2021 March Madness Experimental Prediction Formula

Collapse

Recommended Videos

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Tyro11
    Rookie
    • May 2020
    • 326

    #1

    2021 March Madness Experimental Prediction Formula

    I have watched ZERO college basketball this year, but I always enjoy March Madness. I also enjoy mathematics, statistics, and analytics. So, I wanted to try to look at key team stats and come up with a prediction formula. And, I did it. I will explain the process below then provide the "Team Rating" and their matchups. I'll bold the ones that are close enough to be interesting.

    Each metric respectively rated from 0-10 based on its relation to the national best (10) or national worst (0).

    Metrics are weighed by importance (my choice) and tallied for a total rating:

    Strength of Schedule: 25%
    Longest Winstreak: 5%
    Wins in Last 10 Games: 10%
    Offensive Efficiency: 20%
    Defensive Efficiency: 20%
    Effective FG%: 5%
    Opp. Effective FG%: 5%
    Turnover %: 5%
    Opp. Turnover %: 5%
    Offensive Rebound %: 5:%
    Opp. Offensive Rebound %: 5%

    Ratings/Matchups

    1 - Gonzaga (8.3) ~ highest rated team
    16 - Norfolk State (4.0)
    16 - Appalachian State (3.9)

    8 - Oklahoma (5.7)
    9 - Missouri (5.0)

    5 - Creighton (6.2)
    12 - UC Santa Barbara (5.8)

    4 - Virginia (6.1)
    13 - Ohio (5.3)

    6 - USC (6.2)
    11 - Drake (6.0)
    ~ upset alert
    11 - Wichita St. (5.3)

    3 - Kansas (6.5)
    14 - Eastern Washington (5.1)

    7 - Oregon (6.0)
    10 - VCU (5.6)

    2 - Iowa (6.8)
    15 - Grand Canyon (5.0)

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    1 - Michigan (6.9)
    16 - Texas Southern (4.6)
    16 - Mt. St. Mary's (3.6)

    8 - LSU (5.7)
    9 - St. Bonaventure (6.1) ~ lower-seed favorite

    5 - Colorado (6.4)
    12 - Georgetown (5.3)

    4 - Florida State (6.1)
    13 - UNC Greensboro (5.1)

    6 - BYU (5.6)
    11 - UCLA (5.4)
    ~ upset alert
    11 - Michigan St. (5.0)

    3 - Texas (6.3)
    14 - Abilene Christian (5.9)
    ~ one of the closest 3/14 matchups

    7 - Uconn (6.2)
    10 - Maryland (5.6)

    2 - Alabama (6.6)
    15 - Iona (4.5)

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    1 - Baylor (7.7)
    16 - Hartford (3.9)

    8 - North Carolina (5.9)
    9 - Wisconsin (5.8)
    ~very close; great seeding by committee

    5 - Villanova (6.1)
    12 - Winthrop (5.8)
    ~upset alert

    4 - Purdue (6.0)
    13 - North Texas (5.6)
    ~another interesting 4/13

    6 - Texas Tech (6.4)
    11 - Utah State (6.2)
    ~upset alert

    3 - Arkansas (6.6)
    14 - Colgate (6.0) ~very high rating for #14 seed; watch out

    7 - Florida (5.5)
    10 - Virginia Tech (5.6)

    2 - Ohio State - (6.0) ~ very low rating for #2 seed
    15 - Oral Roberts (4.2)

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

    1 - Illinois (7.3)
    16 - Drexel (4.2)

    8 - Loyola Chicago (6.8) ~ excellent rating for #8 seed; dark horse
    9 - Georgia Tech (5.9)

    5 - Tennessee (6.2)
    12 - Oregon State (5.0)

    4 - Oklahoma State (5.9)
    13 - Liberty (5.9)
    ~big-time upset alert

    6 - San Diego State (6.9) ~great rating; dark horse
    11 - Syracuse (5.4)

    3 - West Virginia (5.7) ~bad rating for a #3 seed
    14 - Morehead State (4.9)

    7 - Clemson (5.7)
    10 - Rutgers (5.5)

    2 - Houston (7.5)
    15 - Cleveland State (4.0)
  • Tyro11
    Rookie
    • May 2020
    • 326

    #2
    Re: 2021 March Madness Experimental Prediction Formula

    For anyone wondering, I filled out a bracket on the ESPN using these ratings or only these ratings. It currently sits in the 94.4 percentile a 1560 max (quite a lot compared to other brackets). What I am seeing is that it was pretty shoddy the first two rounds (specifically its calls on San Diego State, Illinois, Oregon State, Oral Roberts, etc.) but I believe it did a good job picking the deeper round teams.

    Comment

    • Tyro11
      Rookie
      • May 2020
      • 326

      #3
      Re: 2021 March Madness Experimental Prediction Formula

      The bracket finished in the 93.5 percentile despite Gonzaga losing. If the Zags and Michigan would have won, it would have finished in 99.9 percentile.

      Comment

      • markohenry
        Rookie
        • Dec 2021
        • 7

        #4
        Re: 2021 March Madness Experimental Prediction Formula

        Hello, I am very glad to see this post about the 2021 March Madness Experimental Prediction Formula. I really like this mathematical formula, Actually, I am a final year student of mathematics, Now I am working as a contributor at this math educational forum where we shared different types of mathematical equations, questions with answers that can help the student solve their academic math problems.

        Comment

        • markohenry
          Rookie
          • Dec 2021
          • 7

          #5
          Re: 2021 March Madness Experimental Prediction Formula

          Originally posted by markohenry
          Hello, I am very glad to see this post about the 2021 March Madness Experimental Prediction Formula. I really like this mathematical formula, Actually, I am a final year student of mathematics, Now I am working as a contributor at this math educational forum where we shared different types of mathematical equations, questions with answers that can help the student solve their academic math problems.
          I also love to play basketball.

          Comment

          Working...