    Not true.

    http://espn.go.com/nfl/player/stats/_/id/12576/brandon-lafell

    Went from 88 targets to 119, not exactly double. But more importantly, his catch % went from 55.6% to about 62% under Brady.
     
  24. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Not true? I said "threw him the ball ALMOST TWICE AS MUCH," although, I should have said he caught almost twice as many, as I think that was the start I was looking at. He had 49 receptions his last year in Carolina, and 74 receptions his first year in NE. So I never said double. So trying to play the Gotcha game. Secondly, why give an exact amount for Carolina, 55.6%, but not the exact amount in NE? I assume it was lower than 62%, since you said "almost." So, his catch percentage went up about 6%? What does that mean in terms of catches? How many more? How dies it really affect anything I was saying? The bigger impact in his stats was the amount of balls that went his way. Just like Welker.
     
  25. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    First off, it wasn't a gotcha, which is why I didn't emphasize it, just correcting you. Second, the exact number was because that's the one on my calculator still. I calculated his NE numbers first but it wasn't on my calculator anymore. Your assumptions fail you (again). It was 62.18%. I didnt even say "almost" I said "about." Lol man, you are really losing it. Notice how you're misreading and assuming my intentions are all negative

    You can discount a 10-11% increase in catches per target that's fine. Just putting stats out there.
     
  26. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    55% up to 62% isn't 10-11%, man. How do you increase catches per target? Anyway, having almost twice as many catches is certainly a bigger factor in the higher yardage total than a 7% better catch percentage.

    Please recall how we started talking about LaFell, it was the their that Brady took him from garbage to almost an 1,000 yard receiver. That theory is patently false.
     
  27. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    Not even an apology for misreading and then assuming my negative intentions? I'm catching feelings.

    And your math skills suX0Rz.

    55.6 to 62.18% = ~6.5%.

    6.5%/55.6% = 11.69% increase.

    You do understand you have to do it that way. Because a 25% catch rate to 50% catch rate is a 25% nominal increase but a 100% increase. 50% to 75% is also a 25% nominal increase but is just a 50% increase. Same as 5% to 10% (5% nominal/100%) or 85% to 90% (also 5% nominal but merely 5.8% increase)
     
    keypusher likes this.
  28. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Dude, that's some serious math skills, to explain that his catch percentage went up 6.5%.

    Regardless, none of that changes what I was saying. LaFell had far more yardage because he was thrown the ball more. He had almost twice as many catches. Someone used LaFell's yardage as an proof that Brady made him better. That is disingenuous at best.
     
  29. jacquesstrap

    jacquesstrap New Member

    2
    3
    0
    Oct 31, 2015
    Georgia
    The definition of "franchise" is being a player so critical that if that player is lost than the whole season is lost too. Tannehill isn't a Franchise QB. If he was injured, the dolphins might be slightly worse but since Tannehill is only good for 7 to 9 wins (with everything being perfect) than you really haven't lost much. How many times has Miami won it's division with Tannehill ? Tannehill's inconsistency will not create any job stability for whoever the coach might be.
     
    Phins Up Wins Up and dolphin25 like this.
  30. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    That's your definition of franchise.

    Losing a couple of wins could absolutely be the difference a good or bad season. Also, for all you know, our team could be a 2 win team without Tannehill.

    How are Hartline and Wallace doing without Tannehill for example?
     
    resnor likes this.
  31. keypusher

    keypusher Well-Known Member

    1,351
    448
    83
    Nov 29, 2007
    Makes sense they would be hit by the adjustment, they both had issues with interceptions. Also, I think (unlike Marino and Namath, but like Griese) they weren't good at avoiding sacks.
     
  32. keypusher

    keypusher Well-Known Member

    1,351
    448
    83
    Nov 29, 2007
    Thanks! Damn, there are some people on this thread who are very good at math.
     
  33. Phins Up Wins Up

    Phins Up Wins Up Banned

    1,471
    269
    0
    Nov 27, 2014
    Wallace did much better in Pittsburgh when Big Ben was throwing him the ball.
     
    dolphin25 likes this.
  34. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Further, that is the increase compared to his percent, not compared to if he caught 100%. Like, 6.5% is 11% or so of 55%, but his total increase, was 6.5%, as compared to the total amount thrown him.
     
  35. Chuck Wilson

    Chuck Wilson New Member

    89
    54
    0
    Oct 20, 2015
    How the hell did we get on mathematic equations on the increase in Brandon Lafell's numbers? Its not rocket science, he's more productive because he's getting more looks and he has the greatest QB of all time slinging him the rock.

    The debate is Ryan Tannehill the answer or not. I voted no, obvious if his circumstances change so could that answer. He needs better support and a coach around him who doesn't have his head up his *** to be a winning QB. That's not rocket science either, this past off-season should've been about solidifying the OL to protect him and get the running game going.

    Joe Philbin had this vision in his head that he was going to turn us into the Miami Packers except he's a terrible coach and Ryan Tannehill is far from Aaron Rodgers.

    Watching the Bengals last night all I could think about was how fortunate Tannehill would be to work in a system created by Hue Jackson. Just beautiful design and play calling.
     
    adamprez2003 likes this.
  36. dolphin25

    dolphin25 Well-Known Member

    6,338
    2,400
    113
    Nov 22, 2014
    except Tannehill won't throw down field, nor does he avoid sacks like Dalton does. Simms even mentioned something about Dalton throwing the ball down field not worrying about getting an 80% completion rate like Tannehill
     
  37. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    Not in 2012. But hey...facts....
     
    resnor likes this.
  38. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Yep, Wallace in 2012 looks early similar to Wallace every year since he left Pitt.

    Strange, I know.
     
  39. RGF

    RGF THE FINSTER Club Member

    6,066
    3,436
    113
    Nov 24, 2007
    NY
    And how exactly do you know this since everything hasn't been perfect since we drafted him?
     
    resnor likes this.
  40. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    Antonio brown and Ben All did worse in 2012. Big Ben and Haley sniped each other in the press. Facts. I like diggs better though.
     

Share This Page