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

Possible Controversy Coming?

Discussion in 'Miami Dolphins Forum' started by Galant, Oct 6, 2019.

  1. KeyFin

    KeyFin Well-Known Member

    6,340
    8,302
    113
    Nov 1, 2009
    Are you guys sure we can't all just root for a different team this season? New Orleans, Seattle, Cleveland, Jacksonville...I don't really care. This non-stop argument over arbitrary points is driving me nuts!
     
    Sceeto, Puka-head and Irishman like this.
  2. Galant

    Galant Love - Unity - Sacrifice - Eternity Staff Member

    11,940
    7,402
    113
    Apr 22, 2014
    I'm watching the games, but I've turned my attention this year to college ball (for the first time ever) in order to try to look at the various QB's who may declare for the draft. I've really enjoyed it. Just the games themselves. Less talent and discipline makes for more ups and downs and big plays. Fun games to watch. They're typically on Saturdays though, so Sundays are still free.

    I also took the chance to invite a few friends to an NFL night to introduce them to American Football, and for that we obviously didn't watch a Dolphins game. I've also been turning Red Zone on - if you have access to it.
     
    resnor and KeyFin like this.
  3. cbrad

    cbrad . Club Member

    7,393
    8,673
    113
    Dec 21, 2014
    No one in statistics would agree with a phrase like "the art of statistics". Statistics is NOT an art, at least not to statisticians. It's almost like confusing the distinction between philosophy and science. Tell a scientist they're really philosophers and you've insulted them BIG time.

    And you don't need to determine likelihoods (which apply to hypotheses) before calculating probabilities (which apply to events). Statistics can calculate probabilities directly from the data. So that statement isn't correct either.
     
  4. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    Pertaining to the part in yellow in the above quote, what about the instances in which the QB on the losing team has the higher passer rating? What is passer rating measuring in those instances? It certainly isn't measuring winning.
     
  5. Vertical Limit

    Vertical Limit Senior Member

    10,599
    3,700
    113
    Nov 25, 2007
    I can watch football all day without having to root for anyone.. thats how i spend my friday nights, my whole saturday and my sundays... i give my family all the attention they want for 7 months and then on football season they gotta leave me alone
     
    Puka-head likes this.
  6. KeyFin

    KeyFin Well-Known Member

    6,340
    8,302
    113
    Nov 1, 2009
    I started watching college this season too and it's definitely different than the pros since there's so many teams and the rankings are down-right broken. I live in Clemson territory so that was an easy choice on who to root for, but they literally play nobodies for the rest of the season- what fun is that? So I don't watch full games of the team I'm rooting for which is strange as heck. I'm also rooting against Alabama since I want Tua to prove his worth against powerhouses, so I watch them more than Clemson...college ball is strange that way.

    My brother's a Florida fan and they're playing back to back to back top 10 teams, and I have enjoyed watching their ride so far. But once they lose a game to one of these powerhouses their season is basically over...which is why I HATE college football to begin with. A 10-1 or a 9-2 team is dominant, yet in college you're losers who blew the entire season. That really, really sucks in my opinion and they need a 16 team tournament like the NCAA. Once that happens though then I'll be all-in on college football again.
     
  7. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    It's. Measuring. Successful. Plays. Which takes more than the QB. Again, give me an actual example of a high passer rating without there being points scored. Regardless, what do YOU believe it's measuring?
     
  8. Galant

    Galant Love - Unity - Sacrifice - Eternity Staff Member

    11,940
    7,402
    113
    Apr 22, 2014
    LSU play Florida this week. Joe Burrow is rising in many estimations. I think I'll probably watch that and/or Alabama vs. Texas A&M
     
    KeyFin likes this.
  9. Vertical Limit

    Vertical Limit Senior Member

    10,599
    3,700
    113
    Nov 25, 2007
    Dont miss out on the Red River game at 12..
     
  10. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    What I can give you is significant variation in points scored, despite high season passer ratings:

    https://www.pro-football-reference....1comp=gte&c1val=100&c5val=1.0&order_by=points

    Note that points scored among those teams with high passer ratings (greater than or equal to 100) varies from 367 to 565, and the correlation between passer rating and points scored in that sample is a moderate 0.49, meaning that 76% of the variance in points scored in that sample is due to things other than passer rating.

    So again, what is passer rating measuring there? It isn't measuring points scored.
     
  11. KeyFin

    KeyFin Well-Known Member

    6,340
    8,302
    113
    Nov 1, 2009
    I haven't seen LSU at all this season so I'm looking forward to it. I'm NOT IMPRESSED with Florida's QB and he's playing on a hurt knee as well. LOL, he scrambles like Marino and it's painful to watch at times. Florida's defense is legit though and should make it a great game!

    I might go to the Florida/South Carolina game the week after...haven't fully decided yet since the tickets are so much. If we find a deal though then we'll be there!
     
  12. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    1,696
    1,475
    113
    Sep 12, 2015
    I believe the art aspect comes in terms of finding what data produces the most accurate information when assembled. In other words, creating the system of analysis itself.

    Also I'd consider it more an art applied to football because these arent hard numbers that actually mean something, they are human approximations of performance that may or may not always be accurate as to what happened. Especially in smaller sample sizes.

    Though I agree it's not really an "art" but a science. There is art within science though.
     
    Pauly likes this.
  13. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    It's measuring successful plays. How many times do I have to say it? You look at that and see "QB." But you ignore everything I've said in regards to what passer rating is actually grading versus what you think it's grading.

    But let's see a specific example of a QB in a game with a high rating but very few points scored.
     
  14. cbrad

    cbrad . Club Member

    7,393
    8,673
    113
    Dec 21, 2014
    Statistics is a branch of mathematics, not science. The application of statistics to actual data may be done by scientists instead of statisticians (many statisticians are actually quite bad at that part lol), but the statistical methodology itself is a form of math.

    The reason statistics is not an "art" by any stretch of the meaning of the word is because there is a ground truth: either what is stated is logically implied or it is not. That is, the inferences are either "correct" or "incorrect" given the assumptions made and with no room for interpretation, and you never see that in art.

    To be clear, statistical analysis of data may lend itself to multiple interpretations (i.e., different interpretations of the analysis might suggest different things about the physical system being studied), but whether that statistical analysis is valid or not does not lend itself to interpretation – either what is stated is logically implied by the assumptions or not – and it's the method of analysis that constitutes statistics.
     
  15. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    The last thing I'm doing is ignoring everything you're saying.

    First you said quarterbacks have high passer ratings because they tend to be on winning teams, that the "winning" the team does inflates their passer ratings. That was disproven.

    Then you said the same thing regarding points scored, that the "scoring" the team does inflates passer ratings. That was disproven.

    Now you're saying merely that high passer ratings are indicative of successful plays. Well no ****.

    And what player's ability do you think is responsible, predominantly, for the success of those plays?

    There is only one player who is the constant among all the plays that contribute to passer rating.
     
  16. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    1,696
    1,475
    113
    Sep 12, 2015
    I dont think you understand what calling something an art actually means. You're taking it very literally when it's just a phrase that means there is a specific talent and creativity to it. Not that it is literally an art and it's generally just meant as a compliment.
     
  17. Unlucky 13

    Unlucky 13 Team Rosen Staff Member Club Member

    26,100
    27,210
    113
    Apr 24, 2012
    Troy, Virginia
    I watch less and less college football every season, for various reasons. I used to be nearly as passionate about it as the NFL, and used to know who all of the upcoming prospects were for the next draft, even if I'm no scout. But these days, I have weeks where I don't watch a single game on a Saturday, and if I do, I don't get up or down at all about it.
     
  18. cbrad

    cbrad . Club Member

    7,393
    8,673
    113
    Dec 21, 2014
    It would be taken as an insult by most statisticians IMO because it suggests there's a subjective component to determining whether someone did statistical analysis correctly. That is, I don't think most people in the field could ignore the implications of labeling statistics in any way an art by just saying "oh we meant it as a compliment".
     
  19. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    You've proved nothing about QBs on winning teams having inflated ratings. That's the ****ing point.

    If you can't grasp that after the QB lets go of the ball, another player MUST CATCH IT in order for the play to be successful, then wtf are we still talking for? The ONLY WAY the QB is going to have a great rating is IF OTHER PLAYERS CATCH THE BALL AND SCORE TDS. Good teams generally have better players...this is why they are better teams. If you have better players, they are certainly making more catches and doing more after the catch than the less good players on bad teams. So yes, being on winning teams is beneficial to the QB and his rating. Honestly man, this is really frustrating.
     
  20. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    Last edited: Oct 10, 2019
  21. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    Dude. Stop. You haven't answered anything I've asked. You keep insisting that passer rating is a good judge of QB because QBs with high passer ratings correlate to wins... While refusing to accept the simple fact that the rating is not judging the QB, it's judging the success of the play...WHICH IS COMPLETELY DEPENDENT on a player OTHER THAN the QB. That's my point. So if you don't agree that's fine. Just stop responding with the same circular argument.
     
  22. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    That’s actually not what I’m saying. I’m saying that passer rating is a valid measure of quarterbacks’ individual ability, and that belief on my part has nothing to do with its correlation with win percentage.

    However, what you’re saying is that passer rating isn’t a valid measure of quarterbacks’ individual ability because of its correlation with win percentage, and as I’ve shown, that belief on your part is faulty.
     
    Last edited: Oct 10, 2019
  23. djphinfan

    djphinfan Season Ticket Holder Club Member

    96,511
    49,371
    113
    Dec 20, 2007
    Fu## them all
     
    Puka-head likes this.
  24. djphinfan

    djphinfan Season Ticket Holder Club Member

    96,511
    49,371
    113
    Dec 20, 2007
    No one is stopping you, root away
     
  25. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    You for some reason think you've shown that. If you can't understand that what the receiver does with the ball either increases a QBs passer rating or decreases his rating, then why are we still talking? You've shown that passer rating, which is grading how successful plays are, correlates to win%.
     
  26. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    Do you believe that the fact that receivers contribute to passer rating can invalidate it as a measure of quarterbacks' individual ability over large samples of play, or just small ones? If it's the latter, then we can simply agree.
     
  27. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    I have a problem using rating to predict win % IF you then translate that into saying that the rating is an indicator of actual QB ability. Passer rating is not grading QB ability. It is grading whether or not a play is successful.
     
  28. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    OK, well I have yet to say that win percentage is what validates passer rating as a measure of quarterbacks' individual ability. So we agree there.

    It looks like we may not be agreeing here, however:
    I'll agree that passer rating is a measure of successful plays and isn't always valid as a measure of QBs' individual ability over small sample sizes of play (one game, for example).

    However, over larger samples of play (several seasons, for example), the contributions to passer rating from players other than the QB don't vary significantly enough across QBs to invalidate it as a measure of quarterbacks' individual ability.

    What you're left with is one constant -- the quarterback -- and many other players and situations and teams over that larger sample of play. The common denominator among all of those other many players and situations, and across teams, is the quarterback.

    You can then safely say that passer rating, over that long haul, is measuring the ability of the quarterback. The fact that he needs someone to catch his passes then becomes irrelevant, because, unlike in the case of the single game, the variation in "catching passes" around the league isn't large enough over that larger sample of play to differentially affect passer ratings of individual quarterbacks.

    Over that larger sample of play, in terms of the validity of passer rating as a measure of QBs' individual ability, saying "the QB needs someone to catch his passes" then becomes no different from saying "the QB needs to tie his shoes before the game." Yeah, they all tie their shoes before the game, and they all have their passes caught at roughly the same frequency over the long haul.

    We don't run around thinking Dan Marino had a higher career passer rating than Chad Henne because he tied his shoes tighter. Likewise we shouldn't think he had a higher career passer rating because his receivers performed differently. We should simply think Dan Marino had a higher career passer rating than Chad Henne because he was better, and passer rating then becomes a valid measure of their ability.
     
    Last edited: Oct 11, 2019
    Pauly likes this.
  29. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    Disagree completely. I think you guys are way off base in regards to the individual abilities of even the worst QB in the league. As far as what they are capable of, even the worst QBb in the league is elite as compared to 99.999999% of the people who play football. If you ran a straight skills competition, I don't think you'd see a huge difference in the actual skills from the worst to the best. The team around the QB definitely is important. Tom Brady throwing to Randy Moss is going to produce very different ready us than Tom Brady throwing to Legedu. If you think that that balances out over the long term, then we disagree.
     
  30. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    Yeah well, then we certainly have to agree to disagree, because what you are saying there is that there are no differences in individual ability among NFL quarterbacks.

    That’s a pretty radical position that I imagine an extremely small percentage of planet earth would agree with, but of course like all of us, you possess the personal freedom to believe whatever you’d like, and no one can stop you.

    At least the discussion has evolved to the point where you simply put that out there, and now we can stop.
     
  31. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    I'm talking physical ability. Generally speaking, NFL QBs can make all the throws, if they're say throwing at targets, stuff like that. Obviously the mental side, being able to handle pressure, make good decisions quickly, especially under pressure, etc, is where QBs really separate from each other, imo. However, that's exactly what I'm saying passer rating is NOT grading.

    So you believe that Tom Brady is going to have very similar stats throwing to Moss as he would throwing to Greg Camarillo?
     
  32. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    If those quarterback characteristics make successful plays more likely, and if as you say passer rating is a measure of successful plays, then why wouldn’t passer rating measure those characteristics? Just because a characteristic isn’t physical doesn’t mean it can’t be measured numerically.

    I think Brady would have a higher passer rating throwing to Moss than Camarillo, but over the long haul, across the league, that variation in ability among receivers becomes equivalent enough that we shouldn’t attribute passer ratings to receivers, when the sample of play is that large.

    I don’t think for example that we should believe that Ryan Tannehill’s seven seasons with the Dolphins would’ve been just as good as Dan Marino’s first seven seasons in the league if Tannehill would’ve only had Duper and Clayton to throw to.

    I think what we should believe, instead, is that Duper and Clayton would’ve played much more like the receivers Tannehill had if Tannehill would’ve been throwing them the ball instead of Marino.
     
  33. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    So you believe that QBs make WRs physical abilities better, but the converse isn't true? So then you have to believe that the only thing you need to win is an elite QB. But we KNOW that isn't true. Of course Tannehill would have looked way better throwing to Clayton and Duper. It's the height of ridiculousness to suggest that what actually would have occurred is that the receivers would have looked worse. Great receivers run great routes, which allows them to get open, giving the QB better windows to throw into, and they generally have physical abilities (speed, better hands, higher jump) that allow them to catch balls that other receivers won't catch. If a QB is throwing to receivers who are open, any QB can generally make those throws. It's completely ludicrous to believe otherwise.
     
    Last edited: Oct 11, 2019
  34. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    But would he have played as well as Dan Marino, in terms of passer rating (era-adjusted)?

    And if not, why not?
     
  35. Finatik

    Finatik Season Ticket Holder Club Member

    1,596
    1,542
    113
    May 2, 2014
    SO Cal
    We need a stats forum to dump all these posts into.
     
    KeyFin, Hooligan, Sceeto and 2 others like this.
  36. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    No he wouldn't. But the rating isn't why. But, he would APPEAR, based on rating, to be better than he actually is, because, AGAIN, passer rating is NOT grading the QB and what he's doing. It's heading the success of the play.
     
  37. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    You’ve implicitly acknowledged in that post that passer rating is a measure of quarterbacks’ individual ability.

    If you gave Ryan Tannehill Dan Marino’s surroundings and Tannehill’s passer rating was nonetheless lower than Marino’s, then obviously the difference in their passer ratings would have to be due to none other than Marino’s greater individual ability.
     
  38. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    You're still not understanding. Just because passer rating is affected to any extent by individual ability in no way shape or form proves what you say it does. This entire argument is you trying to prove that passer rating isn't affected by the other players. This had been my argument with you all along. I'm not saying that ability has no affect on passer rating. I'm saying it's not ONLY heading the QB and his abilities. It's grading the success of the play, of which the QBs ability plays a part. I think you know exactly what I'm saying, you just won't admit that passer rating is not only, NOR PRIMARILY, rating the QBs ability.
     
  39. The Guy

    The Guy Well-Known Member

    1,635
    839
    113
    Oct 1, 2018
    I never said passer rating wasn't affected by other players. What I said was that over large samples of play, the variation in those other players' performance becomes equivalent enough across the league that the only thing left passer rating can measure -- again over large samples of play -- is quarterbacks' individual ability.

    Here's something you may agree with.

    The best season passer rating (117) of Tom Brady's career was in 2007, when he was part of an offensive machine with Randy Moss and Wes Welker, perhaps the best receiving tandem of Brady's career, and the team went 16-0.

    The best season passer rating (106) of Andy Dalton's career was in 2015, when he was throwing to AJ Green and Tyler Eifert, both of whom made the Pro Bowl that year, as well as Givoani Bernard as an effective third-down back.

    So, one might believe that based on those numbers alone, Brady and Dalton's individual ability isn't all that different. Both players played very well when surrounded by good players.

    However, when we look at the careers of both players, we find that Dalton's career rating is 88.7, while Brady's is 97.6, which is a fairly large difference.

    So what is passer rating measuring, between them, on a career basis?

    I think you have to conclude that, based on their career passer ratings, even with Moss and Welker and the offensive machine that was the 2007 Patriots, Dalton wouldn't have played as well as Brady did. And I think you have to conclude that, even with Dalton's typical surroundings, Brady would play significantly better than Dalton.

    Passer rating is simply measuring that difference between them, in individual ability.

    Again, however, over small samples of play, all that goes out the window, your position becomes correct, and passer rating becomes far more unreliable as a measure of individual ability.

    You can certainly find many games in which Dalton played better than Brady in terms of passer rating, but of course that shouldn't make us conclude that Dalton has more individual ability than Brady. We need a much bigger sample for passer rating to achieve its validity as a measure of individual ability.
     
    Last edited: Oct 11, 2019
  40. resnor

    resnor Derp Sherpa

    13,274
    7,491
    113
    Nov 25, 2007
    New Hampshire
    Uh, the Patriots have consistently had a better team than the Bengals. Take a look at who the Patriots like to use for receivers. They prioritize route running, smarts, and hands over height/speed/jumping because the first three things get receivers open and give the QB much better windows to throw in.

    There's literally no way you can argue that playing on better teams doesn't overall improve QBs ratings.
     
    Irishman likes this.

Share This Page