Page 4 of 16 FirstFirst ... 23456 ... LastLast
Results 37 to 48 of 187
  1. #37
    Join Date
    Aug 2008
    Location
    Pasadena
    Posts
    14,123
    Blog Entries
    4

    Re: You guys are allergic to math

    Quote Originally Posted by Unibomber View Post
    It's wrong because it's wrong. Even when it worked, I still thought it was a risk not worth taking. Nothing against math, but this isn't chess. Football analytics are not yet advanced enough to beat common sense and experience.
    This thinking is one of the big reasons why Belichick and the Pats have been so successful.

    This is from a 2015 article

    “Owner Robert Kraft worked with a former colleague in the 1990s to create statistical models for player valuation. And for the past 15 years, Belichick has relied heavily on his football research director, Ernie Adams, a former Wall Street trader who collaborates with the coach to develop a variety of cutting-edge approaches to team building and game play.

    Belichick recently told The Boston Globe: “Ernie’s really a great sounding board for me personally and other members of our staff. Particularly coaching staff. Strategy, rules, decisions. Ernie’s very, very smart.”

    “One major strategy employed by the Patriots has been an arbitrage system in personnel, whether multiplying draft picks via draft day trades or moving their veteran players (such as defensive tackle Richard Seymour in 2009, receiver Randy Moss in 2010 and offensive lineman Logan Mankins in 2014) before they lose value. Based in part on such moves, the Patriots have had unmatched success in the Belichick era, with four Super Bowl rings and counting.”

    Owner Robert Kraft has always been the gregarious side of the Patriots, compared to Belichick’s Gregg Popovich, but Kraft’s business acumen and Ivy League education fit this approach like a glove. Assets, depreciating value, return on investment, elastic demand, and all those other words that your portfolio manager uses that you don’t really understand all come into play in the team’s final roster.

    Also, it’d be completely logical to expect that NFL teams employing a heavy dose of analytics post more wins, hoist more Lombardi trophies, and get more parades.
    https://chowderandchampions.com/2015...ers-analytics/





  2. #38
    Join Date
    Jun 2011
    Location
    South Florida
    Posts
    64,935
    Blog Entries
    1

    Re: You guys are allergic to math

    Quote Originally Posted by BigPlayReceiver View Post
    This year, the NFL is celebrating 100 years of existence, while dem Greek fellers credited for the discovery and application of mathematics preceded Christ, soooo....
    yeah this new fangled math thing is hurting some of these guys heads...

    World Domination 3 Points at a Time!





  3. #39
    Join Date
    Dec 2009
    Location
    Baltimore
    Posts
    3,009

    Re: You guys are allergic to math

    I was told by Bc to stop by and receive lessons in remedial math. Where do we start?





  4. Re: You guys are allergic to math

    Quote Originally Posted by Trackmaster View Post
    It all depends on how high you think your odds of converting the 2 and the 1 are. Historically over the past two years, the 2 point conversion has been about 50%, and the 1 point conversion has been about 94% with the new placement. But that may differ for every team and every offense. In our case, JT is more like 99.9%, so that really makes it more of a toss up, it just depends on what our 2 point conversion percentage should be.

    Assuming that we give ourselves 50% for the 2, and 99% for the 1, the 2 point conversion should be favored unless there's strong situational evidence for the other. And if you think about, its not as clearcut as it sounds.

    You're down 11, so it sounds great to make it a cool 10, with no real need for 9 points. But, let's not forget something: in that case a FG and TD only sends it to OT (assuming nothing else happens), with 9 points, the FG and TD wins it outright. If you value the odds of picking up the 2 at more than half of your odds at the 1, going for 2 makes sense, as it really does in most cases at that point.
    Loba has an excellent post in a different thread that hits on this point, but when there's such a small difference in expectation, the secondary considerations are magnified (don't listen to me bumble through, go read about why Loba played the lottery despite negative expected value).
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





  5. #41
    Join Date
    Jun 2011
    Location
    The Land of Verdite
    Posts
    52,649
    Blog Entries
    1

    Re: You guys are allergic to math

    Math can’t capture gamers, the way players play through injuries, because you can’t quantify those things. That doesn’t mean they don’t exist and it doesn’t mean they aren’t very important.





  6. #42
    Join Date
    Jul 2011
    Location
    Right Where I need to be
    Posts
    2,252

    Re: You guys are allergic to math

    Um. I guess that MATH does not work for passing vs rushing.

    Hairbaugh cant have it both ways man.


    Not bad for a RUNNING BACK!!! Now that is funny.
    #FIREROMAN





  7. #43

    Re: You guys are allergic to math

    Quote Originally Posted by Paintballguy View Post
    Harbs calls for going for it were great. My only gripe were the actual play calls. I don't like calling a rollout on 4th and 2.

    4th and 2 or 1 should always be automatic go for it calls unless you're deep in you're own territory.

    I highly recommend giving this article a read. https://www.advancedfootballanalytic...4th-down-study
    Nice read. None of these are taking into account the Mahomes factor in going for it on 4th in you're own territory.

    The expected points of we made the first down from our 38 IF we made the first down at the 40 would be 1.3.
    Punting would be a -.5

    Failing to convert using this person's NFL data then gives the Chiefs Expected Points of 2.5 right away.

    This is where analytics can easily fail due to variables entering the equation. While a chart on Expected Points throughout NFL history is one set of data
    Patrick Mahomes completely is a outlier. His expected points from the opponents 38 are not just 2.5. Probably over 4.

    That changes the decision dramatically





  8. #44
    Join Date
    Jun 2011
    Location
    The Land of Verdite
    Posts
    52,649
    Blog Entries
    1

    Re: You guys are allergic to math

    Analytics can be a good research point. One of my key issues with analytics and the way they are portrayed is that they aren't perfect either. It's like somebody trying to tout how marijuana is better for the body than opiates. That doesn't mean that marijuana doesn't carry its own health risks. Just because analytics says it was the best percentage chance, doesn't mean it automatically was and was automatically the right decision. Yet, that's how analytics are portrayed.
    "Please take with you this final sword, The Excellector. I am praying that your journey will be guided by the light", Leon Shore





  9. Re: You guys are allergic to math

    Quote Originally Posted by BearArms View Post
    Nice read. None of these are taking into account the Mahomes factor in going for it on 4th in you're own territory.

    The expected points of we made the first down from our 38 IF we made the first down at the 40 would be 1.3.
    Punting would be a -.5

    Failing to convert using this person's NFL data then gives the Chiefs Expected Points of 2.5 right away.

    This is where analytics can easily fail due to variables entering the equation. While a chart on Expected Points throughout NFL history is one set of data
    Patrick Mahomes completely is a outlier. His expected points from the opponents 38 are not just 2.5. Probably over 4.

    That changes the decision dramatically
    His expected points after a punt are almost certainly not 0.5 either. You don't get to make adjustments on one side only to prove your point.
    Shared Google Folder with Ravens spreadsheets, nextGen charts, and more! Please share my content! (attribution to Twitter requested)

    Knight of the Kingdom of Perfect Play, Student of The Bill James School of Stamping Out Bullshit. Main Sources: PFR, particularly the Play Index; for cap stuff, RSR's Brian McFarland (secondary: OverTheCap, Spotrac)





  10. #46

    Re: You guys are allergic to math

    Quote Originally Posted by organizedchaos21 View Post
    His expected points after a punt are almost certainly not 0.5 either. You don't get to make adjustments on one side only to prove your point.
    Yeah, should have stated that But again the EP can be skewed by Koch dropping in the coffin corner. Mahomes probably has his EP pretty set no matter what field position he starts at.

    I still think you don't give him a short field early in the game. I like going for it on 4th, just not to that extreme even if the math favored it by only 1% against league history. Add Mahomes and the field position and it's a bad choice





  11. #47

    Re: You guys are allergic to math

    Somewhere in the roots of this whole analytics argument is a reasonable estimate of the probability of success for a two point conversion.
    I think JH and his analytics department have over estimated this success rate for his offense.
    They need better two point plays and execution for the "analytics" argument to make sense.

    They should definitely not have Hollywood on the field on two point conversions, too small to block, too small to throw a fade or back shoulder to, too small to make a high leaping grab in the back of the endzone… but there he was doing nothing and challenging no one in the JH/ Roman 2 point offense.





  12. #48

    Re: You guys are allergic to math

    Quote Originally Posted by BearArms View Post
    Nice read. None of these are taking into account the Mahomes factor in going for it on 4th in you're own territory.

    The expected points of we made the first down from our 38 IF we made the first down at the 40 would be 1.3.
    Punting would be a -.5



    Failing to convert using this person's NFL data then gives the Chiefs Expected Points of 2.5 right away.

    This is where analytics can easily fail due to variables entering the equation. While a chart on Expected Points throughout NFL history is one set of data
    Patrick Mahomes completely is a outlier. His expected points from the opponents 38 are not just 2.5. Probably over 4.

    That changes the decision dramatically
    If we are quoting from this article, it was written in 2014, using stats from 2003-2008. So completely worthless.

    And taking into account Mahomes is very easily statistically, and all it does is further back up the point of going for it because his expected value on any drive is going to be higher than any other qb. Therefore keeping it away from him is always going to be gto for us.





Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Link To Mobile Site
var infolinks_pid = 3297965; var infolinks_wsid = 0; //—->