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

Football Outsiders Hates Tannehill's Performance on Sunday

Discussion in 'Miami Dolphins Forum' started by GARDENHEAD, Sep 13, 2016.

  1. GARDENHEAD

    GARDENHEAD Season Ticket Holder Club Member

    11,681
    10,413
    113
    May 7, 2008
    New Orleans
  2. Rocky Raccoon

    Rocky Raccoon Greasepaint Ghost Staff Member

    30,224
    36,965
    113
    Dec 2, 2007
    Jersey
    Meh, I thought he played OK. Wasn't great, wasn't bad, made some big plays in crunch time in a tough environment.

    I also know this thread will turn into 10 pages by lunchtime so I'll see myself out from here.
     
  3. dirtylandry

    dirtylandry Well-Known Member

    4,214
    1,750
    113
    Aug 2, 2015
    if Stills makes the catch and they convert on4th, what happens?
     
  4. GARDENHEAD

    GARDENHEAD Season Ticket Holder Club Member

    11,681
    10,413
    113
    May 7, 2008
    New Orleans
    Read the article.
     
  5. Dol-Fan Dupree

    Dol-Fan Dupree Tank? Who is Tank? I am Guy Incognito.

    40,533
    33,035
    113
    Dec 11, 2007
    Wow Marcus 'give the game away to the Vikings' Mariota was 17th!?!?!?!

    I can't take this site seriously.
     
  6. gunn34

    gunn34 I miss Don & Dan

    21,755
    3,475
    113
    Jan 5, 2008
    Oviedo FL
    This article is more about their very own DYAR then anything Miami. They just used the Phins game as a token to get their message out about their metric.
     
    resnor likes this.
  7. emocomputerjock

    emocomputerjock Senior Member

    5,649
    1,853
    113
    Nov 23, 2007
    DC
    It happens. I thought they were going to integrate some EPA/WPA data in their stuff, but they didn't. Overall though it's silly to try to use a metric like DYAR or DVOA at this point in the season.
     
    3Pmi likes this.
  8. josh

    josh Well-Known Member

    1,138
    1,297
    113
    Oct 8, 2013
    Raleigh, NC
    That article was nearly devoid of content. They start by stating the fact that Miami had the worst passer and rusher of the week. Then it's two paragraphs qualifying that statement, since some other team ended up worse. Finally the meat of the article was about other times in history when that's happened, like anyone gives a crap.
     
    resnor likes this.
  9. KeyFin

    KeyFin Well-Known Member

    10,488
    12,821
    113
    Nov 1, 2009
    Yeah, this was basically one person's opinion using a formula that's based on season-wide stats...and ignoring how solid Seattle was on D last year. Can you really call it a "formula" when half the formula is ignored? I don't think so...especially when there's an update halfway through saying that the whole article is lie anyway.
     
    resnor likes this.
  10. heylookatme

    heylookatme Well-Known Member

    902
    438
    63
    Sep 12, 2012
    Wait, what? That was the worst game of Mariota's career. He was dreadful.
     
    resnor likes this.
  11. smahtaz

    smahtaz Pimpin Ain't Easy

    That's weird. Cian Fahey posted these about Tannehill's game.

    [video=twitter;775419884826230784]https://twitter.com/Cianaf/status/775419884826230784[/video]
    [video=twitter;775421131390128129]https://twitter.com/Cianaf/status/775421131390128129[/video]
    [video=twitter;775424110809735168]https://twitter.com/Cianaf/status/775424110809735168[/video]
    [video=twitter;775437101781422080]https://twitter.com/Cianaf/status/775437101781422080[/video]
     
  12. josh

    josh Well-Known Member

    1,138
    1,297
    113
    Oct 8, 2013
    Raleigh, NC
    Thanks! It's nice to see something positive after just reading some crap about how no one would trade for Tannehill. The real reason why no one would trade for our starting QB is this thing called psychology. We have no alternative starter, so the act of putting him up for trade signals that we, who are most familiar with him, think he sucks. You trade your back up QB, after he has a good game or three, not your starting QB... Jeez!
     
  13. vt_dolfan

    vt_dolfan Season Ticket Holder Club Member

    Cian Fahey will have is QB grades out Thursday.....Im gonna go out on a limb and predict he will not say that 17 was the second to worst QB this past week.

    Whats DYAR anyways?
     
    Shane Falco, resnor and smahtaz like this.
  14. smahtaz

    smahtaz Pimpin Ain't Easy

    Not really sure. I read the definition and still don't know.

    http://www.footballoutsiders.com/info/methods#dyar

    There are a few number crunchers around here. Maybe one of them can explain it in English.
     
  15. finserg

    finserg Well-Known Member

    1,690
    252
    83
    Dec 16, 2007
    Linden,NJ
    Everyone would thing Tanns the next Marino
     
    resnor likes this.
  16. Fin D

    Fin D Sigh

    72,252
    43,684
    113
    Nov 27, 2007
    Its a wholly bananas bat**** cuckoo bird grade, that takes other more standard stats like DVOA, then through the magic of Bud Lite induced bro banter, they decide how much better that number is than what they're back up or replacement would do in that their place.
     
    resnor and smahtaz like this.
  17. DolphinGreg

    DolphinGreg Season Ticket Holder Club Member

    5,227
    6,527
    113
    Dec 7, 2014
    I don't think many people pay attention to that long-winded dbag to be honest.
     
    smahtaz likes this.
  18. smahtaz

    smahtaz Pimpin Ain't Easy

    Are they saying MM probably would of had a better game?
     
    Fin D likes this.
  19. cbrad

    cbrad .

    10,659
    12,657
    113
    Dec 21, 2014
    Personally I have no respect for DYAR or the DVOA it's based on so if it were up to me I'd never report it.

    That link isn't transparent about what DYAR or DVOA is and that was my first reason for objection (it's imperative you explain your methodology.. otherwise the default assumption is you're just making stuff up), but Stringer Bell sent me a PM long time ago with a link that provided greater clarity about what it is:
    https://codeandfootball.wordpress.c...alized-adjusted-net-success-rate-per-attempt/

    Basically, DVOA is substituting success rate (defined by whether EPA = expected points added was positive or not) for yards in AYA. Long story short, AYA has problems of its own. One link I went to showed AYA formula was:

    AYA = (YDs + 20*TD's - 45*INTs)/ATT

    The problem is that these values of "20" and "45" are not justified in any way I can accept. In mathematics, a "+" means "or", meaning the value of 1 TD = 20 yards in that formula. The way they get numbers like that is by taking the EPA of say 1st and goal (e.g. 6.8 points), looking at the difference between that and the 6 points in a TD (so now you have 0.8), and dividing it by a linearly extrapolated value of 0.08 points per yard (taken from 6 points for a TD to -2 for a safety = 8 points difference divided by 100 yards), and getting a value like 10. A newer version of their model got 20.

    Two major problems with that approach are:

    1) Using arbitrary anchor points, like 1st and goal. Any analysis that gives you some value of "20" yards = 1 TD or so should work no matter what anchor point one uses and that's not true here.

    2) Assuming linear extrapolation (every point between the two anchor points has equal weight) for all 100 yards when the number of attempts and EPA at different points on the field are different. What I mean is that they say 6 points for a TD and -2 for a safety, so that means 8/100 = 0.08 per yard?? No, you have to look at EPA at every point on the field and then weight it by how often that actually occurs. Can't just linearly extrapolate when you have the data for attempts and EPA at different points.


    So in summary, you can call AYA, DYAR and DVOA a "model", and its utility can be measured by how well it helps you predict whatever you want. But you CANNOT call it a stat because it's not a logical implication from data (at most you can allow some simple assumption about the distribution of the data but no more.. and in this case even that isn't allowed because you HAVE that distribution).
     
    djphinfan, resnor, Two Tacos and 3 others like this.
  20. 3Pmi

    3Pmi IT Specialist Staff Member Administrator

    2,619
    7,083
    113
    Mar 23, 2008
    Michigan
    Eyeball test beats stats, every day, all day.
     
    adamprez2003, MAFishFan, Agua and 2 others like this.
  21. I would of liked to see RT handle the presure of the rush better and he could of made better reads but he is far from the bottom. I saw wome things from him tjat were not there last year. Its only been o.e game but he looks improved yo me and last year i comsindered him a middle of.the pack starter. He still in that area just maybe a few rungs higher


    Sent from my SM-G920P using Tapatalk
     
    smahtaz likes this.
  22. Unlucky 13

    Unlucky 13 Team Raheem Club Member

    51,930
    63,009
    113
    Apr 24, 2012
    Troy, Virginia
    Their system is so bizarre that its beyond understanding. Its mostly built on assumptions on who they think will be good, and who will be bad, and they hate the Dolphins on an annual basis, so we're always ranked at the bottom.
     
  23. RevRick

    RevRick Long Haired Leaping Gnome Club Member

    7,191
    3,940
    113
    Nov 25, 2007
    Thomasville, GA
    C'mon, Man. Tell us how you REALLY feel about that data analysis...

    For my own, I will quote a Navy Chief I remember very well. It was about "40 pounds of hammered horse sh!t." The rest of the quote I cannot write on the forum.. It gets rather obscene from this point on.!
     
    resnor and Fin D like this.
  24. Fin-O

    Fin-O Initiated Club Member

    11,375
    11,392
    113
    Sep 28, 2015
    10x out of 10.

    These opinions are good entertainment though.


    Sent from my iPhone using Tapatalk
     

Share This Page