2022 RPI and NCAA Tournament Bracket

Discussion in 'Women's College' started by cpthomas, Aug 19, 2022.

  1. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    It’s a new season, so here we go .... It is all pretty speculative, so take it for what you will.

    Here are Chris Henderson’s pre-season bracket projections:

    [​IMG]

    And, at the RPI and Bracketology for D1 Women’s Soccer blog, I have three new posts:

    2022 PRE-SEASON ASSIGNED RPI RANKS AND RESULTING SIMULATED END-OF-SEASON RPI RANKS

    2022 PRE-SEASON STANDINGS AND NCAA TOURNAMENT AUTOMATIC QUALIFIERS FOR CONFERENCES

    2022 PRE-SEASON PROJECTED NCAA TOURNAMENT SEEDS AND AT LARGE SELECTIONS

    LET THE WILD RUMPUS BEGIN!
     
    NewbieSoccerMom, mpr2477 and Rank Cleats repped this.
  2. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
  3. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    At the RPI and Bracketology for D1 Women’s Soccer blog, I have two new posts with:

    Simulated end-of-season RPI ranks and

    Simulated NCAA Tournament participants,​

    both using the actual results of games played through August 28 and simulated results for games not yet played. Both have lots of room for change between now and the end of the season.
     
    NewbieSoccerMom and mpr2477 repped this.
  4. Enzo the Prince

    Sep 9, 2007
    Club:
    CA River Plate
    That's interesting, you have UNCW as an at-large team? The CAA has often been on the verge of getting a second team in, but I can't remember the last time it happened?

    I'm looking at their schedule and it looks like only Hofstra has potential for a glamor win/RPI boost?
     
  5. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    I believe the Colonial had 3 teams in the Tournament in 2015.

    Something I have noticed over the years is that for the mid-majors, teams that look very good based on my system’s pre-season expectations do not end up getting the expected game results. So, over the course of the season some of the mid-majors that my system has in the tournament end up dropping out. This is one of the reasons I warn about taking my system’s results too seriously early in the season.
     
  6. Cliveworshipper

    Cliveworshipper Member+

    Dec 3, 2006
    are you saying that fewer mid majors make the tournament than your system predicts, or just not the teams the system predicts?
     
  7. Sledhead

    Sledhead Member

    Atalanta
    United States
    Jul 14, 2019
    Drake?
     
  8. NewbieSoccerMom

    NewbieSoccerMom New Member

    Chicago
    United States
    Nov 4, 2021
    I saw the same thing but with Loyola gone out of the MVC; I guess the stats say Drake will win the conference for the bid. I assume thats what he meant.
     
  9. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    To answer a couple of questions in posts:

    1. I am expecting fewer teams from mid-majors to end up in the Tournament as compared to the current projection from my system, as distinguished from the same number but not the same teams.

    2. My system is strictly based on ranking teams and then assigning them ratings based on those ranks. Each team rank comes from a trend formula applied to its historic rank data. As a somewhat simplified explanation, it looks to see whether a team has been trending better, worse, or about the same and then assigns a rank (and corresponding rating) based on where the trend indicates the team will be this year. I was a little surprised when I saw Drake as an AQ, but when my system plays out the season, determines the end-of-season positions within the conference, populates the conference tournament, and plays out the tournament, Drake ends up as the tournament champion. Once I have set up the program, everything else gets played out with no judgment or discretion on my part. Thus the only judgments I make are in the process of setting up the program.

    As I always say, this is just one way of projecting where things will end up. Chris Henderson has a completely different way. The USC has another way. Individual fans have their own ways. They all have some value. My system, however, is completely transparent -- over time, I have provided a comprehensive explanation of how the system works. This lets anyone see what it does and what it does not do. As the season progresses, my system comes closer and closer to where things will end up. At this stage of the season, it at least in come cases is pie in the sky.
     
  10. Sledhead

    Sledhead Member

    Atalanta
    United States
    Jul 14, 2019
    MVC is definitely wide open with Loyola gone. Just not sure Drake would be my early season pick. Then again, I don't know all the criteria used for the ranking, so maybe.
     
  11. Sledhead

    Sledhead Member

    Atalanta
    United States
    Jul 14, 2019
    Makes sense. Thanks for the explanation.
     
  12. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    #12 cpthomas, Sep 5, 2022
    Last edited: Sep 5, 2022
    At the RPI and Bracketology for D1 Women’s Soccer blog, I have two new posts with:

    (1) Simulated end-of-season RPI ranks and

    (2) Simulated Top 60 teams plus automatic qualifiers outside the Top 60, placed in order based on how my simulation says they will score for NCAA Tournament at large selection purposes based on the Committee’s historic patterns

    both using the actual results of games played through September 4 and simulated results for games not yet played.

    The NCAA Tournament post is different than what I have done previously and I believe is likely to have the teams closer to how the Committee will place them for NCAA Tournament purposes.
     
  13. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    At the RPI and Bracketology for D1 Women’s Soccer blog, I have two new posts with:

    (1) Simulated end-of-season Adjusted RPI ranks and

    (2) Simulated Top 60 teams plus automatic qualifiers outside the Top 60, placed in order based on how my simulation says they will score for NCAA Tournament at large selection purposes based on the Committee’s historic patterns

    both using the actual results of games played through September 11 and simulated results for games not yet played.

    In the ARPI rank post, I have provided some additional context that I think those interested in the RPI will find worth reading.
     
  14. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    At the RPI and Bracketology for D1 Women’s Soccer blog, I have three new posts with:

    (1) Current actual RPI ratings and ranks and additional data for each team,

    (2) Simulated end-of-season Adjusted RPI ranks,​

    (3) Evaluations of teams for each of #1, 2, 3, and 4 seeds and at large selections based on Committee historic decision patterns, plus simulated conference champion automatic qualifiers​

    using the actual results of games played through September 18 and simulated results for games not yet played.
     
  15. Tom81

    Tom81 Member+

    Jan 25, 2008
    Really?
    Alabama at #1 and FSU at #40?
    Is that what your data tells you?
     
  16. Number007

    Number007 Member+

    Santos FC
    Brazil
    Aug 29, 2018
    Lets assume it is, what does that tell you about RPI as a measure of relative strength between teams? If FSU played Alabama tomorrow, who is your money on? more importantly who would the Alabama coaches bet on if it had to be an amount meaningful to them?
     
  17. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    Yes.

    Here is a general idea why it looks that way:

    1. The system determines future game results based on the comparative current RPI ratings of the opponents, as adjusted for home field advantage.

    2. Alabama has a better current RPI rating than Florida State. The three current RPI elements for each are:

    a. Alabama: .8500 - .6390 - .6151

    b. Florida State: .8571 - .5941 - .5606

    The last two elements for each team go into the RPI computation of strength of schedule. Thus according to the RPI, Alabama has had a much better bunch of opponents than Florida State.
    3. When you plug the current RPI ratings into all teams’ remaining schedules and determine what all the game outcomes should be based on those ratings, Alabama ends up with a 19-1-2 record and Florida state with an 8-4-4 record.​

    Of course, if these and other teams do differently in their remaining games than their current RPI ratings indicate, things will change.

    As the season goes on from week to week, what my system shows will come closer and closer to the final RPI ratings and ranks.

    Something I have emphasized at the RPI and Bracketology blog: The RPI does not measure how strong teams are. It (supposedly) measures their demonstrated performance. As of today, it says Alabama’s demonstrated performance has been substantially better than Florida State’s, which is why you see the outcome in my current numbers. For Florida State, this means that if it is better than its RPI currently indicates, it is going to have to demonstrate it by outperforming its current RPI in conference play.
     
  18. sweepsit

    sweepsit Member

    Oct 25, 2016
    SF, California
    Nat'l Team:
    United States
    Measures a team’s win/loss performance. It of course misses that fsu has been blowing teams out with ease since their first games of the season.
     
  19. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    Very true, the RPI does not consider goal differential. There are some reasonable policy reasons for that, but it may diminish the RPI as a rating tool.

    On the other hand, Massey does consider goal differential and has a good rating system. I have a modified version of the RPI, however, that does not consider goal differential but that is at least as good as Massey. So, the RPI problem seems not really to be its failure to consider goal diffential.
     
    sweepsit repped this.
  20. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    At the RPI and Bracketology for D1 Women’s Soccer blog, I have three new posts with:

    (1) Current actual RPI ratings and ranks and additional data for each team, including - new this week -- indications of teams within the current ranges for potential seeds and at large selections,

    (2) Simulated end-of-season Adjusted RPI ranks,

    (3) Evaluations of teams for each of #1, 2, 3, and 4 seeds and at large selections based on Committee historic decision patterns, plus simulated conference champion automatic qualifiers

    using the actual results of games played through September 25 and simulated results for games not yet played.
     
  21. Tom81

    Tom81 Member+

    Jan 25, 2008
    I'll take Florida State at #29 for the end of season RPI ranking for $2000 Alex.
    I'll take FSU for top 10 for sure!
    System needs a little tweaking IMO.
     
  22. sweepsit

    sweepsit Member

    Oct 25, 2016
    SF, California
    Nat'l Team:
    United States
    FSU have seven games left. Five of those games are against teams currently in the top seven of the rpi. Does seem unlikely that their rpi would drop that substantially from that stretch.
     
  23. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    If you understand what the system is -- there are very detailed explanations of it in earlier posts -- it is not intended to do what you appear to be thinking it is supposed to do. You are operating on the belief that Florida State will do better in its conference games than its current RPI, as compared to other conference teams’ RPIs, indicates. The system is explicitly designed not to do that, just as the RPI does not do that. If you are right and Florida State is better than its current RPI rating says, then as and when Florida State outperforms its current rating, the system will incorporate that change and you will see an improvement in where it ranks Florida State. That is all it is designed to do.

    As an aside I appreciate the reference to Alex.
     
  24. cpthomas

    cpthomas BigSoccer Supporter

    Portland Thorns
    United States
    Jan 10, 2008
    Portland, Oregon
    Nat'l Team:
    United States
    The big unknowns, I think, are (1) which teams will perform in accord with their current RPI ratings and which teams will not and (2) which games that opponent rating differences indicate are likely to be ties will not be ties and, of those games, who will win them and who will lose them. However, if all teams perform in accord with their ratings over the balance of the season (which, of course, will not happen), and if all games where the rating difference is within tie range end up as ties (which likewise, of course, will not happen), then that is where their RPI will end up. This would not necessarily mean that is where they really are in terms of team strength, it simply would mean that is where the RPI places them based on their winning percentange and the winning percentages of their opponents and opponents’ opponents.

    Like you, I doubt their RPI rank actually will drop to #29 from the current #14, but if everything plays out according to current RPI ratings, it will. Also but, however, as the current candidate ranges for seeds and at large selections indicate, there still are very large changes that will occur.
     
  25. SpeakeroftheHouse

    PSG
    Italy
    Nov 2, 2021
    What is the thought process behind having 148 teams on the bubble? 64 make the tournament with 34 (?) automatic bids.
     

Share This Page