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

Great Older Article on Fitzpatrick

Discussion in 'Miami Dolphins Forum' started by KeyFin, Jun 6, 2019.

  1. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    What if we did a partial correlation with YPA, win percentage, and INT rate? If we found that controlling for INT rate increased the correlation between YPA and win percentage, would that address one or both of the above points?
     
    cbrad likes this.
  2. KeyFin

    KeyFin Well-Known Member

    10,488
    12,821
    113
    Nov 1, 2009
    Here's where I dislike stats over human observation since the answer is so blatantly clear. In those first 5 games, the offensive line was so bad that we couldn't complete a pass....Tannehill was running for his life and basically playing school yard football. Of course his stats were off in that stretch...any QB's would be.

    The only team we beat in that stretch was Cleveland and they literally gave us the game on turnovers- we did almost nothing to win it on our own. In that Cincinnati game, (game 4), Tannehill was pressured on almost 100% of his drop backs in the first half- I remember distinctly because I drove almost an hour to find a sports bar on a business trip, only to walk out disgusted at halftime. It felt like we had a 0% chance of completing a pass...much less winning a game. Two linemen were fired a week later and that's when everything turned around for the 6 game win streak.

    In short, that stretch was the worst in RT's career because it's when he was pressured the most. And as Brad said, I don't think a 5 game stretch is statistically relevant enough to draw any conclusions from...other than to say Tannehill plays especially bad under pressure (which we have his whole career to corroborate). Over the next 12 or so games (spanning 3 seasons for RT), his numbers were right there with an elite, top 5 QB because that's when protection was at it's best. Again, maybe that's not enough data either for a long-term prediction of performance, but it was enough for me personally to say if you protect the guy, you can win with him at a fairly high percentage.

    It's no coincidence when we got to 2nd and 3rd string linemen in 2018,his performance plummeted once again. That's just who the guy is- you pressure him and he makes critical, game-changing mistakes.
     
    Last edited: Jun 9, 2019
  3. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    I'll give you credit for the suggestion – you're trying to shift the burden of association between Y/A and "risk" to INT% which is a much easier argument to make – but it won't work for two reasons: 1) INT% and Y/A for Tannehill are too weakly correlated at -0.1, and the confidence interval on the correlation between Y/A and win% is too large (the 95% CI goes from 0.1567 to 0.525), meaning that you'd need an absurdly strong effect by controlling for INT% (which by reason #1 is unlikely to be strong) to make the argument.

    Either way, this is the result: the partial correlation between Y/A and win% when controlling for INT% is 0.3418 so actually a tiny bit SMALLER than when not controlling for it (0.3545). But again.. that 95% CI is huge.
     
    Surfs Up 99 and The Guy like this.
  4. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    What's confusing about that explanation for his performance during those five games, however, is that he exhibited what looks to be (by just glancing through his career game logs) the highest YPA over any five-game stretch of his career. Like cbrad said, that's a very small sample, so that may be a random finding, but it's certainly counterintuitive. It's hard to make the argument that extremely poor offensive line play increases YPA well beyond what a player typically exhibits.

    I think an equally plausible explanation is that Adam Gase, in his first year as a head coach, attempted to have Tannehill play the more aggressive style he prefers, and that style contributed to both 1) a significantly increased YPA, and 2) a greater interception rate. Gase, having started the season 1-4, then had to rein Tannehill in.

    It also makes little sense that Tannehill's YPA decreased significantly after the supposed "correction" of the offensive line. And that was over eight games.
     
    Last edited: Jun 9, 2019
  5. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Yeah, there's no question Tannehill plays bad under pressure, and we don't have good "pressure" stats (though Football Outsiders does report them and you can see with their pressure stats that Tannehill plays well without pressure but bad with it), but answering the original question of "risk" isn't the same as talking about pressure even if pressure is one variable that matters. But yes, in general I agree it's difficult to answer these types of questions using stats alone, though it's essential any narrative be consistent with available stats (e.g., those deep ball and 3rd and 10+ stats were a good example).

    Anyway, good discussion.
     
    The Guy, KeyFin and Surfs Up 99 like this.
  6. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Yeah.. I think we should just go with what statistical tests say about this instead of trying to over-interpret it. We can use a two sample t-test with unequal sample sizes to see if two samples come from populations with the same mean.

    Testing Y/A for first 5 games in 2016 vs. last 8 games in 2016: p-value = 0.5848
    Testing Y/A for first 5 games in 2016 vs. all 13 games in 2016: p-value = 0.6935
    Testing Y/A for first 5 games in 2016 vs. all 88 games in Tannehill's career: p-value = 0.2251

    The p-value needs to be less than 0.05 to make a case it's not random variation alone that could account for this.. so I think it's best if we don't make much of the Y/A for the first 5 games (5 is really small so the above results are no surprise lol).
     
    The Guy likes this.
  7. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    OK well that does it. Thanks.
     
    cbrad likes this.
  8. KeyFin

    KeyFin Well-Known Member

    10,488
    12,821
    113
    Nov 1, 2009
    YPA can lie at times because it doesn't necessarily mean deeper passes...it simply means the ball ended up further down the field (due to QB, receiver or both). That was the case in that five game stretch-

    The Seattle game was a 12-10 loss...to put it in perspective how bad the offense was, we had 327 yards of punting and 150 yards passing. RT completed 10 passes total on the day (we punted 7 times).

    Game 2 was New England where Tannehill went berserk in the 2nd half with bombs and deep passes...I still believe it was his best game ever (even though he played horrible in the 1st half). That's the game where Parker dropped the bomb on the goal line as time ran out to send it to overtime.

    Game 3 was RT for 311 yards in an OT win, but it was an ugly afternoon filled with 3 and outs. RT had 3 TD's but the stats lie here- his 93 QBR came from short passes with big effort. Landry has a massive game with yards after the catch, plus there was one one beautiful bomb RT to Stills.

    Game 4 was Cinci where again, the D carried the day. RT had 160 yards passing.

    Game 5 was the Titans, 149 yards passing and zero TD's thru the air.

    Y/A per game-

    Seattle- 5.2
    NE- 8.6
    Browns- 8.0
    Cinci- 6.4
    Titans- 8.3

    Week 10 Chargers- 9.6
    Week 14 Arizona (Moore)- 9.2
    Week 15 Jets (Moore)- 13.1

    When we break this 5-game span down, you have 3 games sub-160 yards, a 311 yard day that came off short passes, then an INCREDIBLE 2nd half with RT grabbing chunk yards drive after drive as a pure gunslinger. The NE game is an anomaly because that simply wasn't his play style under Gase...he NEVER had that much freedom. Cleveland was also an anomaly because the receivers/RB's did all the work. And in the other 3 1/2 games (counting the 1st half of NE), we could barely get a pass off to anyone but Landry on screens (which sometimes went for great yardage).

    So the numbers you're quoting are due to two things- Landry's effort in a horrible offense and that HUGE 300+ yard second half from RT in New England. His best full game that season was week 13 against San Diego (in Y/A and overall QBR), but the 1st and 3rd best Y/A per game were actually Matt Moore in weeks 14 and 15. So you're really looking at anomaly after anomaly in this short span to give a very unexpected result.

    Mostly though, you're seeing the line starting out horrible, but eventually growing in chemistry. They were peaking around week 13 or 14...which explains that awesome 3-game stretch (under two different QB's) among the win streak.
     
    Last edited: Jun 9, 2019
  9. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,652
    67,546
    113
    Dec 20, 2007
    You would think with such a smart brain that Fitzpatrick would really improve with the more experience he gets.

    Do we see that upward trend in his game, I know last season was his best..

    For me it will be interesting to see if his arm strength is diminishing..he’s not the type that can lose anything off the fast ball
     
    Surfs Up 99 likes this.
  10. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Once you adjust ratings to a common year.. no improvement:
    [​IMG]

    ** 3 games where Fitz threw at least one pass but did not start and did not have a minimum of 10 passing attempts aren't plotted.
     
    The Guy, Surfs Up 99 and djphinfan like this.
  11. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,652
    67,546
    113
    Dec 20, 2007
    It will be interesting to see if the smartest guy in the room throw these Interceptions in stupid fashion, ya know the ones that make you go wtf was he thinking or because he thinks he has more arm strength then he actually does..

    Cb, ?, Overall statistically speaking in what area do you see his game is flawed.?
     
    Surfs Up 99 likes this.
  12. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Easy.. his INT% is way too high. Best he does is middle of the pack but more often it's near or at the bottom of the pack. For example, in 2018 and 2016 he was literally the worst by that measure among starting QB's:
    https://www.pro-football-reference.com/years/2018/passing.htm

    For reference, team average INT% (so treating starting and backup QB's as a single QB for each team) was around 3% when Fitz came in the league and is now down to about 2.5%. However, starting QB's tend to have lower INT% than team INT%, and if you have anything around 4% or so.. you're at the bottom or very close to it among starters. Out of 12 years Fitz played, 5 years he had greater than 4% INT%!!
    https://www.pro-football-reference.com/players/F/FitzRy00.htm

    Oh.. and when looking at those stats remember to keep in mind sample size. In 2017 he had a career low 1.8 INT% but that's with only 3 games starting. So that one and his 5.9% his rookie year (also with 3 games) are probably more due to sample size than anything else.
     
    Hooligan and The Guy like this.
  13. KeyFin

    KeyFin Well-Known Member

    10,488
    12,821
    113
    Nov 1, 2009
    Let me ask a more technical question since it's the one we're all thinking anyway-

    1) Is Fitzpatrick better than (or at least equal to/similar to) Tannehill?

    2) Does Fitzpatrick have more potential than Tannehill?

    Obviously I'm asking from a statistical perspective only, but you can weigh that however you want (YPG, PPG, QBR, whatever).
     
    Last edited: Jun 9, 2019
    Surfs Up 99 and The Guy like this.
  14. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Statistically.. Fitz is slightly worse than Tannehill. Doing this properly by z-scores (standard deviations above/below mean) and weighting by passing attempts, Tannehill's weighted career z-score rating is -0.1166 (45th percentile) while for Fitz it's -0.4662 (32nd percentile). So Tannehill is slightly below average while Fitz is better than only about 1/3 of starters.

    I also have those z-scores adjusted for defense (slope of best-fitting line between z-score points allowed and z-score passer rating is 0.3043 so you can subtract z-score rating points given defense points allowed), and Tannehill's hardly changes at -0.1104 while Fitz improves just a bit to -0.3097. Either way, Tannehill > Fitz.

    As far as potential.. I guess the best I could do is go back to those graphs I had that show that on average (obviously not for everyone) ability plateaus by year 5 so statistically at least you'd expect no better/worse with large sample size for either. So I guess no there too.
     
    Surfs Up 99, KeyFin and The Guy like this.
  15. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,652
    67,546
    113
    Dec 20, 2007
    Jmo You can’t measure these two qbs accurately based on stats..

    I think one has an awareness of situational football and the other was clueless.
     
    Surfs Up 99 likes this.
  16. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    Here’s a link to the p-f-r page that has the splits of offensive stats for 2018
    https://www.pro-football-reference.com/years/2018/splits.htm

    Eyeballing the numbers there is a slight down tick with sack% according to estimated win%, but it’s nowhere near as dramatic as INTs.
    Without crunching the numbers being in second and 10+ and/or third and 4+ (i.e. obvious passing downs) more predictive of sack rate than win%. Obviously bad teams and teams coached by Adam Gase are going to be in second or third and long more than well coached teams,
     
    The Guy likes this.
  17. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    What's interesting about that is the dramatic decrease in passer rating as a function of win probability. In 2018 if your team was 80 to 99% likely to win, your passer rating was 114.7. If on the other hand it was 0 to 19% likely to win, it was a mere 68.9. Fascinating.
     
    Pauly likes this.
  18. Surfs Up 99

    Surfs Up 99 Team Flores & Team Tua

    1,950
    1,785
    113
    May 5, 2016
    If we are going to lose I would rather have a guy who goes down fighting. That's why I always liked Matt Moore. However, as I said before, if Fitzpatrick can let other guys earn their paychecks too ( let his fire to win fuel the others), and save his bag of tricks for when they are really needed he can be a good QB. It's not the Fitzpatrick show. He needs to put the team first and play his role.
     
    resnor and KeyFin like this.
  19. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,652
    67,546
    113
    Dec 20, 2007
    Tannehill did not know how to win football games, he didn’t understand how to manage a game and convert using his whole skillset, he just went with the play call, most of the time first read stuff..
     
    Hooligan and Surfs Up 99 like this.
  20. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    And the Dolphins must believe this as well, hence Josh Rosen.
     
  21. ExplosionsInDaSky

    ExplosionsInDaSky Well-Known Member

    3,163
    2,325
    113
    Sep 13, 2011
    This is how I see it too...Physical talent, arm strength, potential...Obviously Tannehill checked all those boxes before Fitzpatrick did. With that said, I think Ryan Fitzpatrick is the better football player. That's kind of hard to explain, but that is how I feel.
     
    Hooligan, KeyFin and djphinfan like this.
  22. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    It's analogous to the difference between a receiver who has nothing but speed and one who actually plays the entirety of the position. Who would you rather have, Darrius Heyward-Bey or Danny Amendola?

    Tannehill was all arm. Fitzpatrick plays the entirety of the position. Not well, but at a "Danny Amendola" level. When you need a play made you want Amendola, not Heyward-Bey.
     
    djphinfan likes this.
  23. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Yeah.. what the stats are measuring is the effect of the QB (or technically.. QB-led passing offense). That word.. "effect".. is the key. I agree that stats are totally insufficient to measure stuff like football smarts or how well a person uses their entire skillset. But you can watch and evaluate all you want, it won't allow you to predict the effect of that QB on the passing offense as much as looking at the stats.

    And ultimately results are what matter, not the perceived talent.
     
    KeyFin likes this.
  24. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Yet, at the end of the day, neither QB is viewed as anything more than a backup, really, by the league. So, does it matter if Fitz is better? I mean, it doesn't translate into wins.
     
    The Guy likes this.
  25. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    cbrad may be able to correct me on this, but I don’t think you get to anything significant in terms of wins from the quarterback position until you go from the average level up to about the top six QBs in the league.

    There are essentially five or six QBs in the league who give their teams the probability of making the playoffs every season, based on the number of wins they would be expected to account for over and above the average 8-8 team. It takes uncommonly poor play around them, almost always (if not always) by their team’s pass defense, to keep them from making the playoffs.

    This of course why those quarterbacks are so valuable.
     
  26. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    It's more like top 4 in most years if the question is what passer rating does a QB have to have so that IF he were on an otherwise average team that team is expected to have 10 wins. I chose 10 because the average win% of all WC teams in NFL history is 64.92% which translates to 10.4 wins in a 16 game season so for this question let's say 10 wins. Using the best-fitting line between adjusted passer rating and wins that translates to about 15 passer rating points above average that you need, which in 2018 means a 108 rating or top 4. In 2017 it's a 102 rating which is also top 4 and in 2016 it was about 104 which is also top 4.. so top 4.

    Just so it's clear, this is assuming the rest of the team is average. The average z-score for a playoff team, both in points per game as well as points allowed per game is between 0.71 and 0.72 while the average z-score for a playoff team QB measured by passer rating is 0.627. The calculation above of +15 passer rating points corresponds to an average z-score of 1.37. So you can see what's going on here. If you assume the defense z-score and non-passing component of the offense z-score goes from 0.7 or so to 0, then the QB needs to be about double above league average as your average playoff QB to expect to make the playoff with 10 wins.

    Keep in mind this is top 4 in any given year, not across all years.
     
    The Guy likes this.
  27. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Awesome work. So last year that was Brees, Mahomes, Wilson, and Ryan.

    The biggest mistake this franchise made in recent history was drafting Jake Long instead of Matt Ryan.
     
  28. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    If you're willing to go back to 2006 instead of 2008.. easily the biggest mistake this franchise made was selecting Culpepper instead of Brees. Matt Ryan is a QB that's consistently slightly above average with so far only 2 years that were awesome, especially 2016. Brees on the other hand.. he and Peyton are approximately tied now for the most "added wins" by a QB in history. In other words, not just efficiency (Steve Young is #1 in efficiency all time) but efficiency times longevity.

    I'm firmly convinced this team would have at least one and probably multiple SB wins by now had we chosen Brees in 2006.. or had we done what practically every fan wanted in the 2001 draft and drafted him back then instead of.. (gulp) .. Jamar Fletcher!?%@#!
     
  29. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Yeah that’s a tremendous point, because Brees is even better than Ryan. And I think with him you would’ve even had Nick Saban’s choosing to stay and probably seeing to it that the team had one of the league’s best defenses as well.
     
    resnor and cbrad like this.
  30. djphinfan

    djphinfan Season Ticket Holder Club Member

    111,652
    67,546
    113
    Dec 20, 2007
    Just my personal opinion the biggest mistake was not taking Luke Kuechly and Russel Wilson 1 and 2, instead taking Ryan Tannehill.
     
  31. ExplosionsInDaSky

    ExplosionsInDaSky Well-Known Member

    3,163
    2,325
    113
    Sep 13, 2011
    I agree, we ultimately know what we are getting with Fitzpatrick. We're not expected to win the Super Bowl or make the playoffs for that matter. Both he and Tannehill have had multiple opportunities at being "the guy." As a fan of this team and as someone who never roots for losses, I hope we magically catch lightening in a bottle with him and he leads us to a winning season full of excitement. In reality I doubt that happens, but one can hope for a Hollywood ending with this until it becomes that reality. We're in June right now....I haven't even begun to evaluate this or that lol. I'm still watching the NBA finals.
     
    resnor likes this.
  32. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Yeah that's a big one.

    Brees vs. Wilson I'm neutral on, but add in Kuechly and you're right that could tip the scales. However.. keep in mind that missing out on something where many other GM's would have made the correct decision is the bigger mistake, all other things being equal. And many GM's would have either drafted Brees or chosen Brees over Culpepper (either way landing us Brees) while it's unlikely any more than a few GM's (if any) would have picked both Kuechly and Wilson. So you have to consider the odds too.

    Either way, I do know you're one of the few people who advocated for picking those two, so kudos to you on that!
     
    Last edited: Jun 10, 2019
    resnor and djphinfan like this.
  33. KeyFin

    KeyFin Well-Known Member

    10,488
    12,821
    113
    Nov 1, 2009
    I'm pretty sure Tannehill will start most of this season- they didn't trade for him with the intention of him not playing. Mariota spends a lot of time on IR anyway (five different injuries to five different areas in 5 years) so even if he gets the nod in pre-season, RT will likely have his chance either way. So far in camp though, Tannehill is the man (since there's no real QB pressure in camp.....RT will ALWAYS win out in a two hand touch scrimmage).

    I still believe that if you protect him behind a truly elite line, RT is a top 5 quarterback. Somebody just needs to teach the guy to tuck and run when pressured and he'd be okay. And I still think he need hearing aids...just my 2 cents.
     
    resnor likes this.
  34. Finatik

    Finatik Season Ticket Holder Staff Member Club Member

    4,323
    4,012
    113
    May 2, 2014
    SO Cal
    This whole thing is possible or probably hilarious. LoL
     
  35. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    The thing I always come back to on the Brees -v- Culpepper thing is that betting on the recovery of a knee is much safer than betting on the recovery of a shoulder.
    The other issue is both QBs had significant character/reliability questions. The Chargers had basically given up on Brees and drafted Philip Rivers before Brees started playing top tier football then injured his shoulder. The falling out of the Vikings and Culpepper may have been due to club or player issues.
    Looking at their passer rating prior to 2006 (discounting their rookie years as both players weren’t starters as rookies). Brees had gone 76-67-104-89. Culpepper had gone 98-83-75-96-110-72.

    So I don’t see the Brees -v- Culpepper decision to be so one sided as hindsight makes it out to be.

    As for drafting Brees, if we’d have done that we still had Wannstadt as HC, and he refused to take risks on rookies, so if we had drafted Brees there is a good chance we would have given up on him before he had an opportunity to establish himself as a starter.
     
  36. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    The Brees vs. Culpepper thing was closer to a 50/50 proposition at that time so it's not "one-sided" in the sense that it was clear Brees would succeed and Culpepper would not. However, both Saban and the Saints thought Brees was the better candidate. Where Saban failed was he didn't think Brees was worth his asking price, including guaranteed money, while the Saints did. And the details of the contract essentially tell you how that team balances risk vs. potential reward. So sure, the risk was greater with a shoulder injury but that's beside the point. Neither the Dolphins nor Chargers accurately estimated the potential reward. Evals are hard.. people make mistakes, but this was a HUGE mistake in the history of this franchise.

    As far as Wannstedt and drafting Brees.. we didn't draft Brees BECAUSE of Wannstedt. Our scouting department was high on Brees but Wannstedt and our coaching staff didn't think he was an improvement over Fiedler. So there's no real hypothetical here involving drafting Brees while Wannstedt was our coach. It's funny because win% is ultimately what you measure a coach by, but one really has to make an exception for Wannstedt in Miami: 42-31 record yet was in many ways responsible for the fall of a once proud franchise. And that includes his decision not to draft Brees.
     
    The Guy and Pauly like this.
  37. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Same same. I was only pointing out what seems to be how the league is treating them...but I do still think the actual issue keeping most teams away from Tannehill was the two seasons missed for essentially the same injury (I know it was technically different). I don't think it's his performance that kept teams away.

    But I could be wrong.
     
    Last edited: Jun 11, 2019
    KeyFin likes this.
  38. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    If it’s true that Tannehill becomes top-five with an elite offensive line, then you simultaneously have to hope that none of the more typically elite quarterbacks also have elite offensive lines, because if so, then they are certainly outdueling Tannehill in the playoffs.

    If Tannehill jumps into the top five with an elite offensive line, then those other quarterbacks jump into the stratosphere with one.
     
  39. Pauly

    Pauly Season Ticket Holder

    3,696
    3,743
    113
    Nov 29, 2007
    Not really. Those already elite QBs can already perform at that level without an Elite OL. An elite OL covers/masks Tannehill’s greatest flaws. Tannehill will get more benefit from being behind an all-pro line than an Aaron Rodgers or a Tom Brady.

    Yes if you put Dan Marino behind the 1991 Hogs he would have had another 1984 like season. But putting Mark Rypien Behring them turned a journeyman into an All-pro. You are getting into the law of diminishing returns.
     
    Surfs Up 99 and The Guy like this.
  40. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    I know we don't have good "QB pressure" stats, but for a pocket passer like Tannehill sacks should be correlated to pressure. The best-fitting line between Tannehill's 2018-adjusted passer ratings and sacks is PR = 99 – 2.5*S where S = number of sacks in the game.

    In other words, for S=0 sacks (elite OL?) Tannehill's expected rating in 2018 is 99 which would put him 11th among starting QB's, not top 5. Not saying this is the result you'd get if we had good pressure stats, but at least it's an estimate based on data instead of intuition.
     
    Surfs Up 99, Pauly and The Guy like this.

Share This Page