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

Ryan Tannehill

Discussion in 'Other NFL' started by bbqpitlover, Oct 16, 2019.

Ryan Tannehill is...

  1. A terrible QB

    0 vote(s)
    0.0%
  2. A below average QB

    4 vote(s)
    5.7%
  3. An average QB

    7 vote(s)
    10.0%
  4. An above average QB

    39 vote(s)
    55.7%
  5. An elite QB

    16 vote(s)
    22.9%
  6. The GOAT.

    4 vote(s)
    5.7%
  1. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020
    Another thing to remember with Rodgers... He sat on the bench 3 seasons after being drafted and was mentored by a hall of fame QB He had a Hall of fame coach one of the best offensive minds in the game to teach him. When he stepped on the football field as the starter he was surrrounded by elite talent

    Point is that because of this we did not see the growing pains from Rodgers and he does not have 3 years of lesser numbers

    Let's examine something.

    One QB was born in December and the other in July (Tannehill) So an age comparison is not exact but very close. Here we go.

    Rodgers in 2008 turned 26 during the season.

    28 touchdowns. 13 interceptions 252 yards per game QB rating 93.8 Completion % 63.6

    Tannehill in 2014 He turned 26 in training camp.

    27 touchdowns 12 interceptions 252 yards per game. QB rating of 92.8 Completion % 66.4

    This is not me saying he is a better QB than Aaron Rodgers. This is me saying look at other great QB's at the same age as Tannehill that had better coaching, better mentors, better players around them to throw the football to.

    Tannehill absolutely compares and his numbers hold up... They are legit. Looking back to the 2014 season it is a shame.. We had Damien Williams as a rookie and they did not use him... Another talented player that had to get the hell out of Miami to shine.
     
  2. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Yeah the problem you're going to have there is that Rodgers has distinguished himself from the average QB historically to such a degree that the best explanation for the discrepancy in his performance is his ability, since surroundings don't vary to that degree for that long. You'd be better off picking someone less distinctive for the comparison if your aim is to illustrate how QBs' environments are important. Like if you want to say Tannehill could've been no different from let's say Philip Rivers, that would be more digestible.
     
  3. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Actually, this is where stats do matter most. Yards aren't meaningless (it does change time of possession and also average starting position for the offense), but that pales compared to the effect of points allowed. If points allowed is not that bad the offense really has no excuse.

    You misread. I agreed with you on the OL, just not on the rest of the team. No way the rest of the team was that bad and averages ~8 wins per season across many seasons. Only the OL was as bad as you make it out to be.

    Have to adjust for era. Otherwise Marino's not even in the top 20 all time, which I'm sure you'd have some trouble accepting. And once you adjust there's absolutely no comparison whatsoever. Not adjusting for era is like comparing median household income today to 50 years ago without taking into account inflation.
     
  4. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020

    The main point was to show that at the same age without an ideal situation around him Tannehill matched Rodgers. I mention this because you have used the excuse in this thread that Tannehill needed a perfect or ideal situation to be successful as he was last year in Tennessee.

    I was just burying you with your own shovel... Comparing apples to apples and proving you were wrong. They were the same age... Rodgers had the better situation and Tannehill matched his production.

    You mention Rodgers distinguishing himself after his 2008 season. Yeah because he continued to be in an ideal situation where he had one of the best head coaches in football.. One of the best offensive minds in the game... surrounded by talented receivers. Things Tannehill has never had all around him at the same time. You continue to prove the point that a handful of us keep making to you clearly over and over again.

    Tannehill gets the hell out of Miami and the first time he has an above-average head coach... an average offensive line and a running game that can be counted on he Suddenly distinguishes himself as something more than average. ( as you mention Rodgers did. )

    So by your very definition, Tannehill just proved he has the ability. Leading the league in half a dozen categories and posting the 4th highest QB rating all time in the NFL... That was anything but average.

    Thank you for playing.
     
    resnor likes this.
  5. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    No he doesn't hold up. I did this analysis many years ago. Without re-doing it to incorporate everything up to 2019, Tannehill fell right around average both as a function of age and years starting (posting just the graph for age). All ratings adjusted to 2016. That won't affect things too much with Rodgers and Tannehill except in their early years where you can just add about 5 to their ratings.

    Rodgers is WAY above average age-wise while Tannehill is just around average.

    [​IMG]
     
    The Guy likes this.
  6. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    I don't mind adjusting for eras...like pre-2007 and post-2007.

    But I don't believe that you need to adjust all QBs each year post-2007 to compare to each other.
     
  7. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    You don't want to have subjectivity in the process. You don't want two people arguing over when to adjust and when not to adjust. Besides, reporting the mean and standard deviation (z-scores) is just a very common description of the data that naturally allows you to compare across eras. So you simply do it and compare. And for years close to each other, "adjusting" usually does almost nothing (e.g., going from 2016 to 2019 would add only 1.1 passer rating points). Why complain when there's no effect worth mentioning?

    You're wrong about 2007 though. That's a 7.8 passer rating point difference to 2019, which can make a huge difference in interpretation.

    Anyway, it's the proper thing to do from a statistical analysis point of view.
     
  8. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Yeah you just don't seem to get it. The following career average QBs had 11-game stretches that were indistinguishable from Tannehill's in 2019:

    Nick Foles
    Chris Chandler
    Jim Harbaugh
    Wade Wilson
    Mark Rypien
    Drew Bledsoe
    Vinny Testaverde
    Steve DeBerg
    Tommy Kramer

    So unless you're prepared to argue that those QBs have anything other than average ability, you shouldn't be doing so for Tannehill either, at least on the basis of his 11 games in 2019.

    What Rodgers has that Tannehill doesn't is a far more lengthy history of elevated performance, that no career average QBs in history have. So Rodgers is very reliably distinguishable from Tannehill and other career average QBs, whereas Tannehill can't be distinguished from the average QBs noted above.

    Again, the only folks here who are "stomping around like three year olds" as you put it earlier in the thread are the ones like yourself, who continue to attempt to extract greater meaning from Tannehill's 11 games than is warranted. There is some acceptance of the reality of the information at hand that you should be yielding to at this point, but aren't.
     
  9. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020
    CBrad

    I am all for adjustments for Eras... But The season I quoted from Tannehill was before 2016...

    As resnor stated if this was a pre 2007 and post 2007 adjustment both seasons I mentioned for Rodgers and Tannehill at the same age fall into the Post 2007 argument.

    The numbers are identical for those same age seasons.. So even if there is a slight adjustment for era... The fact that Rodgers had superior coaching, mentorship and talent around him... Still puts them about even. My goal was not to say that Tannehill is better than or equal to Rodgers. I was merely showing that when Tannehill had half decent receivers as he did in 2014.. ( Landry, Hartline, Clay ) an average running game (Lamar Miller ) Even with ****ty coaching and a bad O line... He started producing at a young age.

    Without knowing what the adjustment factor is for the environment... I wonder if the superior coaching, mentorship, surrounding talent balance out any adjustments you would make for Era? That might be a difficult calculation to make.

    Obviously we cannot go multiple seasons to compare Tannehill and Rodgers because Tannehill changed coaches soon after and then got hurt...

    Guy's comments about needing ideal situations to succeed prompted this side by side apples to apples comparison of Tannehill compared to arguably the best QB of his generation at the same age. If you want to give the edge to Rodgers I can support that... But the fact of the matter is that the numbers compare...

    Nobody here is arguing that Tannehill had seasons where he did not look special in Miami... There are many factors for this... Really, really bad football teams... Adam Gase ( Arguably one of the worst coaches in the league ) and injury... Tannehill did not take a step forward after two strong seasons in 2014 and 2015. None of that is up for dispute.

    What creates a great debate is that when he leaves Miami and for the first time in his career has a half-decent coach... and an average O line with a dependable running game... His production takes that step forward and we see an improvement on the promising numbers from 2014 and 2015. Suddenly Tannehill is leading the league in many offensive categories. Playing at a level that would make him a legit MVP candidate.

    Guy would argue this a fluke or luck... But we saw promising numbers from a Young Tannehill on a bad football team in a less than ideal situation multiple times in the past... Suggesting that the ability was there. ( As he has argued we have not seen evidence that the success of 2019 was based on Tannehill's ability )
     
    resnor likes this.
  10. Patster1969

    Patster1969 Season Ticket Holder Club Member

    679
    788
    93
    Nov 8, 2017
    Did I miss something - I was sure that RT had left Miami but here we are at 217 pages :smile:
    All I know is that he seems like a decent guy, who needed a fresh start and who had a great stretch of play after relieving Mariota (who had stunk up the place). We all know that in Miami, he didn't have any consistency on the Oline or coaching (either systemically or personnel) but we also know that he isn't the QB if you want off the cuff moments of brilliance to bring your team back from the brink.
    As a "team" (including a good to great back & good Oline), they rode that great stretch but lost to KC in the AFC Championship and in that time, he statistically was very good (& the Titans are hoping that he can extend that play). Will he carry on that play, no idea but as long as he has support from his team, he has shown that he can be better than average.
    We can all agree to disagree chaps - lets just be excited for the Fins for this season!!
     
    Bumrush and resnor like this.
  11. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    None of the above distinguishes Tannehill from a sufficient percentage of career average QBs. Not what he did in Miami during his pockets of elevated performance, and not what he did in Tennessee in 2019. If you want to reduce the number of consecutive games from 11 to those in which Tannehill's performance was elevated from time to time in Miami, you'll find scads of average QBs throughout history who've done the same thing.

    You just don't seem to understand that Tannehill has still done nothing that reliably distinguishes him from an average QB. Might that be different in the future? Who knows, but as of now this is the case.
     
  12. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    The only valid reason I see for adjusting, is to account for rules changes that have a significant effect. For instance, 2007.

    Like I've argued, record books don't adjust. If you want to look at averages, per year, and rank QBs each season off that, that is fine. But overall volume stats I don't agree with being adjusted.
     
  13. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Tannehill lead the league in Peru much every important statistical category, but you claim that it doesn't distinguish him.

    And you can't see why that position is contentious.

    How many average QBs lead the league in all those categories?
     
    The_Dark_Knight likes this.
  14. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    What categories did those QBs lead the league in?
     
  15. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    What meaning does leading the league have if the QB leading the league is indistinguishable statistically from the QB who's let's say second in the league? If hypothetically the QB leading the league in passer rating has a season passer rating of let's say 120, and the QB who's second in the league has a season passer rating of 119.9, what meaning does leading the league have?

    The point is, Tannehill is indistinguishable statistically from the QBs listed above, regardless of who led the league. But if you think leading the league has some special meaning, let's hear what it is.
     
  16. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    YOU are claiming that what Tannehill did last year is no different than many other "average QBs." So, I'm asking, of the "average QBs" you listed, what statistical categories did they lead the league in those seasons? Surely if what Tannehill did is so common, then those other QBs lead the league in similar categories.
     
  17. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    How can you tell whether a rule change has a "significant effect"? The only way is to look at the stats. So you START with the stats, not the other way around because there is no independent way of determining "significant effect".

    Also, rule changes happen constantly. 2009 for example had two that intuitively should affect passing stats: 1) it became an illegal hit on a defenseless receiver if the defender uses his helmet, forearm or shoulder to the receiver's head or neck area, and 2) a defender cannot initiate a roll and lunge to forcibly hit the passer below the knee. Yet you dismiss that. And that's just one example (one year).

    When you look at the stats the effect of rule changes and interpretations are all very gradual, not sudden, except for 1978. In other words, it's actually warranted to adjust by year because the observed effect is constant and gradual, not sudden.

    Before 2016.. so from 2012-2015 Tannehill was 24-27 years old. Rodgers didn't play at 24 but played from 25-27 and put up WAY better numbers than Tannehill. No the numbers are not anywhere close to similar.
    https://www.pro-football-reference.com/players/T/TannRy00.htm
    https://www.pro-football-reference.com/players/R/RodgAa00.htm

    It's an impossible calculation at this time. So there's room for interpretation obviously, but statistically Tannehill was average as a function of both age and years starting while Rodgers was way above average.
     
  18. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Stop. The rules changes in 2007 was significant. It changed IMMENSELY how defenders could play defense on receivers. Having ratings creep up over seasons could be attributable to many things, but as I recall the change in rating pre-07 to post-07 was not creeping.
     
  19. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    No you stop. The rule change in 2007 led to only a 2.2 increase in passer rating. There were many years with much bigger changes. You've got some idea of reality in your head that doesn't fit with the data. The data show gradual increase since 1978 (which actually took about 2 years to fully manifest itself).
    https://www.pro-football-reference.com/years/NFL/passing.htm
     
  20. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    So if you would've said Tannehill is distinguishable from the other QBs listed above because he combs his hair more neatly than they do, would it also be incumbent on me to investigate that and determine whether that's true, or would it be incumbent on you to show why combing one's hair neatly has any relevance to the topic?

    The point is, you haven't established why "leading the league" in anything has any meaning. Why should we assume that's relevant to the topic?

    Additionally, it's highly likely that the QBs listed above led the league in passer rating during the 11-game stretches in which they were statistically indistinguishable from Tannehill in terms of passer rating. So even if we were to ascribe some special hypothetical meaning to "leading the league," Tannehill still very likely wouldn't be distinguishable from them on the basis of his 11 games in 2019. They likely all led the league in passer rating during their 11-game stretches.

    The problem here you all are having, again, is that you're attempting to extract a great deal of meaning from 11 games, when that just isn't possible. You should just wave the white flag on that one and start talking about 2020.
     
  21. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,320
    113
    Nov 24, 2007
    Rockledge, FL
    Ok, here's where your "adjusting for era" makes ZERO sense to me. I'm going to just provide a small example to illustrate my point.

    The 1984 Miami Dolphins regular season;
    572 passing plays
    484 rushing plays

    The 2019 Kansas City Chiefs regular season;
    576 passing plays
    375 rushing plays

    I don't care if we're talking about 2020 or 1920...
    -A first down is still 10 yards
    -Each team has four down to achieve that first down
    -They can run the ball
    -They can throw the ball

    Now back to the adjusting for era, did the 1984 Dolphins run the ball more than the 2019 Chiefs? Sure, the number of actual plays clearly illustrate that, but did the 2019 Kansas City Chiefs throw the ball SIGNIFICANTLY more than the 1984 Dolphins? Sure...a whopping 4 more times.

    I still don't understand WHAT your point is with the "adjusting for era". A 15 yard first down pass by Dan Marino and a 15 yard first down pass by Patrick Mahomes is the exact same thing, period.
     
    resnor likes this.
  22. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    As you know the nature of the game has changed. Often, it's due to rule changes or re-interpretations of rules already on the books, and when you look at the number of rule changes and re-interpretations over time it's huge. The question is what the effect of all those are. Well, that depends on the stat. In terms of points scored or points allowed, you get very little change over time, and while one can technically "adjust", I would have no issue with someone not doing so because the difference is quite small.

    But that's not true for many other stats, specifically many passing stats. Go through this link:
    https://www.pro-football-reference.com/years/NFL/passing.htm

    Look at the league average passer rating in the 1980's. It was in the mid-70's. That's a really bad rating today. Look at league average passer rating in the last 5 years. It's around 90. We're talking a 15 point difference on average here! That is, an "average" QB today can be expected to have a passer rating about 15 points higher than an average QB in the 1980's. You can't ignore that difference.

    Similar things are seen for completion percentage. However, Y/A hasn't changed too much, etc.. Point is, the stats show you the effect of any changes in the game, and "adjusting" for that just means you have a way of putting everyone on the same scale (which by now you know is called a z-score).

    Point is, don't think every stat needs to be "adjusted", but it has to obvious to you that putting up huge passing stats in today's pass friendly environment is a completely different story than when Marino was breaking records (it's much easier today). Hope that helps a bit.
     
    Last edited: Jun 8, 2020
    Irishman likes this.
  23. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020
     
  24. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Then I take back what I've said. I don't support ever adjusting for eras.
     
  25. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    That can be a result of different things, though, cbrad. For instance, it can be a result of QBs pushing the ball further and getting away from "3 yards and a cloud of dust" mentality. More lower percentage of completion passes because of the bigger payout, vs 40 years ago being more concerned with safe and reliable passing.
     
  26. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    OK then Marino in Miami = Tannehill in Miami. Personally I think that's ludicrous. Marino was one of the best. Tannehill was average. But now you'll have to argue they're similar (and I mean Tannehill in Miami excluding Tennessee).
     
  27. resnor

    resnor Derp Sherpa

    16,327
    9,874
    113
    Nov 25, 2007
    New Hampshire
    You've argued, I think it was you, I might be wrong, that Marino in his very early years, especially 84, was unreal, but was not nearly like that the rest of his career.
     
  28. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020
    CBrad

    I think one difference is that you are attempting to take one apples to apples comparison and change it into another.

    I tried to look at the seasons when both QB's turned 26... and showed that the statistical numbers are identical. Now if you want to give the advantage to Rodgers slightly because he did it in 2008 and Tannehill did it in 2014 and you are arguing that the league shifted towards passing and therefore needs to be adjusted slightly... I have no issue with that.

    There are benefits to different types of studies and comparisions... But there are also more variables and complications. One reason that OPS is still used as a golden standard for offensive production in baseball it that it is a simple thing to calculate without a ton of variables... There are better statistics to measure certain things. But overall considering the simplicity and accuracy of OPS it is hard to beat as a statistic for overall offensive performance in a baseball hitter.

    Applying that logic to this conversation... I looked at their seasons when they turned 26 to get an apples to apples comparison that was simple to do without a ton of variables. The longer the comparison goes the more variables that get introduced like coaching changes, injuries... Rebuilds by the front office... Etc.

    Nobody is denying that from the age of 27 through 30 Rodgers did not put up better numbers than Tannehill.

    I merely looked at the 26 year old seasons from both because Guy has pointed out that the success of Tannehill in 2019 was not enough to go on in regards to saying he was anything more than undistinguishable or average. I wanted to show that Tannehill had in fact in the past showed indications that he was not merely average and undistinguishable. His 2014 season at the age of 26 proves that.

    I remember when Rodgers took over in 2008 and everyone was raving over him at the age of 26.... Like he was the next sure-fire hall of famer to come along. Even with some adjustment... Tannehill puts up the same numbers at the same age and he is average?

    Yes there is a slight adjustment for era... But as you said there is consideration to me taken account for ... in terms of surrounding talent, head coach, the type of offense...etc. ( we both agree these are hard to calculate.. but they are there and they lessen the impact of the adjustment for era.. .considering that the seasons were only a few years apart... Not decades like Tannehill and Marino. )

    Come on.
     
  29. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    That's correct, but he was still averaging around a z-score of 1 for much of the 1990's, excepting his last few years. That's around the average for a SB winner. Tannehill never approached that. z-scores are MUCH better resnor.
     
    Irishman likes this.
  30. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Dude.. at the age of 26, even without adjusting, Rodgers put up a 103.2 rating while Tannehill put up a 92.8 rating. In what world are those similar?? And at the age of 25, it was 81.7 for Tannehill to 93.8 for Rodgers. In both cases no similarity.
     
  31. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020

    Dude

    Because I explained this if you took the time to read... I looked at the seasons in which Rodgers and Tannehill turned 26. Not their age before the season. I did this because they were not born on the same month. The season in which Rodgers turned 26 he had a QB rating of 93.8 The season in Which Tannehill turned 26 he had a QB rating of 92.8.

    And if you look at all of the other numbers they are identical Yards per game Touchdowns Interceptions completion %
     
  32. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Tannehill was born in July, Rodgers in December. You want to use the age at the beginning of the season in that case. The comparison I made is (with the exception of a few games) the correct one.
     
  33. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020

    No... I do not want to use the age before the season. Do not put words in my mouth thank you very much. If someone was born in August and someone in September you would be looking at two different seasons. There was a reason I said the season in which they turned 26...

    Either choose to discuss that with me... Or we have nothing to converse about. I appreciate your statistical analysis and the time you take to map out and make charts. But please do not tell me what I want to do or think.

    One player turning 26 in preseason/training camp and the other during the regular season... Is a solid comparison. It is not perfect... In an ideal world they would both be born on the same month... But we do not always get what we want when trying to make comparisons.
     
    resnor likes this.
  34. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    I don't care what you "want" to use. The question is what the proper analysis is. You do faulty analysis we can dismiss it as such. Why don't you do the correct thing and calculate passer rating for all games by age for those 2 QB's. Do that work and I guarantee you it shows the QB's are not similar. That should be obvious based on the stats already presented.

    In any case, if this becomes an important enough issue (which it isn't based on the stats presented — the result is obvious), I can do an analysis on "days from birth" lol. The two QB's will NOT be similar.
     
  35. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,320
    113
    Nov 24, 2007
    Rockledge, FL
    Ok, I get your point. I do hear you, but with all of that being the case, you're adjustments are backwards!

    Rule changes have BENEFITED, not hindered the younger generations of quarterbacks and receivers and as a result have produced higher numbers than their predecessors. So the extrapolation that has to be made and adjusted for is what if today's quarterbacks had to play with the rules established in the 70's and 80's...or what if yesteryear's quarterbacks were playing with today's more protective rules.

    Those are merely hypotheticals which can be discussed and argued. "What if" is always an argument of perceptions. I would argue that with today's rules that are so protective of quarterbacks and receivers being applied to the 80's, the NFL still wouldn't know where the NFL Quarterback Record Book is...because Marino would have it in his own private library. That's my perception based on the actual games I watched.

    Anyhow, the "adjustment" is merely speculative.
     
    resnor likes this.
  36. Etrius24

    Etrius24 Well-Known Member

    682
    685
    93
    Mar 4, 2020

    Brad

    Dude, I did a simple age apples to apples... The season in which they turned 26

    Deal with it or not. Not a ton of variables... Just looking at young qb's the season they turned 26 years old.

    You want to turn it into something else then that shows your bias. I have clearly set the parameters for this discussion... You get to choose whether or not you want to participate actively within those parameters.

    Cheers.
     
  37. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    There is no "direction" in z-scores. Each QB is being compared to all other QB's in that particular year on a z-score axis. When you see me assign a "2019-adjusted rating" to some QB from the 1980's that's just transforming a z-score to a rating in any given year. We could use a "1984-adjusted rating" too if you wanted. I only adjust it forwards because people are watching games today and are more comfortable with 90 being average rating.

    z-score however are timeless. They're neither adjusted forwards or backwards. They just "are".

    Oh, and to be absolutely clear, the adjustment is technically not telling you how a QB from the 1980's would have played today or vice versa. A z-score is only a measure of how "impressive" that QB was relative to his peers in that year. No one knows how well a QB from one era would actually adjust to another. But we can measure how "impressive" that QB was in his own era = z-scores.
     
    Pauly likes this.
  38. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    You're including games from when they weren't the age that you categorize the games as (technically I was too, but only a few games while in your case it's most of the season!). My bias is doing proper statistical analysis, and you're not even attempting to do that. Cheers.
     
  39. The_Dark_Knight

    The_Dark_Knight Defender of the Truth

    11,817
    10,320
    113
    Nov 24, 2007
    Rockledge, FL
    No, it doesn't work Brad. Scores are being applied to quarterbacks based on what they've actually produced...based on rules that govern the game during their era.

    Marino in his prime in 2020 would have different scores based on actual production compared to Marino in his prime in 1984, so to say that a score given to Marino then and Brees now are accurate representations of their individual abilities is not accurate whatsoever. The actual hard numbers governing such scores are completely different.
     
    resnor likes this.
  40. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    It's a difficult concept to understand, I know. But z-scores aren't answering the question you're asking. Everything you said is correct, but what I said in the post you quoted is correct too.

    Maybe let me try a different approach using an analogy: suppose you measure one set of objects in centimeters and another set of objects in inches. The objects aren't the same (that's what you're referring to), but the bigger problem is that the units of measurement aren't the same! We know the conversion from centimeters to inches (2.54 cm = 1 inch), but you don't know that a priori for passer rating from different eras. What z-scores are doing is equating the units of measurement, NOT the objects being measured.

    So yes, Marino today would have had (most likely) a different z-score than back in his day. That's what I meant by we have no way of knowing how a QB from one era would perform in another era. However.. the unit of measurement represented by the z-score is the same in both eras. That is NOT true with passer rating. The passer rating points themselves mean something different in different eras while a z-score means the same thing regardless of era.

    Does that help? We're equating units of measurement, which is necessary for comparison.
     

Share This Page