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

Why Tannehill should be a better QB now Lazor is gone

Discussion in 'Miami Dolphins Forum' started by Pauly, Mar 27, 2016.

  1. Fin4Ever

    Fin4Ever Season Ticket Holder Club Member

    9,297
    2,738
    113
    Aug 26, 2014
    Vero Beach, FL
    And I totally agree with you on this.
     
    Fin D and Pauly like this.
  2. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    That wasn't analysis, only my opinion based on what I saw, Which is why I did the initial analysis.

    To do it properly you would have to get NFL stats as whole for that game situation and then compare Tannehill to the different tiers of QBs. But that's an analysis, if it were me, I'd really want to do side by side with game tape because of the small sample sizes. Also, I'd also want to match the tape to the play call as well.
     
  3. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    doublepost
     
  4. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    For less than 100 attempts the numbers are really volatile.
    They start to become stable in the 100 to 150 range.
    From 150 to about 250 only really big plays like a dropped 50 yard TD, or a dropped red zone TD turning into an INT can swing the numbers by 3 or 4 points.
    By 300 attempts is when you start to need to change multiple outcomes to swing the rating by a really noticeable amount.

    For example if we take the gif I posted earlier and hypothesize that if the pass route went through they right A gap and Pouncey could slide to help Thomas block then Tannehill could step up into the pass lead Landry for a 50 yard TD.
    If we plug that scenario into 2015 and within +/- 7 in the 4th quarter it changes Tannehill's passer rating from 81.2 to 91.1 (63 attempts). But it changes his combined 2 year rating from 90.7 to 95.4 (133 attempts)
    If we put it into his 2015 tied/ahead numbers it changes his rating from 96.3 to 99.8 (181 attempts). However it only changes his combined 2 year rating from 101.2 to 102.7 (438 attempts)
     
    cuchulainn, resnor, Fin4Ever and 2 others like this.
  5. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Cool! Like I said, good you took some time to do some simulations like that.

    Not saying you have to do this, but if you really wanted to quantify how "volatile" or "stable" passer rating is as attempts increase, you'd first get distributions (from actual data) of the individual components other than attempts: COMP, YDS, TD, INT. Now, randomly sample from each of those distributions, creating a randomly sampled passer rating.

    Since you want all this as a function of attempts, you now let the computer generate a distribution of simulated passer ratings for N attempts (where N is fixed). The variance for that N is what you want. For each N=1,2,...,300 or whatever.. you plot that variance as a function of N. That should quantify "volatile" and "stable" because as it is now the meanings of those terms are subjective.

    Either way, nice to see someone play around with the formula as you did.
     
    Fin4Ever and Pauly like this.
  6. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    I was really surprised at how much the numbers could jump around by altering the outcomes of one big play even with around 200 attempts. Which is why I moved my cutoff for the initial analysis from 200 to 299.

    As you say stable and volatile are subjective descriptors. So I'm really using 'volatile' to mean one big play can make a significant difference in passer rating and 'stable' to mean it takes changing several events to make a significant change in passer rating.

    I do think there is some value in using the terms that way, but we have to remember that they are subjective.
     
  7. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    Same Ole laughable hyperbole from the same ole fan boys.

    "Defense routinely gives up 30".....That same defense has given up less than our offense has scored 3 out of 4 years your guys man crush Ryan has been a Phin.

    "worst OL in recent memory"....Just rolls off the tongue doesn't it, but while they are certainly BAD, they are NOT the worst in the league let alone "ever"

    Sadly I must repeat myself in saying it is no secret ryan has been in a bad situation, but he also hasn't shown much in the way of improv, field awareness, or ability to escape a rush. I'm hoping for a big jump (again) under Gase and being given more freedom, but no matter what he does moving forward? He has been an average at best NFL QB his first four years, and all the "but...but...but...buts" in the world will not change that.

    Enjoy your clams.
     
  8. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    Why are you rehashing stuff from 50+ posts ago that people have moved on from?

    For example if we want to explore Lazor's playcalling becoming more predictable we could look at Tannehill's stats compared to the AFC east vis-a-vis the rest of the league. Now if Lazor's playcalling was a weakness that other teams could exploit then Tannehill's performance against the AFC East should be worse than expected in 2014/2015 even accounting for the strength of the defenses in our division. This should also be true when comparing the Lazor years to the Sherman years. Basically our most common opponents should have the most game film on Lazor's offense and the most reason to try to crack it. If my hypothesis is correct then this should be true.

    Do you want to have a crack at contributing something meaningful to the thread by doing some analysis or do you want to go around in the same old circles we've seen in dozens of threads in the past? Here's the starting point http://www.pro-football-reference.com/players/T/TannRy00/splits/2015/
     
    cuchulainn likes this.
  9. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    See, this is the problem, I didn't say or imply all stats were worthless. I said the specific stats you were using were worthless in this specific argument, because they are. The reason they are worthless in this specific argument is because we're saying because of X, Y & Z Thill has struggled and your arguing against that with stats that don't contain or account for X, Y & Z.
     
  10. Shane Falco

    Shane Falco Banned

    916
    468
    0
    Nov 22, 2015
    These guys are going to be mad this year. And they won't be allowed on the bandwagon when Tannehill is in the pro bowl.
     
  11. Fin4Ever

    Fin4Ever Season Ticket Holder Club Member

    9,297
    2,738
    113
    Aug 26, 2014
    Vero Beach, FL
    Well, I am now down to hoping that we draft a couple of really good guards for our oline as it has sucked for the most part for many years. We DO have to Protect Tannehill better to help him be more successful as a passer. Any QB needs good protection to get the most out of his game. There are several guards in this draft that would drastically improve our line play and finally give him decent protection in the pocket. I feel it is also vital that we draft another high quality RB as well to help Ajayi and still cannot believe that they let Lamar Walk away for such little money difference...especially seeing other offers they have made as well as what they have paid for some of these FA's.
     
  12. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    See for me, and I know some will think this goes against what I've been saying......but we don't need greatness at guard, we just need average.

    I feel like people tend to view the line between great and horrible at a given position as this perfect 45 degree line. Its not always that (in fact, its rarely that). I think, especially at guard, the difference in W/L record between great and average is virtually nothing, but the difference between average and horrible is practically a straight up and down cliff.

    When we had an aggregate average level of play across the oline, Thill played lights out and we were one of the hotter teams in league...with Philbin and his poor staff, no less.
     
    cuchulainn and resnor like this.
  13. Fin4Ever

    Fin4Ever Season Ticket Holder Club Member

    9,297
    2,738
    113
    Aug 26, 2014
    Vero Beach, FL
    No I agree with you that good guard play will be plenty..but hey, if they turn out above average, all the better,right?
     
  14. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Yeah.. the stats are not worthless. They provide objective (as opposed to subjective) information you'd otherwise not have about Tannehill, so for that reason alone they're not worthless.

    And yes you implied all stats are worthless because one can always show there is some possibility not accounted for by every stat about the very subject one is talking about. That's true no matter what the context is because stats always throw away (relevant) information.

    Anyway, the debate with you so far has been pleasant but this last post of yours reminds me of the type of "defend-at-all-costs" stuff I won't get into with you. So let's end this pleasantly till the next time. As always you can have the last word.
     
  15. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    No I didn't imply that.

    Again, I specifically qualified my statement so you couldn't make that generalization, but you're ignoring that qualification and making the generalization anyway. This is a smaller scale example of what you're doing in this argument, which is ignore the specific qualifications in favor of over generalizations.

    Your stats don't fit the argument. You're arguing against stance that has specific qualifications and you're using stats that don't address those qualifications. That means they are invalid. It would be like arguing the how many people cheat on their spouses by using stats about orange blossom growth in the Florida summer.

    Also, I've been respectful, but objectively your last post wasn't.
     
  16. jw3102

    jw3102 season ticket holder

    7,760
    3,486
    113
    Sep 4, 2010
    Maui, Hawaii
    Because I see Luck as the type of top tier QB who can lead mediocre talent to the playoffs. I don't see Tannehill as a top tier type QB. I see him more as a game manager type QB, much like Smith in KC. With a lot of talent around him, Tannehill should be efficient enough to help this team get into the playoffs. I just don't see him ever leading the Dolphins to the Super Bowl.

    This thread started out by stating Tannehill has top 10 talent at the QB position. I just don't see it. He is and I think will always be a middle of the pack starting QB in the NFL. Certainly he is far from the worst starting QB in the league, but my desire is to see the Dolphins draft a young QB who can become an elite QB in the NFL and lead this team a Super Bowl win in the near future .

    Tannehill will suffice until that elite QB is found, but the sooner they find that QB, the sooner this team will become a legitimate SB contender, IMO.
     
    Finster likes this.
  17. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    Nothing meaningfull in 95% of these Tannehill threads.
     
  18. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    Just to compare Luck to Tannehill. Their career Passer ratings are 85.0 and 85.2. Without
    They play in different divisions, so I have done up some quick nasty averages.
    Tannehill's rating against divisional opponents is 74.4 indicating his division is much tougher than average his rating against the NFC is 91.5
    Luck's divisional average is 89.8 indicating he gets to feast on worse than average opponents 6 times a year and his rating against the NFC is 77.1

    Maybe Tannehill is already a better QB than Luck when you even out divisional differences. This is further reinforced when considering they've played basically a full season against the NFC and Tannheill is much better against the NFC than Luck.

    Luck has had the advantage of being labelled 'future HOFer' when he was drafted and Tannehill was labelled 'project'. People look at Luck and look for signs of greatness whereas people look at Tannehil and look for deficiencies he needs to improve, yet statistically Tannehill has been a better QB against common opponents than Luck.

    EDIT
    Another way of looking at it which isn't statistically robust, but useful as a thought experiment.
    Luck gets a bonus of about +5 passer rating points per divisional game by playing against the AFC South over the last 4 years. Thill gets a -10 passer rating points penalty per divisional game by playing against the AFC East.
    If their divisional opponents had NFL average defenses Luck's career rating would drop to around 83 and Thill's would go up about 88
    If Miami and Indianapolis had swapped divisions Lucks career passer rating would be about 80 and Thill's about 90
     
  19. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    Ah, so you were just coming in and being a seagull without actually reading the thread.
     
    cuchulainn, number21 and resnor like this.
  20. dolphin25

    dolphin25 Well-Known Member

    6,338
    2,400
    113
    Nov 22, 2014
    That was a good laugh. Thank you!
     
  21. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    I read enough.

    Maybe start commenting on the repetitive garbage posts rather than mine which is at a much lesser volume these days.

    Or maybe it is because I do not polish Ryan's knob and apologize for his flaws that you decide MY post was the one that needed attention. Nice agenda
     
    dolphin25 and jdang307 like this.
  22. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    I'm sorry I am a Ryan Tannehill realist and not an obitual nut hugger, in which case you would likely thank every post I make.

    Enjoy your "thread"
     
  23. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Edit: deleted so as to not further derail.

    What does "obitual" mean?
     
  24. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    My guess would be "regularly dead"?
     
    Pauly and resnor like this.
  25. Conuficus

    Conuficus Premium Member Luxury Box

    18,044
    19,676
    113
    Dec 8, 2007
    Well away from here
    Or, maybe he meant habitual.
     
  26. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    O rly?
     
  27. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    I'm sure that's what he meant.

    The question however, was what does "obitual" mean. Since its not a word, i thought my definition was good.
     
  28. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    My friend,

    I gave you an opportunity to do some actual analysis and contribute something new and meaningful to the debate, and you're still acting like a seagull. You claim to have read enough of the thread yet you've made no comment on the actual analysis of the thread which is a about whether Lazor's playcalling can be quantified and how it affected Tannehill's game, for good or bad.

    Your only contribution so far has been to fly in and drop insults on people who don't share your opinion about Tannehill. If you want to keep being a seagull complete the job and fly off now.

    However, I would prefer you to provide opinion and analysis.
     
  29. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    Obitual is a word. Not the word he meant, I'm sure. But it's a word.
     
    Pauly likes this.
  30. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015

    Are you new to this site?

    I have given months of analysis and provided facts that go by the waist side with moronic excuses EVERY. SINGLE. TIME. It is not a productive approach as it is met with ignorance and not a solid debate.

    Go back and read FinD and Res's last 150 posts about Tannehill and see if you can HONESTLY think they bringing reasonable analysis to this topic. You happen to take a very optimistic approach to Ryan so you lean toward that side and conveniently pass through the REAL problem with these threads.
     
    dolphin25 and jdang307 like this.
  31. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    So seagull,

    How do you explain the difference between Tannehill's passer rating according to game situation?

    I'm trying to isolate what the root cause of the inconsistent production is. There are basically four possible explanations for the difference in between his performance when he is behind and when he's not behind.

    1) null hypothesis, it's just random. With over 1200 attempts included random variation is a extremely unlikely to be the cause.
    2) Hypothesis 1: It's his team mates fault. His team mates are the same whether he is ahead or behind so [short version] this doesn't fly for me.
    3) Hypothesis 2: Tannehill melts under pressure. The Sherman years indicate that Tannehill is able to perform consistently regardless of game situation.
    4) Hypothesis 3: Lazor's playcalling turned to crud when he was under pressure. This is supported by game logs showing that Miami abandoned the run when behind

    You've said Tannehill is a very average QB. I disagree with that. Under Lazor he's been Dr Jeckyl when the phins aren't trailing and Mr Hyde when they are.

    PS If you are going to call people moronic you should know the difference between "way side" and "waist side" first. It doesn't help in establishing your intellectual credentials.
     
    cuchulainn, number21 and resnor like this.
  32. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    You're a riot.
     
    P h i N s A N i T y and Fin D like this.
  33. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    It's an archaic word. Most people can probably go their whole life without ever seeing or hearing that word.
     
  34. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    Well then we can thank Fin-O for providing us all with a little more enlightenment which is never a bad thing, even if unintentional.
     
    Pauly likes this.
  35. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    It's not enlightening when it's used incorrectly...which prompted my tongue-in-cheek question.
     
    Fin D likes this.
  36. Finster

    Finster Finsterious Finologist

    3,087
    2,038
    113
    Jul 27, 2013
    I have to disagree with hypothesis 3, QB17 was inconsistent in clutch situations with Sherman as well;

    '12) 3rd down=65.3 rating, trailing=75.9 rating, 4th qtr=83.0 rating
    '13) 3rd down=74.6 rating, trailing=81.5 rating, 4th qtr=64.3 rating

    career;

    3rd down=73 rating, lowest TD total of 1-3rd downs(22), highest INT total of 1-3rd downs(22).

    trailing=62.1 rating, has thrown 10 more TDs than when winning or tied(38vs48), has thrown about twice as many INTs than when winning or tied(19vs36).

    4th quarter=78.4 rating, 2nd lowest TD total per qtr(24), 2nd highest INT total per quarter(16).

    This is the crux of QB17, his inability to perform well in clutch situations, which is tied to his lack of leadership, he's above avg when not in clutch situations, and below avg when in clutch situations.

    This is Gase's and QB17's real challenge, getting past this, if he/they can do this, they're golden, but easier said than done.
     
    dolphin25 and Pauly like this.
  37. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    I'm not a believer in 'clutch', which is normally defined as stepping up in high pressure situations.

    I am however a strong believer in melting under pressure.

    You can't look at third down as a group in isolation. 3rd down passer raring is very heavily influenced by distance to go. So you have to compare Ryan Tannehill to NFL averages for down and distance to go before drawing any conclusions. Some people have said RT faces more 3rd and long than average, I don't know if that's true or not but it needs to be looked at first.

    Now if RT does melt under pressure it should show up in all pressure situations.
    If look at his rating in the last 2 minutes of a half in 2014/20125 under Gase his rating were 108.3/91.1 and 98.8 combined. Meaning RT performed slight;y better than average after the 2 minute warning.
    Under Sherman his ratings were 34.0/73.3 and 62.6 combined. So a bit less than average in 2013 but sucking balls as a rookie.
    This isn't consistent with the narrative that RT contunues to melt under pressure. It s a story of a QB who has made big strides in his game with experience.

    If we look at his stats for being within +/-7 in the 4th quarter his ratings for 2014/2015 are 99.3/81.2 and 90.7. Because of the small sample size the 2015 number is quite volatile and one dropped TD catch could bring his rating up to within a point of his season average. Looking at his combined number I think it's safer to draw the conclusion that RT performed about his average when within +/7 in the 4th quarter under Lazor.
    Under Sherman the numbers are 73.3/79.6 and 77.3 combined, Which is a drop off about 5% from his season averages.

    So what I take away from this is to say that in these high pressure situations he basically maintained his normal level of play under Lazor, which was an improvement over his years under Sherman.

    I discussed his rating when trailing in my initial post. Under Sherman RT's rating when trailing was within 2 rating points of his non trailing rating, which is better than the NFL average of losing 5 rating points when trailing.

    If RT is consistently bad in high pressure situations I'd expect it to be reflected in all his high pressure stats. Passer rating needs fairly large sample of attempts (about 300 by my estimate) to become stable enough to draw solid conclusions,so I don't care very much for slicing the data into very small slices and I try to avoid reading too much into smaller samples.

    Under Lazor there's some good evidence to suggest it was Lazor who melted under pressure, because the drop in passer rating when trailing is consistent over a large sample, and RT's stats in other high pressure situations is consistent with his normal performance.
     
    Finster and resnor like this.
  38. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    We can fairly easily find stats to discuss RT's performance as a passer. To some extent we can discus if the inconsistency has been due to coaching or individual performance. The stats can give us an indication but not a definitive answer on this question, so I'm all for debate on this point and looking at different ways to try and quantify an answer.

    I'm not sure of any way to quantify the effect of a QB as a leader. I agree it is a very important part of being a QB, but is there any way that it can be measured?
     
    Finster, Fin D and resnor like this.
  39. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    You try being a leader in your work environment when your boss and second supervisor are both actively undercutting you to the CEO, and seeking to get you fired and bring in a replacement, and all your coworkers know it.

    And let's not forget this started in his third season as an NFL player, after making huge strides.
     
    Fin D likes this.
  40. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Don't see the logic here. If one person can "melt under pressure" while another doesn't, then that means some handle high pressure situations better than others. That is, some are "clutch" while others "choke".

    There's a reason I make the argument about Tannehill not being clutch using the <2 minute and <4 minute trailing stats in the 4th quarter. That's the one stat I can defend.

    3rd down stats for Tannehill actually make him look average when adjusted for distance. Here's the data for 2013 and 2014:
    http://www.footballperspective.com/which-passers-were-most-effective-on-third-down-in-in-2013/
    http://www.footballperspective.com/which-passers-were-most-effective-on-third-down-in-in-2014/

    Basically they calculate league average percent success (getting a 1st down) as a function of 3rd down distance, then calculate the "expected" and "actual" 1st down success % for all starting QB's. Tannehill's "expected" and "actual" are essentially the same for 2013 and 2014, meaning he's an average 3rd down QB.

    As far as actual distance to go, it's not that bad. In 2013 he had it a bit worse than average while in 2014 he had it a bit better than average.
     
    dolphin25 likes this.

Share This Page