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. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    Wow......

    Teams that take longer to throw the ball ask their OLs to hold their blocks longer than teams the throw the ball quickly. What you never seem to understand is that team's are adapting their game plans to deal with issues. Those adaptations come at a cost that are often not measured in the new magic stat. Here is an article that sums it up nicely.

    https://www.windycitygridiron.com/2...n-t-pass-the-eye-test-the-curious-case-of-the

    From the article:

    My problem with PBWR is it makes a fatal assumption. And it is an assumption over something with huge variability.

    PBWR assumes that all pass plays need the same amount of time. In the eyes of this metric a snap and throw WR screen and a seven step drop with a play fake both require 2.5 seconds of pass protection to be successful. That is insane. And that is the fatal flaw of PBWR.

    ......

    After an embarrassing opener, the Bears went to a short/quick passing game week 2 against Denver. Because the ball is out before the defense has time to get pressure, these short passes go as wins in PBWR. This made a bad o-line look better statistically. The irony is that the o-line’s inability to keep a clean pocket long enough to throw down field was a key driver of why Nagy went to a quick passing game. Put another way, the o-line was so bad that Nagy adjusted his game plan to cover for them, and their PBWR stats looked better as a result. That is a problematic stat.
     
    Irishman likes this.
  2. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    I made no such assumption. You were reading too much into a fake example designed to simply show you the math in how PBWR can be misleading. Once again, should have been obvious but sadly no.
     
  3. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    Are you still upset about getting completely exposed over your misinformed take on Wilson? PBWR as a lone measure is invalid. Always has been. I simply blew up your argument with your own stats and words and you cannot handle it. I'm embarrassed for you.
     
  4. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    Let me see if I can summarize the problem with using PBWR as some magic stat:

    1. Doesn't account for differences in time to throw
    2. Doesn't account for differences in number of players in pass protection
    3. Doesn't account for differences in number of pass rushers
    4. Doesn't account for differences in route depth

    The reason PBWR varies less than you might think across the league is because of the factors above. Teams will adjust based on passing game success or failure. Those adjustments come at a cost that is NOT measured by PBWR.

    Take the Dolphins for much of 2013 to 2018. The used a high percentage of quick routes with extra blockers. That may raise the PBWR but at the expense of the passing game. Shorter throws to receivers that are facing more defenders causes a reduction in both air yards and YAC. This results in an offense that needs a ton of plays to complete a scoring drive. This leads to stalled drives and an ineffective offense. Didn't we see far too many bubble screens. All because the OL sucked, despite what a lone stat might tell you.
     
    resnor likes this.
  5. Cashvillesent

    Cashvillesent A female Tannehill fan

    770
    641
    93
    Dec 8, 2019
    If Titans were such a GREAT team how come Mariota couldnt put them over the top. I mean lets be honest, Titans had a ****load of picks in 2016 after they traded with the Rams for the 1st pick in 2015..... but if we know this to be true, if the roster was filled with bunch of talent how come TN has failed for the last 4 years to find a Ryan Tannehill clone? Let me guess, because it is hard to find a talent like Tannehill. Guy is shruging this off as if its easy to find a player like RT17 that can play the QB position.

    How about the Browns? I mean they scored A+ in the offseason last year, but they sucked throughout the whole regular season, could it be the QB? Or was it the talent????? Hmmmmm........
     
    Hiruma78 and FinFaninBuffalo like this.
  6. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    I'm referring to your post where you specifically state that you are guessing about something, and then drew conclusions off of your guess.
     
  7. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    https://www.pro-football-reference.com/players/F/FoleNi00.htm
     
  8. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    There are two significant problems with the article referenced above:

    1) The article says the following: "The advanced stat most people point to measure offensive line in the passing game is Pass Block Win Rate (PBWR), which measures the rate that the line holds their blocks for 2.5 seconds or more. If the ball is thrown before 2.5 seconds and the line has held the block, that’s a win."

    Based on my understanding of the statistic, the boldfaced portion above constitutes neither a win nor a loss. Blocks have to be held at least 2.5 seconds to be a win.

    2) The article also says: "In the majority of games, just about everything looked terrible for the Bears offense. They couldn’t run, they couldn’t pass. They couldn’t run block, they couldn’t pass block. At least that was what the eye test said. But advanced metrics disagreed. PBWR said the Bears were above average pass blocking. They ranked 13th in the NFL with a 61% PBWR."

    The average team's PBWR in 2019 was 58.5%, and the standard deviation in the league was 5.89%. The Bears weren't even a half a standard deviation better than the league average -- they were hardly "above average" in a way that had any meaning. In fact they weren't even a standard deviation better than the 26th-ranked team in the league in PBWR, the Saints, with a PBWR of 56%.

    Again this shows just how little variation there is across the league in this area.

    And of course we don't know the validity of the author's "eye test" he references. For all we know he was watching an average line, which is what PBWR attested to, and the author's non-systematic viewing of the rest of the league prevented him from determining how the Bears' line was truly functioning, relatively speaking.

    This is of course why the "eye test" typically becomes garbage the moment comparisons among teams need to be made. It's impossible to "eye" every team in the league in a systematic manner, and so you have no idea where your team -- the one you're likely eyeing the hardest -- falls in relation to the average team in the league.
     
    Last edited: Apr 17, 2020
  9. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    You're still stuck trying to show that there isn't a linear relationship between holding blocks until 2.5 seconds after the snap, and holding blocks beyond that point.

    The difficulty you're facing there is that it's implausible that the teams that are good at holding their blocks until 2.5 seconds are also the bad ones at holding their blocks beyond that point. There very likely isn't an interaction effect among teams as a function of time after the snap, and there would have to be to support your position.
     
  10. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    The article that I read only says that screens are excluded because the OL purposely lets the defender through. That, to me, implies all other passes are counted, including those that are thrown before 2.5 seconds.

    If you can find a description the specifically states that quick passes are excluded, let me know.
     
  11. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    I'm not stuck trying to show anything. You're the one claiming that PBWR is the true measure of OL performance. Prove it.
     
  12. mooseguts

    mooseguts Well-Known Member

    362
    368
    63
    Jan 12, 2018
    Hmm I don't remember using PBWR as a reference point since PBWR only measure's OL who held their block for 2.5 seconds+ and I never used that as something to hang my hat on. I use PFF and I don't believe PFF uses PBWR as their sole evaluation in their OL ranking if at all.....so how again did you use my own stats against me? It's almost as if you're assuming I believe all OL ranking are created equal which I've never said any such thing, so it's almost like you're blatantly making things up. If you use QB rating as your stat of choice and I bring up ESPN's QBR so as to use your own stat against you it doesn't make much sense does it unless you previously stated ALL qb rating formulas are valid which you haven't. Seriously how does this not make sense to you?

    But yes I'm embarrassed because you assumed because I use PFF it MUST mean I care about which OL held their block for 2.5+ seconds. That's logical? I'm super embarrassed.
     
  13. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    I could not find a written definition but I did find this:

    https://www.espn.com/nfl/story/_/id...g-stats-analytics-explainer-faq-how-work#full

    In that article, they list David Bakhtiari as having an 89% PBWR on 156 rushes through week 4 of the 2018 season. Here are the splits for Aaron Rodgers from the 2018 season.

    https://www.pro-football-reference.com/players/R/RodgAa00/splits/2018/

    If you scroll down the the games 1-4 section, you will see that he had 156 pass attempts.

    I check a bunch of other QBs against the players on that list. Not every one is exact, but all were close. It is clearly NOT excluding quick throws. Other examples:

    Taylor Lewan - 75 total rushes, Marcus Mariota - 77 pass attempts
    Ryan Ramczyk - 146, Drew Brees - 161
    Tyron Smith - 123, Dak Prescott - 115

    These numbers do not include sacks, screens, or plays missed by a player, but clearly the play counts for the linemen include passes that took less than 2.5 seconds.
     
  14. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    Well if Seattle ever fixes the OL and they give Wilson a borderline top 10 pass blocking line and he continues racking up the sacks at his usual pace than we'll know for sure that Wilson was the issue.
     
    Irishman likes this.
  15. mooseguts

    mooseguts Well-Known Member

    362
    368
    63
    Jan 12, 2018
    Now show me the part where I said where PBWR is my method of choice in evaluations. Show me the part where I said all OL evaluations are created equal. I'll wait.

    You assumed because I use PFF it MUST mean I care about which OL held their block for 2.5+ seconds. That's logical? I'm super embarrassed.
     
  16. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    How about an analysis of the Seahawks line from after the end of the 2018 season:

    https://www.seattletimes.com/sports...look-to-keep-offensive-line-together-in-2019/

    Just face it, you assumed that Wilson had never played behind a decent line and if he had, his sack rate would drop. You were wrong.

    I didn't assume anything. I knew their line was better in 2018. I simply did a search and that was the first ranking that came up. After that I looked at Wilson's sack rate and almost pissed myself laughing when I found out it was higher than usual. But, my explanation of why holds. They ran slower developing plays and Wilson is excellent at extending plays. It works for Seattle.

    The only point is that you cannot look at any stat (including sack rate or PBWR) in a vacuum.
     
    Last edited: Apr 17, 2020
    Cashvillesent likes this.
  17. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    I'm not sure that provides definitive clarity when on the same page there is the following statement:

    "When blocks are held for at least 2.5 seconds or more, we call that a win for the blocker."

    So it's possible that they're tallying total numbers of pass dropbacks but not including "wins" on passes thrown in under 2.5 seconds in the calculation of pass block win rate.
     
  18. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    Guess you should not be using a statistic that is so poorly understood. Either:

    1. They count all drop backs and the numbers are skewed by different offensive strategies

    or

    2. They only count drop backs > 2.5 seconds and the numbers are skewed by counting differing percentages of total drop backs for different teams.

    Either way, the stat is too flawed to be used in isolation like you are trying to do. This is in addition to all the other flaws I pointed out.
     
    resnor likes this.
  19. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Yeah well, when you compare that to the only alternative -- the "eye test" -- I'll take this.
     
  20. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    There is a third option.

    Use the statistics to give yourself a broad overview and then use your eyes to confirm what actually happened to end up at that number.
     
    resnor likes this.
  21. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    You'd think that would be obvious by now.....

    Another suggestion is to watch some games, use more than one stat, watch some games, research the systems used by the teams you are interested in, watch some games, read some expert analysis, and watch some games.
     
    Last edited: Apr 17, 2020
  22. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    LOL at "only alternative".....

    No surprise that you are against actually watching games.
     
  23. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Unfortunately there's no reliable way of determining a precise deviation from a league norm with the eye test. Give me a z-score based solely on the eye test, for example. You can't, and no one can.
     
  24. resnor

    resnor Derp Sherpa

    16,329
    9,874
    113
    Nov 25, 2007
    New Hampshire
    Z scores don't win games.

    You understand that most people don't care about z scores? When you say z scores, people's eyes glaze over.
     
  25. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    I dont understand how I said using both and you took it to how you cant use just your eyes. I didnt imply that.

    You arent looking for a Z-Score with your eyes. That isn't the purpose at all.
     
    FinFaninBuffalo and resnor like this.
  26. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    For ****s and giggles, I wonder what kind of z-score a QB like Culpepper put up. He is the perfect example of a slightly above average when healthy QB who put up monster numbers due to his receivers and whose skill set meshed up with his cast.

    Then I'd like to see someone like McNabb who I considered a much better QB with a crappy cast.
     
  27. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    Unfortunately, there's no reliable way of determining a precise deviation from a league norm of on field performance of any player or position group using flawed statistics either.
     
  28. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    What I'm saying regards making comparisons among players and teams. As soon as you make a comparison among players or teams, the relevant question becomes the degree of deviation from a league norm (cbrad's all-time highest z-scores for passer rating above are an example). No one can establish a league norm on the basis of the eye test, and certainly no one can determine a precise degree of deviation from it on the basis of the eye test.

    Answer this question using the eye test only -- how much better was the Tennessee Titans' defense than the average defense in the league in 2019? First, good luck with just accomplishing that, and second, good luck convincing anybody you're right solely on the basis of your eye test assessment.

    As for PFF, one would have to have access to the inter-rater reliability of their assessments of players for that approach to have any validity. If there is no inter-reliability, there is no validity. If PFF assessment person X thinks Ryan Fitzpatrick played great and PFF assessment person Y thinks he played far less great, we have a major problem with PFF's player grades.
     
  29. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Sure they do. There is most certainly a degree of deviation from the league norm in scoring that's associated with a high win probability.
     
  30. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    No there isn't, but again, when the two options are the statistics or the personal observations, I'll take the former when it comes to the kind of comparison being made here. At least in the case of the former there were systematic observations being made, as opposed to random ones.
     
  31. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    You're mistakenly thinking that the statistics themselves actually provide an accurate picture. They dont either, they are like a partially torn map on their own.

    Also, you CAN use statistics to evaluate a unit. I wouldnt use the eye test on an entire unit. The statistics of a defense inherently include everything. An individual player on the other hand it is impossible to entirely separate from the factors around them, making statistics less accurate a gauge than they would be for a unit.
     
    FinFaninBuffalo and resnor like this.
  32. Cashvillesent

    Cashvillesent A female Tannehill fan

    770
    641
    93
    Dec 8, 2019
     
    Hiruma78 likes this.
  33. AGuyNamedAlex

    AGuyNamedAlex Well-Known Member

    3,582
    2,579
    113
    Sep 12, 2015
    Right or wrong, you realize an NFL team would fire you as a scout immediately if you said you've seen the numbers but havent watched any film right?

    How do you know if a player fits your offense? How do you know if a back is Reggie Bush or Jerome Bettis without actually seeing them?

    How do you know if a guy is Randy Moss or Anquan Boldin?

    Are all those players great? Yep, some fit in different schemes better than others and if you didnt use their skill set within your offense, their play would suffer.
     
  34. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    Notice in all of the posts I just made above I said some form of "in the kinds of comparisons being made here." What you're saying above falls outside that category.
     
  35. The Guy

    The Guy Well-Known Member

    6,598
    3,323
    113
    Oct 1, 2018
    It's funny how that can be the way a person who has no knowledge in this area can experience what I said in the post you quoted, whereas in a group of statisticians I'd be talking at only a very foundational level, and they would experience it along the lines of 2 + 2 = 4.
     
  36. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    The best option is BOTH. Adding the personal observations will help you understand what you are trying to measure.
     
  37. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    Every other person in the conversation is using observations in the "kinds of comparisons being made here".
     
  38. FinFaninBuffalo

    FinFaninBuffalo Well-Known Member

    2,474
    2,954
    113
    Dec 13, 2007
    I feel exactly the same way when discussing football concepts with you.
     
    Hiruma78 and Cashvillesent like this.
  39. Cashvillesent

    Cashvillesent A female Tannehill fan

    770
    641
    93
    Dec 8, 2019
    Says a guy that doesnt even watch NFL games but looks at the stats all day. Lol
     
    Hiruma78 likes this.
  40. JPPT1974

    JPPT1974 2022 Mother's Day and May Flowers!

    410
    84
    28
    Apr 15, 2012
    Glad he found a team that need his services!
     
    Cashvillesent likes this.

Share This Page