What role should analytics/data science play in our organization?

Discussion in 'Texans Talk' started by Texanmike02, Jun 10, 2019.

  1. Texanmike02

    Texanmike02 Site Contributor

    Joined:
    Apr 20, 2007
    Messages:
    8,933
    Likes Received:
    1,352
    In terms of analytics/machine learning utilization the three major sports look something like this:
    Baseball >>>> Basketball >>>>>>>>>> Football

    I think, however, it is time to start evaluating players/coaches using advanced metrics. I have seen all of the claims that "football is different", but I think we're at the point that we should be able to derive actionable intelligence. I know the Browns are analytics heavy (so I guess we should avoid whatever the hell they are doing lol) but are there any up and coming GM's that are using real ML technology to evaluate players?

    I don't know the answer to that, but what I have in mind is kind of a football outsiders (on steroids) where algorithms are breaking down players movements/tells/breaking ability and play recognition.

    Anyone else have experience in the ml/analytics field? How do you see this playing out?
     
  2. amazing80

    amazing80 Site Contributor

    Joined:
    Oct 13, 2011
    Messages:
    4,706
    Likes Received:
    2,945
    Location:
    Elkhron, WI
    This is going to be my opinion with very little knowledge of analytics in general.

    Baseball and basketball use analytics when it comes to players and situations. Not just player ability. The NFL needs to use analytics during the game. Use it while determining down and distance, play calls, when to punt vs field goal vs go for it, etc. I think using analytics to evaluate player talent will never work in the NFL.
     
    DocBar and TheMatrix31 like this.
  3. ATXtexanfan

    ATXtexanfan Site Contributor

    Joined:
    Apr 14, 2007
    Messages:
    6,641
    Likes Received:
    2,008
    Location:
    leander, tx
    When the likes of Brady and brees figure out your defense pre snap

    Everything goes out the window
     
  4. TheKDog

    TheKDog Hall of Fame

    Joined:
    Sep 30, 2013
    Messages:
    3,716
    Likes Received:
    3,433
    Obrien is pretty dismissive of it
     
  5. ATXtexanfan

    ATXtexanfan Site Contributor

    Joined:
    Apr 14, 2007
    Messages:
    6,641
    Likes Received:
    2,008
    Location:
    leander, tx
    Kinda like to timeouts in critical situations lol

    Lol yeah dude hates analytics
     
  6. Pharaohcious

    Pharaohcious Veteran

    Joined:
    Feb 5, 2018
    Messages:
    369
    Likes Received:
    100
    No comment
     
  7. TheMatrix31

    TheMatrix31 Hall of Fame

    Joined:
    Aug 6, 2010
    Messages:
    10,299
    Likes Received:
    2,069
    Location:
    California
    This about sums it up, I'd say.
     
  8. TexansBull

    TexansBull Hall of Fame

    Joined:
    Jun 16, 2006
    Messages:
    5,502
    Likes Received:
    3,437
    Not his job...

    and

    Doesn’t matter Brian.AFC South Divisional Champs.
     
  9. Texecutioner

    Texecutioner Hall of Fame

    Joined:
    Jun 12, 2008
    Messages:
    22,623
    Likes Received:
    5,894
    ^^^^ This.
     
  10. welsh texan

    welsh texan Site Contributor

    Joined:
    Feb 10, 2005
    Messages:
    2,137
    Likes Received:
    722
    Location:
    Llandudno, Wales
    I would be seeing the Browns as more of a blueprint team than the Pats or Steelers at this point.

    They have finally got it together.

    I was however shocked that they moved for OBJ, as i thought one of the things analytics would tell you is that a great team does not require great WRs (and certainly isnt necessarily composed with a lot of salary cap at the position) but then i guess they are on the rookie QB cap pass atm.
     
    TexansFight and steelbtexan like this.


  11. Toro Bravo

    Toro Bravo Rookie

    Joined:
    Dec 25, 2018
    Messages:
    65
    Likes Received:
    96
    IMO, analytics should definitely be a portion of the process for determining your roster - maybe 40%. It's basically a statistical analysis of player performance vs. expense (which can include cost to acquire, bonuses, salary, etc...).

    The Browns hired Paul DePodesta in 2016 and he has been a major contributor to their roster talent. Their odds to win SB LIV are 5th best according to versainsider.com.
    Per DePodesta; “Analytics is not about sitting behind a computer and pushing enter and having it produce an answer,” he said, per ESPN. “This game is not a simulation. It’s played by real people and because of that, there’s just a tremendous amount of uncertainty that surrounds it. For us, it’s about how we use information, how we use data to really get our arms around that uncertainty.”

    It's only as good as the data used but it should part of the evaluation process.
     
    steelbtexan and 76Texan like this.
  12. Texansballer74

    Texansballer74 Hall of Fame

    Joined:
    Aug 14, 2016
    Messages:
    5,440
    Likes Received:
    4,243
    Please don't involve this analytical approach to our football. The high tech nerds can keep that schit to themselves.
    That nonsense has ruined my freaking Rockets. Yeah they're in the race but when it's money time, that analytic crap doesn't work.
     
    Mr teX likes this.
  13. Speedy

    Speedy Yeller Dweller

    Joined:
    May 1, 2004
    Messages:
    6,647
    Likes Received:
    5,257
    Location:
    The Livable Forest
    Have they?? I get the intrigue with all the moves they've made, but they haven't done squat on the field yet.

    I know sports is about copycatting success, but you have to be successful first. Nobody was going to do what the Astros did if it didn't work. They won a WS, now everybody's trying to do it that way.

    Let's pump the brakes on following the blueprint of a franchise that has been a perennial dumpster fire for two decades, until they actually do something on the field one time.
     
  14. TheKDog

    TheKDog Hall of Fame

    Joined:
    Sep 30, 2013
    Messages:
    3,716
    Likes Received:
    3,433
  15. Texansballer74

    Texansballer74 Hall of Fame

    Joined:
    Aug 14, 2016
    Messages:
    5,440
    Likes Received:
    4,243

    Crazy on how these people view things on paper. The Browns haven't seen success in years and they have them that high. Lol
     
  16. Mr teX

    Mr teX Hall of Fame

    Joined:
    Sep 20, 2006
    Messages:
    11,664
    Likes Received:
    3,801
    Location:
    Houston, TX
    Analytics should always be apart of the pie....No more than 30-40% though imo. The issue is these nerds in positions of power that attempt to make it the majority of the pie when selecting talent & building teams. & that's been my issue with Morey & why I've largely been rolling my eyes at the Rockets every since Morey took over. His analytics that drive his 3 & D philosophy say that we should be chucking 3's @ a ridiculous rate & he uses GSW as his example of why it works. What his analytics don't/can't factor in is that GSW has 3 of the all-time greatest shooters & players anchoring them. His system is using streaky, 1 trick I-can-only-shoot-it-well-from-1-spot guys. That's fine for the regular season, it'll keep you compettive...but there's a limit to that..

    Analytics also can't factor in what's in a guys heart & his drive either. No way any analytic could told folks TB12 was going to be what he was...Altuve, same thing.
     
  17. DocBar

    DocBar Hall of Fame

    Joined:
    Apr 6, 2005
    Messages:
    14,532
    Likes Received:
    5,149
    Location:
    I'm international
    The Browns were analytics heavy under Sashi Brown. They now have John Dorsey calling the shots. Depending on how Freddie Kitchens ends up as a HC, the Browns look very good on paper.

    As far as analytics go, I don't think it's a very useful tool for individual players, but could be very useful in-game as another member posted. It's too dependent on every players execution on a given play in the NFL than it is in MLB or the NBA. You would have to know every players assignment on every single play and then determine whose play was more positive or negative on the play, then try to determine the magnitude of the positive or negative players impact on the overall play call. That's a bunch and it's very subjective.
     
    Last edited: Jun 12, 2019
  18. TheKDog

    TheKDog Hall of Fame

    Joined:
    Sep 30, 2013
    Messages:
    3,716
    Likes Received:
    3,433
    It's all about where the money goes. Vegas doesn't want to put them too low then have a huge bet come in
     
    steelbtexan and Texansballer74 like this.
  19. Speedy

    Speedy Yeller Dweller

    Joined:
    May 1, 2004
    Messages:
    6,647
    Likes Received:
    5,257
    Location:
    The Livable Forest
    It doesn't help that Morey doesn't seem to get the players necessary to apply their 3 & D philosophy. Signing a career 25% 3pt shooter like MCW and 58 year old Melo, doesn't quite fit what you're trying to do. That'd be like the Texans drafting 2 TEs in a limited draft and never using them or bringing in a Lamar Miller and ramming him up the middle.

    I'm far from an analytics nerd, but I don't think that's how it works.
     
    CWTexansFan and Mr teX like this.
  20. Uncle Rico

    Uncle Rico Pro BS'er

    Joined:
    Nov 19, 2012
    Messages:
    4,711
    Likes Received:
    4,431
    Location:
    Near the Mountains
    Analytics when used properly offer a different look at the same situation. IMO it offers the best use in football during the preparation phase when implementing tendency and game theory. Having to sift through data on the sidelines in the 45 seconds between plays is too much to ask of anybody thats where the "feel" that a good head coach has or doesnt have comes into play in football. Only sport where the coach has as much of an impact in winning games as the players themselves. Not embracing analytics in the modern world of sport would only give your opponent an advantage. QB hot zones, offensive line alignment tendency, defensive alignments man there is a plethora of stuff you could probably numerate and install as a teaching tool if nothing else.

    ps: Darryl Morey is so far ahead of the curve and thinking about moves 9,10, and 11 in a series of events that complaining about one or two of his moves is silly. The man has not fielded a losing team in what a decade all the while trying to build a contender which he has. Uncanny ability to find diamonds in the rough and also a master of the NBA salary cap and how to work it. Not afraid to make a SPLASH move, but he cuts his losses fast when he does make a mistake and rarely makes the same mistake twice. Great GM.
     

Share This Page