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

With all the Injuries to Quarterbacks do we...

Discussion in 'Miami Dolphins Forum' started by Dorfdad, Sep 17, 2019.

  1. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    You think the defense wouldn't have an advantage if the offense told them before every play if they were going to run or pass? Interesting.
     
  2. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    No, what I think is that the unpredictability you’re espousing is much more successfully achieved by staying at least close if not being ahead on the scoreboard, managing down and distance successfully, and varying offensive personnel on the field.
     
  3. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Rushing Y/C is definitely the better measure of how good a running game is, but the correlation between win% and rushing Y/C is really low at 0.1613 across NFL history. For passing Y/A it's 0.5525. And if you take the square of those numbers you get the proportion of variance in win% explained: 2.6% for rushing Y/C vs. 30.5% for passing Y/A.

    So rushing Y/C is the better stat to use to evaluate a running game but it actually doesn't matter that much for winning.

    Some stats account for this. For example, analyzing play-by-play data from 2009-2017 you find that both rushing Y/C and passing Y/A decrease the more "predictable" the play call is. For example, passing Y/A is around 6.7 when down by 12.5-17.5 points in the 4th quarter but gradually goes up to around 7.5 when point differential is large positive. Similarly, rushing Y/C is largest in the 4th quarter when down by double digits, around 4.75 while it's around 4.45 if it's quarters 1-3.

    So you definitely can see statistical evidence of decreased efficiency with greater predictability in play calling. In fact, just plotting league average rushing Y/C per year vs. league average rushing percent shows you this:
    [​IMG]

    What's interesting about that graph is that it suggests that you lose the "surprise" effect once you hit around 50% rushing. Beyond that you're just seeing what the rushing offense can do even if the defense (more or less) expects it. But rush less frequently and rushing Y/C increases because it's less predictable.

    Situational football definitely has a huge effect on efficiency. Just look at the splits for any year:
    https://www.pro-football-reference.com/years/2018/splits.htm

    Only data point there one should not take at face value is rushing Y/C in the 4th quarter because they include kneels. Otherwise one can see that the less important it is to stop the run (e.g. distance to go is higher or field position is farther away from the redzone) the higher both rushing Y/C and passing Y/A are.

    As to whether situational stats have a bigger effect than "keeping defenses honest" that depends on the situational stat. For something like rushing efficiency in the opponent's 1-10 yard line (1.9 Y/C) vs. on your side of the field (4.8 Y/C) situational football has a WAY bigger effect. But which down it is doesn't seem to affect rushing Y/C that much so in that case "keeping defenses honest" probably matters more if you go by the stats I posted above.
     
    Surfs Up 99, resnor and The Guy like this.
  4. keypusher

    keypusher Well-Known Member

    1,351
    448
    83
    Nov 29, 2007
    Well, sure there is, if you can average 8 yards per pass. And maybe even if you can't, because if you're averaging 4 you're going to have some -1's and -2's, and you're not going to be running then.

    It's really hard to move the ball all the way down the field with a bunch of fairly short runs. It's easier to hit a few longer passes, especially if the rules favor it, as they do. That's why the game can seem so strange to someone old like me who can remember the 70s, when the Dolphins would sometimes have 8-10 runs for every pass.
     
    The Guy likes this.
  5. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    Holy crap, we agree.

    I think on a practical level, is that when you're rushing at 50% of the snaps two things happen...1) Defenses play a more hybrid defense and 2) I think from a human anticipation standpoint, that run or pass quandary actually becomes MORE predictable.
     
  6. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    True, but I thought we were just talking about evaluating the run game to begin with haha.

    For the run I'd like a more specific data set to judge exactly how useful it is.

    For example, if I just had it in front of me I'd like to take the top 25% of teams in YPA Passing and see if those with a higher YPA Rushing as well had any additional success compared to their peers.

    Basically trying to eliminate teams who rush well but dont do much else and trying to see if there is synergy between both.
     
  7. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,815
    10,319
    113
    Nov 24, 2007
    Rockledge, FL
    I NEVER said that!! of course if you have a poor pass defense, you're leaving yourself open to have the score run up on you. It doesn't take a rocket scientist to figure that out...BUT THAT'S NOT WHAT YOU SAID!!!!!

    You said
    You have YET to explain how improving your team's pass defense also improves your team's offensive line. They are two completely different skill sets

    Defensive backs defend against the pass
    Offensive linemen block against opposing team's pass rush

    So please, don't deflect...don't evade...please explain to me like I'm a 6 year old how having 4, 5 and 6 Xavien Howards in your defensive secondary is going to make Witzmann, Davenport, Isadora, Webb and everyone else better blockers?
     
  8. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Dude... If you're down by 30, are you going to be running the ball or passing? You're going to be passing. Therefore, it's easy for the opposing defense because they know you're passing, and, since the oline is bad, it further exposes them because they can't good blocks long enough for duels tired to develop. A bad defense puts unnecessary strain on the offense, and vice versa.
     
    The Guy likes this.
  9. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    An offensive line’s job is more difficult when the opposing defense knows it has to pass protect more frequently to mount a comeback on the scoreboard.

    If the pass defense improves, opposing teams score fewer points, the above-noted situations decrease in frequency, and the offensive line’s job is made easier, thus improving its performance.

    Think about how easily the Ravens and the Patriots were able to tee off on the Dolphins’ offensive line, knowing how much the Dolphins had to pass the ball because they were down so big on the scoreboard both weeks.
     
    mlb1399 and resnor like this.
  10. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Absolutely. But the reverse is also true. The offense being unable to sustain drives and score points puts unnecessary pressure on the defense, and as the game wears on, the defense gets worse and worse.

    It's a symbiotic relationship.
     
    The Guy and Surfs Up 99 like this.
  11. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    There are definitely a lot of interrelationships in the game.

    However, we can examine this more closely. Between 2004 and 2018, the correlation between passer rating surrendered and points allowed was 0.76.

    The correlation between time of possession on offense and points allowed was -0.53.

    When we control for time of possession and look at the correlation between passer rating surrendered and points allowed, we get a meager increase to 0.78.

    cbrad can correct me if I’m wrong, but what I take this to mean is that, first, there is a much stronger relationship between passer rating surrendered and points allowed, and second, that variation in time of possession on offense is almost a meaningless factor in that equation.

    In other words, you could possess the ball a ton on offense, and if you are surrendering a high passer rating, you are very likely going to surrender a high number of points nonetheless.
     
    cbrad likes this.
  12. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    The easiest way to evaluate this is to look at all teams across NFL history and plot their win% vs. passing Y/A and rushing Y/C. It turns out a plane fits the data well (that is.. you don't need some curvy surface to fit the data). This is what you get:
    [​IMG]

    The answer to your question lies in the coefficients of that equation. For each increase in average Y/A passing you increase win% by 13.81% while for each increase in average rushing Y/C you increase win% by 2.53%, and the relationship is fairly stable whether you look at the best teams, the worst teams or those in the middle. So from this perspective increasing passing efficiency is 13.81/2.53 = 5.5 more important than increasing rushing efficiency. Naturally, this doesn't tell you any interaction effect which we were discussion earlier.
     
    keypusher, AGuyNamedAlex and The Guy like this.
  13. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Let's get extreme. If the opponent had a perfect passer rating, but only had two possessions, they probably aren't winning the game. Of course time of possession is important, whether your stats show it or not, because the more times you have the ball, and the longer you have it, the more chances you have to score points, and the fewer chances your opponent has to score points.
     
    The Guy likes this.
  14. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Right, but if that degree of variation in time of possession actually existed in reality, then the numbers I posted above would likely change. That degree of variation in time of possession doesn’t exist, however.
     
  15. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    That interpretation is correct, and incredibly the correlation between passer rating surrendered and points allowed is extremely stable regardless of era. Across NFL history it's literally 0.7598 lol..

    However, resnor was saying "sustain drives and score points". You're only addressing the first part when controlling for time of possession. The second part definitely affects all kinds of efficiency stats, including obviously OL performance (to some very hard to estimate degree) because of increased predictability.
     
    resnor and The Guy like this.
  16. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    The bottom line to me, is that you want to be dominant in every area. The reality is you cant be.

    There are specific areas that will increase your probability of winning, the other areas simply give an additional edge if you can be good at them too.
     
    The Guy likes this.
  17. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    The two key areas are offensive and defensive passer rating, in today’s game.

    The difference between the two is associated with 90% of the variation in win percentage in the league since 2004.

    This is why the highest-paid players in the league nowadays are quarterbacks, receivers, (usually) left tackles, defensive ends, and cornerbacks.

    Passing the ball well and defending the pass well equal winning.
     
    keypusher and AGuyNamedAlex like this.
  18. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    And correlation =/= causation.

    If Team A has the ball for 15 minutes and Team B has the ball for 45 minutes, that still doesn't tell you anything about who won. Team A might win because every possession they had they had Chuck yardage and big plays, while Team B loses because they had a ton of possessions but they were mostly 3 and outs. Or, Team A loses because all their possessions were 3 and outs, while Team B wins because they had sustained drives ending with points.

    One unit being on the field drastically more than the other often points to a problem with the other side of the ball, making that unit's job exponentially more difficult than it should be.
     
  19. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    You don't need stats to tell you that being good on offense and defense haps you win. Lol
     
  20. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    There are a whole host of deductions you can make about how to construct and manage an NFL team based on the post you quoted that go well beyond the mere “being good on offense and defense helps you win.”

    In fact if you rested only on the notion that “being good on offense and defense helps you win,” you may make your running backs, linebackers, defensive tackles, and interior offensive linemen your higher draft picks and highest-paid players, while making your quarterback, receivers, tackles, defensive ends, and defensive backs your lower draft picks and lowest-paid players.

    That approach would certainly be consistent with “being good on offense and defense helps you win,” but it would be entirely inconsistent with how the league truly functions in terms of winning.
     
  21. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    15 years of statistics about how the league functions as a whole won’t determine what will happen in a single game, no.

    But of course teams are trying to win more than just a single game, and as you increase the number of games considered, the statistics cited become more predictive about what will occur.
     
  22. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    When you control for points scored on offense, the partial correlation between passer rating surrendered and points allowed increases even more, to 0.82.
     
    cbrad likes this.

Share This Page