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

Here's What's Wrong With Cutler...Give Me A Minute...Wait I Know There Has To Be Something...

Discussion in 'Miami Dolphins Forum' started by Rickysabeast, Aug 8, 2017.

  1. danmarino

    danmarino Tua is H1M! Club Member

    15,357
    20,976
    113
    Sep 4, 2014
    "Nothing"? If being an 11 year starter in the NFL or an 11 year back-up in the NFL "means nothing" to you I think you need to take off your blinders.
     
    Last edited: Aug 18, 2017
  2. bigballa2102

    bigballa2102 Well-Known Member

    1,318
    638
    113
    Oct 22, 2013
    So CAL

    Well the only real thing to comment here which says enough about the lack of information you have is cutlers record

    Jay Cutler 141GP 69W 72L .489PCT

    Im no math major but I would say when you have more losses than wins that puts you under .500!!!!!!

    good thread!!!
     
  3. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    Sigh.
     
    Bumrush, TheFinWR, danmarino and 3 others like this.
  4. bigballa2102

    bigballa2102 Well-Known Member

    1,318
    638
    113
    Oct 22, 2013
    So CAL
    Like I've said before many times, the red sea parted all the stars aligned for Jay, if gase was anywhere else and not a head coach jay would be broadcasting this year. Sometimes its better to be lucky than good I guess LOL
     
  5. Finster

    Finster Finsterious Finologist

    3,087
    2,038
    113
    Jul 27, 2013
    I see it as 11 years of teams making a mistake, in case you hadn't noticed, Cutler never panned out in those 11 years, he's played his entire 11 years in the "new age" of QB friendly football and has a losing record and an 85 rating, with 208 TDs and 146 INTs.

    The fact that he started for 11 years means nothing if he doesn't have production, Cutler is one of the longest running 1st round bust QBs in NFL history, this is what happens when you "chase" potential, Cutler is a cautionary tale, he's never been anything but a back up quality QB.
     
  6. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Cutler is garbage, but Moore is good?

    The logic around here astounds me sometimes.
     
  7. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    There's nothing wrong with Cutler. He's not great, he's not bad. He's just around average mostly.

    For some things he's a bit above average:
    - Adjusted passer rating (see post #38) he's a bit above average at 90.39 while adjusted league average was 87.6.
    - TD% he's at 4.6% while league average from 2006-2016 was 4.27%.
    - Completion percentage.. Cutler's at 61.9% while league average from 2006-2016 was at 61.3%
    - YPG Cutler averages 233.6 while league average from 2006-2016 is 226.3

    For some things he's a bit below average:
    - Win% (sample size is large enough to use this) he's at 48.92%
    - INT% Cutler's at 3.3% while league average from 2006-2016 was at 2.8%
    - TD:INT ratio Cutler's at 1.42 while league average from 2006-2016 is at 2.

    I'll just note that all these stats should be adjusted like I did with passer rating (and of course win% doesn't need adjustment), but the overall picture isn't going to change much whether I do that or not (just takes more time to adjust). Point is.. Cutler's an average QB.

    Oh.. and look at where Cutler ranked among starting QB's from 2007-2015 where he had enough games started to qualify:
    https://www.pro-football-reference.com/years/2015/passing.htm

    Using passer rating as the ranking metric you get a range of 12-21 over 9 years with an average of 16! (that is.. an even 16 lol).. which btw is the same rank he had in 2015 with Gase. Cutler = Average QB!
     
    Rock Sexton likes this.
  8. danmarino

    danmarino Tua is H1M! Club Member

    15,357
    20,976
    113
    Sep 4, 2014
    I knew I should have put "" around he, but I figured he'd understand the point. I figured wrong. lol
     
    resnor and Fin D like this.
  9. danmarino

    danmarino Tua is H1M! Club Member

    15,357
    20,976
    113
    Sep 4, 2014
    Starting for 11 years in the NFL is anything but a bust. When you look at Cutler's career you should really do more than look at his stats...(And I'm saying that as a HUGE stats guy). He has played on some really dismal teams, all the while learning multiple different offenses under multiple coaching staffs. Are you telling me that if someone with his skills had been drafted by the Steelers or Patriots they wouldn't have been successful?
     
  10. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Win % is not a QB stat, I don't care how large the sample size. When you win 5 games, and your defense is giving up 35ppg, you don't blame the QB.
     
    danmarino likes this.
  11. Walter

    Walter Well-Known Member

    361
    238
    63
    Aug 6, 2017
    Ya well Dan Marino never won a Superbowl.......

    Yet he is considered one of the best to ever play the game .....and rightfully so.



    Just half qtr. of Cutler and one can realistically see the good possibility of having a much more knowledgeable and skilled QB than the one we've seen on the field the last several years.....Got a feeling if Tanne played for another team at this stage from past evidence not many in here would be ringing the bell for him. Now we have heard Gases rave about his improvement but for us fans it was still only words but all our fingers were crossed. Go figure positive words from the head coach not really something we don't pretty much always expect unless is out right blatant screw ups. Only thing Tanne my have over Cutler is age....
    Like someone has already stated...wait most people have already stated for the last many years ....
    More a concern for the OL which would give any QB a problem....

    Well I also have some concern a DB its weird how we have such a hard time getting off the field after third downs , many times third down and long.....very long.....
    But whats that famous saying ? Rome was not built in a day or year , though maybe after 10- 12 years sufficient......luckily we have a new architect...
     
  12. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    It's not a matter of opinion resnor, it's math (or simple logic). If you influence the probability of an event occurring (win or loss), then the observed probability (frequency) of that event occurring (in your presence) is an indirect measure of the degree to which you influence that event. How to infer how much influence you have depends on how much influence other factors have.

    That logic is used in all kinds of math as well as in computer science. In math look up Bayesian inference. It tells you how to optimally update the likelihood of different hypotheses being true given any new observation (of an event occurring). In computer science look up forms of machine learning such as artificial neural networks or reinforcement learning. They all depend on this logic.

    Sample size is important because there's random variation that you need to average out before you can expect to see a certain sized effect. In general, the greater the influence you have, the smaller the sample size needed to see a certain effect. So for QB's you'd need fewer games than for LB's or OG's (probably for them it's far larger than the number they ever play). So while you're right in the specific example you gave, with sufficient sample size you can see the QB's influence.

    Actually, this is the only theoretically objective way I can see for estimating the relative influence of different positions on W/L record. It doesn't work in practice because you don't have large enough sample size for most positions. For QB's it seems like we do as I pointed out earlier (minimum 80 games seems to work.. maybe less too).
     
    Last edited: Aug 18, 2017
    Rock Sexton likes this.
  13. danmarino

    danmarino Tua is H1M! Club Member

    15,357
    20,976
    113
    Sep 4, 2014
    Of course the QB influences the game. Every player does, as you mentioned. However, can you determine if a QB is good because the team is good or if a QB is bad because the team is bad? I'm sure you remember the post I made that shows the Patriots win games with or without Brady running the offense. So, if that's the "math", how do we know that Brady is, as some claim, a top 3 all-time great QB? I understand that it's almost as if I'm knocking Brady because his teams have always been really good, but I'm not. I'm knocking Brady by looking at his individual performances and over all athletic ability. For example, if Brady's teams have been so great AND Brady himself is so great, why is it that Brady holds not one single major individual NFL QB passing record? Not one. You'd think that a GREAT QB playing on a GREAT team would be able to hold at least one major passing record, right?
     
  14. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    That's a separate question.

    In practice, because a QB plays on only a small number of teams and the strength of his surrounding cast isn't as random as across the league from game to game or from year to year, you can't just look at W/L on its own. You have to take into account all kinds of other factors, and I think you know I've done this in the past.

    For example, I've looked at the correlation between points allowed and passer rating for each year, and adjusted passer rating for each year accordingly, etc.. I've also looked at z-scores (standard deviations above or below the mean) for all kinds of stats, which automatically adjusts for league-wide distributions, etc..

    So never would I say use win% alone. I'm just saying every "QB stat" is a "team stat". Passer rating is, Y/A is, YPG is.. name it. So either there are no QB stats or you use any stat where the QB has enough influence and start looking at correlations and league-wide distributions to see where a particular QB ranks.

    Case in point: Brady is tops on career win% and total wins, but I think you remember well when I adjusted for defense Brady was still pretty high, but multiple other QB's were above him.
     
    Last edited: Aug 18, 2017
    danmarino likes this.
  15. Rock Sexton

    Rock Sexton Anti-Homer

    2,553
    1,793
    113
    Mar 14, 2015
    That statement could potentially be adjusted to " .... is good for THIS team."
     
  16. Rock Sexton

    Rock Sexton Anti-Homer

    2,553
    1,793
    113
    Mar 14, 2015
    Matt Moore has played under some of those same dismal conditions. Why does he not get any BOTD, but Cutler does? Simple - falling in love with arm talent.

    Cutler does not strike me as a guy who would thrive under a structured BB regime. He would have gotten benched a few times for some of the boneheaded decisions he's shown that have killed momentum for his teams.
     
  17. danmarino

    danmarino Tua is H1M! Club Member

    15,357
    20,976
    113
    Sep 4, 2014
    Because under some of those same conditions he wasn't good enough to start.

    You could play QB for the Pats much less Cutler.
     
  18. Rock Sexton

    Rock Sexton Anti-Homer

    2,553
    1,793
    113
    Mar 14, 2015
    Your statement doesn't make any sense to me though. He was undrafted, signed with the Cowboys for a year then the Panthers plucked him off their practice squad and spent time developing him. He couldn't start until DelHomme retired - albeit they kept giving him starts towards the end of the '07 and '09 seasons ...... he went 6-2 (4-1 in '09 with a 98.5 rating, 8 TD's, 2 INT's, and 7.6 YPA).

    Fox's job was on the line in 2010, so they drafted Clausen for insurance. It was a disaster season for that team fresh with not only Moore's injuries that year, but the constant fiddling/rotating with the starting QB's. Again, the guy had a 2 inch leash on a team that ended up a dumpster fire. He flashed potential, but a regime on the outs bailed on him faster than a bulimic blows their lunch to try and save their collective asses. Cam Newton was in the ensuing draft and it's not hard to connect the dots here. New regime shows up - Cam gets drafted and the team moved on.
     
    Last edited: Aug 18, 2017
    jdang307 likes this.
  19. danmarino

    danmarino Tua is H1M! Club Member

    15,357
    20,976
    113
    Sep 4, 2014
    You're once again using wins and a very small sample.

    He can come in and win 3-4 games in a pinch. In 2009 he had 2 great games, one good game, and 2 below average games. He averaged about 150 yards per game in those 5 starts in 2009. Yes, Carolina went 4-1 in those starts with a top 10 defense and the 3rd best rushing team in the league and had the 29th ranked passing team. He completed about 15 passes per game.

    I find it funny that you tout the 5 games in the 2009 season and make excuse for his 6 games in 2010. In 6 games that season he was 1-5 with a 55 passer rating. Over 11 games total from 2009 and 2010 he had 2 great games (120+ passer rating), 2 good games (90's passer rating), 2 below average games (70-80 passer rating), and 5 horrible games (60 and below passer rating). Exactly like what I've been saying.
     
    resnor likes this.
  20. Rock Sexton

    Rock Sexton Anti-Homer

    2,553
    1,793
    113
    Mar 14, 2015
    Why would I in good conscience judge his starting potential entirely off the 2010 season in which AGAIN he suffered a major concussion in week 2 - then was rushed back. They rotated Clausen in for snaps and then Moore tore his labrum in Week 6 or 7. What sense does that make Mr. "We need to be patient with Tannehill"?

    WHEN HEALTHY - he's shown that given the right setup he can run an NFL offense ..... one that features a dependable running game from which he can feed off and visa versa. That involves both the Carolina years and the Dolphins years. You can try to cut up the 2010 season and merge it into others all you want to try to drag his abilities down, but it's obtuse. It's Monday morning boxscore or Warden Norton in Shawshank Redemption level obtuse.
     
    Last edited: Aug 18, 2017
  21. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    So what If you win 5 games and the defense gave up 20. Is it then? 15? 10?

    I think wins and losses are a qb stat with context. Not a litmus test. Look at record. Did Qb contribute? Did he hold the team back? Etc etc. It's a stat to be used with other data points and analysis. Not to be used by itself.

    A bad record with a good team can be held against the qb. Fiedler/Lucas.

    Fiedler an average qb had that team on its way to the playoffs.
     
    Last edited: Aug 18, 2017
  22. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    ...aaaaaaaaaaaand literally no one uses it that way.
     
    resnor likes this.
  23. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    That's different than saying it's not a qb stat. By saying that means I can't even use it in my analysis, at all. It is a qb stat but it needs to be used right. Not a litmus test. There are great QBs that have losing seasons like Brees. But over 15 years generally the great QBs will have winning records. That makes sense because if your team doesn't win you don't get to play 15 seasons. You're Jay Cutler where you're almost good enough but aren't good enough to carry mediocre teams.

    Cutler on those early Dolphin teams instead of Fiedler wins.
     
  24. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    That said, we say Tanny has had bad coaching. Has Cutlers coaching been all that great? Let's see what magic Gase works with Cutler. He's got all the tools. It's just what's in his head that needs help, and Gase is pretty good at putting his QBs in situations where they won't hurt themselves.
     
  25. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    No, the extremely long list of qualifications needed to make w/l a qb stat is different than saying it is a qb stat.
     
  26. TheFinWR

    TheFinWR WR22

    12
    4
    3
    Aug 17, 2017
    In years without Dan Marino due to injury, we had a bad season. Yet when Brady missed, the pats still did absolutely fine. Wins don't really define a QBs quality. By this "wins mean the qb is good" logic, Brady is somehow just average.
     
  27. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    That's silly. Completion percentage is a QB stat. Do you base your opinion on a QB solely on that? TDs are a QB stat. Just that? Just INTs? Just YPA?

    Every single stat has a list of qualifications that affect it. YPA is affected by play call, scheme, oline etc. But overtime, the good ones usually have a decent YPA. Etc. etc.
     
  28. TheFinWR

    TheFinWR WR22

    12
    4
    3
    Aug 17, 2017
    Also have to see the receivers, so in special cases stats won't be enough.
    You can't give a really good qb some 4th string receivers and expect his stats to not take a hit.
     
  29. Finster

    Finster Finsterious Finologist

    3,087
    2,038
    113
    Jul 27, 2013
    No Dan, he still would be Jay Cutler, just like Ben would still be successful if he was drafted by a bad team, like P Rivers, bad team shmad team, dude is still a baller, you can't make talent disappear.
     
  30. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    Of course those stats don't tell the whole story, but they are significantly more close to showing a QBs ability than W/L are. You are acting like they are of equal significance simply because they both have qualifiers even though one set has a few and the other set as a boatload.
     
  31. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    Moral is or should be, that figures lie and liars figure. This isn't baseball....all stats are debatable to an extent.

    Watch your QB (and others) play the game, then base your evaluation on that.

    I do not even look at basic stats without factoring in the situation. I couldn't tell you who has the best QBR etc, but I can list my top 5 Qbs in football with confidence. The eye test is second to none as long as you understand what you are watching.
     
    Bumrush, Mcduffie81 and DHitchens like this.
  32. DHitchens

    DHitchens Active Member

    193
    123
    43
    Aug 12, 2017

    In that sense you could define an "average" QB by how much disagreement there is among the people who watch him closely and use, as you say, the eye test.

    Nobody is disagreeing that Aaron Rodgers is very good, or that Blake Bortles is very bad. But when you have loads of disagreement, you're probably dealing with an average player, someone who isn't very good or very bad.
     
    cuchulainn and Fin-O like this.
  33. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    Seems logical. I can buy that.


    Sent from my iPhone using Tapatalk
     
  34. danmarino

    danmarino Tua is H1M! Club Member

    15,357
    20,976
    113
    Sep 4, 2014
    No. Very few QB's have been able to carry bad teams and Ben and Cutler are not one of them.

    I like Rivers too and he's an All Pro. He can't necessarily carry a bad team, but he can play well on bad teams. Ben and Cutler... not so much. Ben is a very good QB, but like the majority of the so called "greats" he's been fortunate enough to play on great teams.

    QB's like Aikmen, Montana, Bradshaw, and yes, even Brady, all were fortunate to play on all time great teams for all time great coaches. Yes, they are all HoF caliber, but IMO remove any of them from their respective teams and they don't have half the success they had. Unlike guys like Manning, Marino, Rodgers, and a few others, who could play for the Browns and be in the playoffs nearly every year.

    If you put Cutler on the Pats, in the same situation that Brady had, my bet is he has just as many Super Bowls.
     
  35. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    In practice when you have people assign ratings to different items (in our case QB's) on questionnaires, you very rarely get items with no disagreement (no variance) at the tail ends. On a scale from 1-10 we already know most people in the NFL would put Brady at 10 but a good number here put him lower (some much lower). You'll see the same thing with Rodgers because I know lots of people look at Rodgers and say his stats are fantastic but he only has 1 SB win. And in practice, for most questionnaires you'll see similar variance for items across the entire rating scale (with a few exceptions at the tail ends), so item variance won't in general automatically tell you how "good" or "bad" a QB is.

    However, variance in ratings IS informative, but in a very different way than you're suggesting. The problem with rating scales is that for most people the difference between ratings "1" and "2" generally doesn't mean the same thing as the difference between "6" and "7", or between any other two ratings.

    There's a mathematics called Rasch analysis that fixes this problem by using observed variances in ratings (interesting Rasch analysis comes up in multiple discussions btw.. just had one on another topic a few days ago here). Rasch analysis estimates the magnitude of a latent trait for different people or for different items (in this case the ability of a QB) on an invariant scale so that the numbers are directly comparable. For example, a QB estimated to have a value of 8.6 means that QB is twice as good as one with an estimated value of 4.3, etc..

    So the proper way to use the "eye test" to estimate QB ability is to have people assign a rating (say from 1-10) to each QB and then apply Rasch analysis to it.
     
    danmarino likes this.
  36. jdang307

    jdang307 Season Ticket Holder Club Member

    39,159
    21,798
    113
    Nov 29, 2007
    San Diego
    That's why my evaluation will be based on a lot of factors. And I don't take Winn loss record until they're well into their careers. Take Russell Wilson. Great defense, decent run game. Crap o line. Not the best receivers. But is Wilson contributing to wins or along for the ride? That's how I look at it. It's not just "W/L" record.

    Romo's team we're going 8-8 but they had a league worse defense and no decent run game. Demarco Murray was good but he was hurt and they didn't commit to it. So I didn't knock him down.

    For Tanny I didn't use the record against him. I felt he's played at a level where he's good enough to start but I want to see more improvement. More development.
     
  37. rafael

    rafael Well-Known Member

    27,364
    31,261
    113
    Apr 6, 2008
    IMO people are way too skewed by the teams those QBs play on. You could have very good or possibly even a great QB on crappy teams and you'll still have loads of disagreements. Likewise you could put an average QB on a great team and have loads of disagreements. Basically the vast majority of people are blind to how much QB play is affected by the coaching and surrounding cast. Even when they say they don't look at the record, they'll start arguing about how a QB didn't have success with multiple teams or coaches. That's irrelevant if the multiple teams or coaches were all bad. But most people can't evaluate the QB individually so they'll default to some career look and assume the QB must have been the problem.
     
    danmarino and resnor like this.
  38. DHitchens

    DHitchens Active Member

    193
    123
    43
    Aug 12, 2017

    I agree with all that, but what I meant in my post was that Packers fans, let's say, don't disagree as much about the skill level of Aaron Rodgers as do Dolphins fans about the skill level of Ryan Tannehill. So when you use the population that's likely engaging in the most stringent and thorough "eye test" possible (the fans of the team the player is on), and you still have a high level of disagreement about the player, it's likely (but not completely certain) you're dealing with an average player.

    Rather than something like Rasch analysis, this is more akin to crowdsourcing, though in this case the "crowd" is at least somewhat expert, in that it's comprised of the population engaging in the most stringent and thorough eye test of the player, since they are fans of the team he is on.
     
  39. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Yeah, I understand your argument/intuition, but this is a testable claim and I'm highly certain that if we did the study the results would not be what you're suggesting. First, let's be clear: the only way to really test this (so that we can observe a variance) is for people to assign ratings to items (QB's) which puts us in the world I was talking about.

    OK, when you have a large enough sample size (people rating items), which for the kinds of questionnaires we're talking about tends to run in the 1000+ people, you generally see:

    1) Item variance (variance in ratings assigned to QB's) tends to be relatively similar for most items irrespective of where on the scale they lie. That is, correlation of variance to "distance from median rating" is near zero.

    2) With enough items (e.g. over 100 let's say) you usually do see a few cases where there is nearly no variance.

    So let's be clear: I'm not saying there aren't going to be some QB's where the variance is very small and people agree almost completely. I'm saying you find those QB's everywhere on the scale, and most QB's tend to have similar variance. That's what you see in questionnaires in medical research (I have some experience with those) and when you look at the results of Rasch analysis in other fields (e.g. education) the results suggest the same thing is being observed.

    So the question is why your intuition is probably wrong here. It's because people are quirky!!

    Specifically, you'll see some people use only ratings 4-10 (so the worst QB is rated a 4), others only use 1-7 while others might use 3-8 or maybe 1-10. You'll also see people only use the even numbers or only use the odd numbers. And something very common: people use only a few specific ratings such as maybe only {2,4,5,9} or {1,5,10} etc..

    In fact, netflix had exactly this same problem, which is why they reduced their 5 point rating scale to just 2 because there was a large enough subgroup that used two ratings. Personally on netflix, I've always used 2-4 with an extremely rare case of 1 or 5. Point is, people differ so much in how they rate items that you get a lot of variance everywhere.

    Oh, and keep in mind I'm not making a specific claim for Rodgers vs. Tannehill here (maybe you're right on that one, maybe not.. we'd have to find out), just that in general you'll see enough cases of "good" or "bad" QB's where there is a lot of disagreement. In fact, Tebow is a good case as that recent thread on him shows. I'd rate him a 2 out of 10, but I know there are many that would rate him a LOT higher.
     
    Last edited: Aug 19, 2017
    DHitchens likes this.
  40. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    Cutler!?!?

    I disagree, he is too erratic to be able to play within a system like that.


    Sent from my iPhone using Tapatalk
     
    Finster and jdang307 like this.

Share This Page