The Dolphins vs. the Top Four Playoff Seeds: A Comparison

Discussion in 'Miami Dolphins Forum' started by Tannephins, Jan 7, 2015.

  1. cuchulainn

    cuchulainn Táin Bó Cúailnge Club Member

    24,764
    41,770
    113
    Sep 7, 2012
    Hattiesburg, MS
    The OL was ranked 21st in 2013, but we had MUCH better Guards and Center. Just no Tackles.

    Losing Pouncey at Center just before the season, then later Albert at LT were the real killers to the 2014 OL. At that point every OL position was shuffled and was sub-par from there on out. You also saw Tannehill running less as the reads weren't there.
     
  2. emocomputerjock

    emocomputerjock Senior Member

    5,649
    1,853
    113
    Nov 23, 2007
    DC
    He does have an exceptional success rate. Matter of fact, he's rated top 5 in DVOA, DYAR, and #3 in WPA. In any event, the line was rated much better at run blocking than pass blocking by sources that aren't pff, and other sites are tracking that Miami lead the league in combined sacks hits and pressures as well. It'd seem that the metrics are consistent.
     
  3. Fin-Omenal

    Fin-Omenal Initiated

    36,936
    10,264
    0
    Mar 25, 2008
    Thee...Ohio State University

    Man if he gets that kind of attention on the read option than our passing game must be lethal.
     
  4. dolphin25

    dolphin25 Well-Known Member

    6,852
    2,718
    113
    Nov 22, 2014
    Miller does a good job catching the ball. I remember him not catching some thrown at his feet.

    He does run much better out of the spread offense, but that has a lot to do with the line.
     
  5. Fin-Omenal

    Fin-Omenal Initiated

    36,936
    10,264
    0
    Mar 25, 2008
    Thee...Ohio State University
    Whoa, whoa, whoa hold on!!

    PFF said Miami had the 31st ranked run blocking in the league, all you need to know. Lamar Miller is a god
     
    dolphin25 and roy_miami like this.
  6. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    Those figures correlate relatively poorly with measures of quarterback play (DVOA, QB rating, YPA, A/YPA, N/YPA, AN/YPA), and so there seems to be a great deal regarding measurements of play at the quarterback position that has nothing to do with measurements of play of offensive lines.
     
  7. cuchulainn

    cuchulainn Táin Bó Cúailnge Club Member

    24,764
    41,770
    113
    Sep 7, 2012
    Hattiesburg, MS
    Well then... since we know PFF rankings and stats can't be right, let's look at Pro Football Reference.

    PFR: Dolphins Offense ranked 11th in 2014, up from 26th in 2013.

    2014: Scored 388 points (24.2/g), 11th of 32 in the NFL.
    2013: Scored 317 points (19.8/g), 26th of 32 in the NFL.

    Miller, thanks to the 178 yard day against the jets to end the season, leaped up to #10 in rushing yards on the season. Not bad for a guy who averages 45 yards and change a game.

    As a team, we rushed for 432 more yards in 2014 than in 2013.
     
    resnor likes this.
  8. Fin-Omenal

    Fin-Omenal Initiated

    36,936
    10,264
    0
    Mar 25, 2008
    Thee...Ohio State University

    I'm confused on what you're trying to say.
     
    roy_miami likes this.
  9. emocomputerjock

    emocomputerjock Senior Member

    5,649
    1,853
    113
    Nov 23, 2007
    DC
    I propose that's because pressures, sacks, hits and hurries aren't even included in the formulas for any of the above measures.
     
    Tannephins likes this.
  10. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    And pressures, sacks, and hurries (or QB hits) aren't included in those measures of QB play (DVOA, QB rating, YPA, A/YPA, N/YPA, AN/YPA).

    But if pressures, sacks, and hurries (or QB hits) had something to do with those measures of QB play, then what you'd expect to see across the league (in a sample of all of the league's QBs) is that as those measures of offensive line play increase (get worse), the measures of QB play decrease (also get worse).

    That isn't the case, however. The measures are largely unrelated, which suggests that there is a great deal to do with measures of QB play that has nothing to do with measures of offensive line play.
     
  11. resnor

    resnor Derp Sherpa

    17,097
    10,700
    113
    Nov 25, 2007
    New Hampshire
    Are we SERIOUSLY arguing the pressures, sacks, and hurries don't negatively affect QB play? Are we arguing that when a QB can't step into a throw because he has defenders in his face, that it doesn't affect the throw? Are we arguing that QBs that are constantly getting hit don't get nervous in the pocket, and throw earlier, because they are trying to avoid pressure that is coming? If we are trying to use stats to say that this stuff doesn't occur, then there is something wrong with our stats.
     
    speed and jldolfan like this.
  12. Fin-Omenal

    Fin-Omenal Initiated

    36,936
    10,264
    0
    Mar 25, 2008
    Thee...Ohio State University

    And there is....

    This isn't baseball so stats don't always paint the correct picture, in watching games this year I have a hard time believing there was no significant difference in time to throw between Brady and Ryan. Just like I don't believe Ryan throws a good deep ball to anyone other than Wallace because of some 20yards in the air barometer.

    Figures lie and liars figure.
     
  13. resnor

    resnor Derp Sherpa

    17,097
    10,700
    113
    Nov 25, 2007
    New Hampshire
    Well, Tannehill seemed to throw a bunch of good deep balls (not your version, probably) to guys not named Wallace, especially in the last 3-4 games.
     
  14. Fin-Omenal

    Fin-Omenal Initiated

    36,936
    10,264
    0
    Mar 25, 2008
    Thee...Ohio State University
    I liked the fact we finally attempted those throws, while I wouldn't say its a bunch it certainly was better than any other few game stretch in his career. Hell, I even liked the throws to Wallace save the 50 yard completion that should've been an 80yd TD. Hopefully it carries over to next year, because I think most would trade a great completion percentage with a 5 YPC attached to it for an early 60's with some chunk yards.
     
    dolphin25 likes this.
  15. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    The question is how much those offensive line variables affect overall QB play.

    When we measure offensive line play and QB play, and those measurements aren't related to each other very strongly, what can we say about how much offensive line play affects QB play, in a way that can be measured?

    It could be that we can't measure offensive line and/or QB play reliably, or it could be that we're measuring it reliably, and offensive line play has less of an effect on overall QB play than is believed by some people.

    In other words, it's quite possible that, as you said, "something is wrong with our stats," and it's also quite possible that something is wrong with the belief that offensive line play affects overall QB play as much as it's thought to.
     
  16. resnor

    resnor Derp Sherpa

    17,097
    10,700
    113
    Nov 25, 2007
    New Hampshire
    I'll put it this way...I bet if you, for instance, put Tannehill on a field with Mike Wallace and the offensive line, with no defense, that Tannehill would have no issue chucking deep balls into Wallace in stride. So, the question becomes, what changes when there is a defense on the field?
     
  17. roy_miami

    roy_miami Well-Known Member

    1,385
    560
    113
    Oct 11, 2013
    Russell Wilson has already broken some passing records, just imagine what he could do behind the best o-line in the league. I mean, there is a good chance he could the best passer in the history of the sport if he was only under Peyton Manning type pressure.
     
    dolphin25 likes this.
  18. DolphinGreg

    DolphinGreg Season Ticket Holder Club Member

    5,227
    6,527
    113
    Dec 7, 2014
    I think the only easy thing to conclude is that the more mobile/agile the QB, the less protection is necessary. Now, when I say " mobile" I'm not talking about Cam Newton and Colin Kaepernick, I'm talking about being able to move around within the pocket, evade pressure and escape when necessary. That's the basic skill that all good QBs have to some degree.

    Wilson, Luck, Romo and Rodgers all put up great stats and look fantastic because they are able to move away from the pressure exceptionally well. None of those O-lines are as good as fans would think they are in terms of pass-protection. The mobility of each of those 3 QBs is making the protection look better than it is.

    In the case of Brady and Manning, you have 2 QBs who aren't very mobile at all. Any amount of pressure will degrade the success of those QBs quite noticeably. For that reason, they have been supported by some of the better pass-blocking O-lines in recent memory.


    The problem Miami seems to have is that they have "average" pass protection (a la Seattle, Indy, Dallas and Green Bay) while at the same time have a QB that doesn't show a lot of mobility (a la Denver and NE). Therefore, one can safely conclude that either (A) Tannehill needs to start moving around more like Wilson, Luck, Romo and Rodgers or (B) Miami needs to set about building an elite pass-blocking O-line.
     
    roy_miami and Tannephins like this.
  19. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    Certainly at minimum I think one has to believe that there are traits QBs can possess that mitigate the effects of poor offensive line play.

    The question is, how many and how much of those traits does Ryan Tannehill possess, and is what we're seeing on the field a problem caused by a deficiency of his, a deficiency in offensive line play, or both?

    And if it's both, how much of an effect is caused by the former (Tannehill), and how much is caused by the latter (the line)?
     
  20. roy_miami

    roy_miami Well-Known Member

    1,385
    560
    113
    Oct 11, 2013
    I think one aspect of pocket presence that goes unnoticed is the ability to not just feel a rusher bearing down on you but how likely you are to get pressured in a particular situation. I bet on their best days Brady and Manning can feel that the defensive end is gaining momentum on this third and ten so they get rid of it quick and maybe the next third and ten they "know" the protection will hold up so they hold it a bit longer.

    Tannehill is not just not as good as the greats, nor do I expect him to be, but I think he's actually terrible at sensing when he has to get rid of it and when he can hold it a bit longer. In the Buffalo game when they sack/stripped us and stole the win he should have been on high alert. He just changed the play from run to a pass. It was getting down to crunch time, one more first down and we likely win, the pass rushers are pinning their ears back and we've got Clabo one on one against Mario Williams. Multiple alarms should have been going off in his head, he had to throw it almost instantly or start running for his life. Instead he just stood there and let a defender take the ball from his hand.
     
  21. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    And if you believe the Dolphins' problems with sacks have more to do with Ryan Tannehill than the offensive line, you're likely to be more aware of and remember plays like the one bolded above.

    If on the other hand you believe the Dolphins' problems with sacks are due more to the offensive line than to Ryan Tannehill, then you're likely to be more aware of and remember the sorts of plays the Dolphins' offense had in which any quarterback would be engulfed by pressure, with little chance to react at all.

    In other words, your belief determines what you "see."
     
  22. dolphin25

    dolphin25 Well-Known Member

    6,852
    2,718
    113
    Nov 22, 2014
    Where when?

    I think I recall a single pass to Hartline where he didn't have to stop and wait for the ball.
     
  23. dolphin25

    dolphin25 Well-Known Member

    6,852
    2,718
    113
    Nov 22, 2014
    I think it is some of both. Some certainly on Tannehill and some on the OL. With 2 OL coaches, 2 OC's, and 9 new OL players you get the same results...... you gotta start wondering ..
     
  24. emocomputerjock

    emocomputerjock Senior Member

    5,649
    1,853
    113
    Nov 23, 2007
    DC
    This is true, but I believe part of that is because the raw offensive line stats don't reflect other correlations that influence the primary QB stats. For instance, I recall reading an article that showed that QB hit % was pretty strongly correlated with INT numbers (completion % was too). The number of sacks doesn't directly correlate to the primary QB stats, but QB sacks + sack yards lost do impact net ypa, and unsurprisingly Cam Newton, Tannehill, and Kapernick all had high sack numbers and yards lost, and that dropped their..."acceptable" YPA numbers drastically. Unsurprisingly, none of those 3 QBs had a winning season. I'm sure there are correlations to be found between offensive line play and QB play, but it won't be an easy one to one. Some other issues to consider that I'm not sure can be correlated are offensive play design and philosophy, which may also be related to line play but not in any way that I can see being teased apart to give a better picture of how either would impact QB play.
     
    cuchulainn likes this.
  25. DolphinGreg

    DolphinGreg Season Ticket Holder Club Member

    5,227
    6,527
    113
    Dec 7, 2014
    This is when in my view you have to start measuring a player's instincts and reactions on a play-by-play basis. The truth is that you're probably not going to find statistics that speak to what's happening in the pocket with sufficient clarity that it might be possible to identify what a particularly lineman or the QB himself should have, had to or decided to do given some set of circumstances.

    I also think that at Ryan Tannehill's age and level of experience you start to see diminishing returns with "coaching." Tannehill has been in the league for 3 seasons during which he's been under pressure quite a bit. He's shown that he's tough and can take a hit but we've also had 3 years to see that his level of improvisation is more limited than we'd like. I think it's possible that he improves, but I don't think he'll get to the level of Wilson and Rodgers are at now nor will he ever show the level of fluidity and comfort within the pocket that Andrew Luck does.

    I look at Joe Flacco as a QB that has evolved his abilities under pressure quite a bit but he's also willing to sling the ball quite a bit more than the average QB. Flacco doesn't hesitate to let it go. He trusts his arm as much as Luck and Rodgers do. Each of those 3 has confidence in not only their arm strength but their ability to fit the ball into tight windows, even on deeper throws. Ryan Tannehill does not exhibit that type of confidence and propensity to sling it.

    What ultimately demonstrates his ceiling to me, is that when Tannehill has been aggressive we've seen his INT numbers go up. We start to see him throw into double coverage down-field as well. We see him throw the ball into places where the WR has to knock it down so that it's not intercepted. I think what people are realizing is that the way Philbin and Lazor structured the offense this year was not out of their own conservative nature but more so out of a necessity to help Ryan Tannehill.
     
  26. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    I think the issue is that sacks most certainly affect an offense, but they don't affect individual QB play anywhere near as strongly.

    For example, the correlation between sacks allowed by offensive lines and net YPA in 2014 is -0.48, but the correlation between sacks allowed by offensive lines and YPA is only -0.20.

    In other words, obviously sacks are going to influence net YPA, because sacks are an element of net YPA, but when we remove sacks from the measure of quaterback play and focus more on what only the quarterback is doing in the passing game (as measured by YPA, and not net YPA), that relationship diminishes considerably.

    Likewise, the combination of sacks, QB hits, and QB hurries (PFF's "Pass Blocking Efficiency" statistic) correlates with YPA at only 0.27, and at 0.27 as well with QB rating.

    In other words, 92.7% of the variance in both YPA and QB rating is not associated with "Pass Blocking Efficiency." There are other variables that are accounting for the variance in YPA and QB rating a great deal more than pass blocking, if in fact PFF is measuring it accurately and reliably with that statistic.
     
  27. emocomputerjock

    emocomputerjock Senior Member

    5,649
    1,853
    113
    Nov 23, 2007
    DC
    Hmm, what about offensive line WPA? They seem to have an interesting way of looking at things. Of course, WPA isn't predictive by any means, but it has been my personal favorite metric for individual skill players. I haven't looked at the rankings for units as a whole (I usually stick with DVOA for that).
     
  28. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    PFF attributes only a portion of teams' sacks to offensive lines in its "Pass Blocking Efficiency" statistic. The correlation between QB rating and sacks per pass dropback attributed to QBs in 2014 is -0.37. The correlation between QB rating and sacks per pass dropback attributed to offensive lines is -0.29.

    In other words, QB rating appears to be more strongly related to quarterbacks' responsibility for sacks than to offensive lines' responsibility for sacks. Said somewhat differently, the QBs who are responsible for a fewer number of sacks per pass dropback tend to have higher QB ratings than the QBs who have offensive lines that are responsible for fewer numbers of sacks per pass dropback. For YPA there is no difference in those correlations.
     
  29. resnor

    resnor Derp Sherpa

    17,097
    10,700
    113
    Nov 25, 2007
    New Hampshire
    How do you decide if its the QB or oline who's responsible fir the sack?
     
  30. roy_miami

    roy_miami Well-Known Member

    1,385
    560
    113
    Oct 11, 2013
    Its both. Almost always both. The defensive players job is not to beat the o-line, its to get the QB. Its the defense vs the QB, the o-line is just a layer of defense.
     
    shamegame13 likes this.
  31. shamegame13

    shamegame13 Madison & Surtain

    3,451
    903
    113
    Dec 15, 2014
    One hand washes the other.
     
  32. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    I have no idea. I'm just going on PFF's attribution of sacks to QBs and offensive lines in post #68.
     
  33. cuchulainn

    cuchulainn Táin Bó Cúailnge Club Member

    24,764
    41,770
    113
    Sep 7, 2012
    Hattiesburg, MS
    Yeah, but PFF also separates those from the OL and puts them onto the backs and QBs as well.

    Of the total 46 sacks and 98 hits, the OL was deemed responsible for 34 sacks, 50 hits, and 144 hurries.

    The other 12 sacks and misc hits were attributed to the backs and TEs in pass protection, or directly on the QB himself.
     
  34. resnor

    resnor Derp Sherpa

    17,097
    10,700
    113
    Nov 25, 2007
    New Hampshire
    So, if we figured one sack apiece for each tight end and back, Tannehill would be responsible for maybe 8 sacks, maybe? That seems high to me, but for the sake of argument.
     
  35. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    Again, however, none of the measures of QB play that are independent of sacks (QB rating, YPA, adjusted YPA) is strongly correlated with any of those variables across the league. In other words, league-wide, those measures of QB play have a great deal more to do with things other than offensive line play, if you deem what I bolded above to be an accurate and reliable measure of offensive line play.
     
  36. resnor

    resnor Derp Sherpa

    17,097
    10,700
    113
    Nov 25, 2007
    New Hampshire
    I think it's a bold claim to make that sacks have no effect on rating or YPA. You're claiming that getting sacked/hit doesn't result in quarterbacks making quicker throws (to shorter routes) and doesn't routinely have them throwing while under duress.
     
  37. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    Like I said in post #55, there are really only two explanations: 1) quarterback play and/or offensive line play aren't being measured accurately or reliably by these statistics, or 2) pressure (hurries, hits, sacks) on QBs has less of a relationship with those measures of QB play than some people believe.

    Keep in mind that when a QB is sacked, he doesn't make a throw, and the measures of QB rating and YPA, for example, involve only plays in which throws were actually made. This is why I said in one post above that sacks may certainly affect offenses, but they may affect individual QB play less than many people may believe, and clearly the relatively weak correlations between sacks and QB rating and YPA support that.

    In other words, it's entirely possible that a QB can be sacked a lot and still post very high QB rating and YPA numbers. That QB's offense may suffer, but he can still individually post very good numbers in those particular areas, because the plays in which he made throws have nothing to do with the ones in which he was sacked.
     
  38. resnor

    resnor Derp Sherpa

    17,097
    10,700
    113
    Nov 25, 2007
    New Hampshire
    And it's entirely possible that a QB who's under duress on every dropback has his play affected, and posts low numbers, but it isn't tracked by those stats. I think what you're advocating would apply to QBs who are used to having good protection, and aren't often hurried/hit/sacked. So, when they are hurried/hit/sacked, it's an aberration, and doesn't affect their mindset, or the overall gamelan for the offense. What you are advocating is really a convoluted way of once again absolving the oline of their abhorrent play, by saying that their play has nothing to do with how the QB plays. It's again, saying that the offensive woes are all on Tannehill. I'll point you to Brady and Romo as two shining current examples of oline play stabilizing, and allowing the quarterbacks to do great things. Did everyone forget how terrible Brady looked early in the year, with a different line every other series, and everyone talking about Brady being done?
     
  39. Tannephins

    Tannephins Banned

    1,818
    572
    0
    Dec 23, 2014
    I don't think there were really any major offensive woes. The team moved from 18th to 7th in the league in offensive DVOA, scored nearly five more points a game, and Tannehill improved a great deal, despite whatever deficiencies have been perceived regarding the offensive line's play.
     
    resnor likes this.
  40. roy_miami

    roy_miami Well-Known Member

    1,385
    560
    113
    Oct 11, 2013
    The Seahawks are now a staggering 27-1 in games decided by eight or more points with Wilson at QB.
    Rodgers: 50-15
    P. Manning: 108-38
    Brady: 102-24

    Romo: 49-26
    Luck: 15-11
    Flacco: 44-13
    Cam Newton: 21-15

    Dalton: 22-12
    Kaepernick: 20-10
    Rivers: 60-24
    Matt Ryan: 34-23
    Eli: 58-48
    Tannehill: 12-13
    Orton: 24-23
    Fiedler: 27-13
    Henne: 5-30
    Bledsoe: 51-42
    Brady's first 3 seasons: 19-8
    Sanchez's first 3 seasons: 16-10

    Not quite sure what to make of this. The Seahawks clearly have to have one of the greatest records in the history of the game at 27-1 over three seasons, and I guess the record reflects the fact that they great in all phases.

    I think Tannehill's floor is Kyle Orton and his ceiling is probably a Matt Ryan or Drew Bledsoe and right now he's equal with Andy Dalton.

    Bengals and 49ers both have teams similar to the Seahawks (coaching, defense, running game) so what the record shows is how much better Wilson is than Dalton and Kaepernick. Wilson is worth 2-3 more wins a season than Dalton. And the Bengals talent/coaching is worth 2-3 more wins a season than the Dolphins.
     

Share This Page