BCS Rankings: Humans vs Computers
Results 1 to 5 of 5
  1. #1
    Pro
    Points: 72,425, Level: 83
    Level completed: 58%, Points required for next Level: 725
    Overall activity: 0%
    Achievements:
    50000 Experience PointsVeteran
    SvrWxCy's Avatar
    Join Date
    Aug 2010
    Location
    Kansas
    Posts
    2,851
    Points
    72,425
    Level
    83
    Thumbs Up
    Received: 0
    Given: 0

    BCS Rankings: Humans vs Computers

    Did this brief blog post on my personal website comparing the human polls (Coaches and Harris) to that of the unbiased computers and computer average in the initial BCS Rankings. Showing which teams may be over-rated and under-rated by the human pollsters. I knew Iowa State would be under-rated, but was surprised by how much and which other teams are getting no love from the human pollsters.


    BCS Rankings: Humans vs. Computers | Severe Plains



    Hint: Iowa State ends up being the most under-rated team!



  2. #2
    Speechless
    Points: 424,090, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 21.0%
    Achievements:
    SocialVeteran50000 Experience Points
    alarson's Avatar
    Join Date
    Mar 2006
    Location
    West Des Moines, IA
    Posts
    21,809
    Points
    424,090
    Level
    100
    Thumbs Up
    Received: 884
    Given: 423

    Re: BCS Rankings: Humans vs Computers

    Yup. The whole human voting process is flawed, from starting ratings too early, to philosophies that teams can't drop if they don't lose, to philosophies that teams *have* to drop if they do lose even if to a significantly higher ranked team (in other words, performing as they should have for that ranking, yet being dropped)



  3. #3
    Speechless
    Points: 734,915, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 61.0%
    Achievements:
    SocialVeteranCreated Album pictures50000 Experience PointsOverdrive
    Cyclonepride's Avatar
    Join Date
    Apr 2006
    Location
    A pineapple under the sea
    Posts
    49,067
    Points
    734,915
    Level
    100
    Thumbs Up
    Received: 2,184
    Given: 1,040

    Re: BCS Rankings: Humans vs Computers

    Yeah, the human pollsters start with a preconceived notion of how good a team is by how good they've been in the past, which is why the traditional powers get a head start every year in the polls, and have to prove they don't belong in the polls, while teams like ISU have to prove they do.



  4. #4
    Starter
    Points: 18,925, Level: 41
    Level completed: 98%, Points required for next Level: 25
    Overall activity: 5.0%
    Achievements:
    Veteran10000 Experience Points
    LivntheCyLife's Avatar
    Join Date
    Nov 2006
    Location
    St. Louis, MO
    Posts
    779
    Points
    18,925
    Level
    41
    Thumbs Up
    Received: 26
    Given: 7

    Re: BCS Rankings: Humans vs Computers

    I agree that human polls are biased, but I think there are obvious flaws to the computer rankings too. I think in an ideal world for BCS selection you'd have some sort of committee similar to basketball that only meets at the end of the year and preferably would take computer rankings as a starting point and adjust from there.



  5. #5
    Addict
    Points: 240,003, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 6.0%
    Achievements:
    50000 Experience PointsSocialVeteran

    Join Date
    Nov 2011
    Posts
    9,262
    Points
    240,003
    Level
    100
    Thumbs Up
    Received: 717
    Given: 509

    Re: BCS Rankings: Humans vs Computers

    One thing I like about our team is we are 4-2 against what Sagarin ranks as the 10th hardest schedule so far. And the two losses were both close games that could have gone either way.



Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
  • TV: Cyclones.tv
  • HOOPS: Iowa State vs. Mississippi Valley State
  • December 31, 2014
  • 06:00 PM