GauchoDan's 2016 RPI

Discussion in 'College & Amateur Soccer' started by Gaucho95, Aug 10, 2016.

  1. OverseasView

    OverseasView Member+

    Olympique Lyonnais
    France
    Feb 3, 2013
    Club:
    Olympique Lyonnais
    Nat'l Team:
    France
    The GauchoDan talked to us!:geek:
    I go at least twice a day on your site to check whether the RPI of my sons' teams stopped to slip.:confused::cry:
    You are almost part of our family.....;)
     
  2. Sandon Mibut

    Sandon Mibut Member+

    Feb 13, 2001
    This.

    I got there a couple of times a day. A great resource for anyone who follows college soccer. Your work is so appreciated.

    PS - I don't even bother looking at any other rankings. I have no clue who is No. 1 in the NSCAA or any of the other polls. It's yours or nothing.
     
    fknbuflobo and jdgaucho repped this.
  3. WatzBruin

    WatzBruin Member

    Aug 23, 2011
    LA - SF Valley
    Club:
    FC Barcelona
    Truly appreciate your commitment to this ambitious endeavor GauchoDan!

    Your work is my go to destination to view the results of each play date and the correlating dynamic RPI!

    Thanks again!
     
    jdgaucho repped this.
  4. jdgaucho

    jdgaucho Member

    Jan 8, 2012
    San Diego
    Club:
    Los Angeles Galaxy
    Nat'l Team:
    United States
    GauchoDan posts this for basketball, volleyball, women's soccer, etc. as well on the Big West boards.

    See, Sandon? There are good Gauchos!
     
    CCSoccerFan and Sandon Mibut repped this.
  5. gsterp

    gsterp Member+

    Jul 16, 2003
    College Park
    Have you figured out the proper multiplier to account for math having an East Coast bias?
     
    OverseasView and Sandon Mibut repped this.
  6. Sandon Mibut

    Sandon Mibut Member+

    Feb 13, 2001
    To wit:

    The ACC currently has 7 of the top 11 teams in the RPI and two more in the top 25.

    1. Clemson: 9-2-3 --- 6
    2. Notre Dame: 10-2-1 --- 7
    4. Louisville: 10-3-2 --- 8
    6. Virginia: 8-2-4 --- 4
    7. Wake Forest: 11-2-2 --- 37
    8. Syracuse: 10-3-1 --- 15
    11. North Carolina 10-2-2 --- 44
    21. Boston College: 6-5-2 --- 2
    23. Virginia Tech: 9-3-3: --- 52

    Here are the other eight teams in the Top 15 of the RPI. Only two schools from west of Chicago.

    3. Loyola-Chicago: 10-1-1 --- 20 (Missouri Valley)
    5. Maryland: 12-0-2 --- 72 (Big Ten)
    9. Charlotte: 8-2-1 --- 41 (Conference USA)
    10. Stanford: 8-2-3 --- 19 (Pac-12)
    12. Massachusetts-Lowell: 10-0-2 --- 97 (America East)
    13. Kentucky: 9-3-2 --- 14 (Conference USA)
    14. Washington: 9-4-0 --- 30 (Pac-12)
    15. Butler: 8-4-1 --- 17 (Big East)
     
    OverseasView repped this.
  7. scoachd1

    scoachd1 Member+

    Jun 2, 2004
    Southern California
    Not a fan of RPI, but am a fan of GauchoDan.
     
  8. gauchodan

    gauchodan Member+

    Oct 18, 2016
    Thanks, all. I'm actually not much of a soccer fan, but I do like to find a place where I can add value. I'm glad everyone enjoys the product. It's getting down to the nitty gritty time.
     
    TimB4Last repped this.
  9. stphnsn

    stphnsn Member+

    Jan 30, 2009
    which teams should be there? if you were making at top 15, who else would be there from west of chicago? i've only watched iu games so i have no idea.

    creighton, denver, utah valley, and ucla are all within the top 25. that's 6/25. ~25% isn't bad considering the distribution of the teams across the country, is it?
     
    Sandon Mibut repped this.
  10. gauchodan

    gauchodan Member+

    Oct 18, 2016
    NCAA has released its latest calculation of the RPI, through games of Nov. 6, 2016. My calculation matches exactly. All 206 positions are correct.
     
  11. WatzBruin

    WatzBruin Member

    Aug 23, 2011
    LA - SF Valley
    Club:
    FC Barcelona
    Impressive stuff GauchoDan! The NCAA should have you on a retainer!

    Thanks for the dynamic RPI!
     
  12. gauchodan

    gauchodan Member+

    Oct 18, 2016
    The final release is out, and once again I have all 206 positions correct...both normal and adjusted.

    I discontinue updating the RPI at this point. The NCAA does not...they will have a final release after the tournament is over with updated numbers. I don't understand why they do that, but whatever. If the purpose of the RPI is to assist the Selection Committee in selecting and seeding the tournament, its work is done.

    I will continue providing daily scoreboards and follow/document the tournament in that sense, but the RPI's will not move on my website.
     
  13. collegesoccer

    collegesoccer Member+

    Apr 11, 2005
    Gaucho Dan - You are the Man!
     
  14. scoachd1

    scoachd1 Member+

    Jun 2, 2004
    Southern California
    [QUOTE="gauchodan, post: 34828308, member: 296748"I discontinue updating the RPI at this point. The NCAA does not...they will have a final release after the tournament is over with updated numbers. I don't understand why they do that, but whatever. [/QUOTE]
    They probably do this is that it helps the final RPI to be more correlated with the final results since teams in the tournament have higher Win percentages and OOP each round. It gives more of an illusion that the system is primarily predicated on results as opposed to what really serves as the primary driver of the numbers which is scheduling. Nice work efficiently gathering all the data and reverse engineering the bonus structure. It makes it a lot easier for fans to see what is going to happen with their team.
     
  15. espola

    espola Member+

    Feb 12, 2006
    What part of the RPI is not based on results?
     
  16. scoachd1

    scoachd1 Member+

    Jun 2, 2004
    Southern California
    It is amazing that after posting for years, you still don't realize that 3/4 exclude the teams results? No wonder so many of your posts described things as VooDoo.
     
  17. espola

    espola Member+

    Feb 12, 2006
    The 3/4 are the results of their opponents and their opponents' opponents. What other method would you use to set the value of their own record?

    I think the 3/4 is too high and the committees for other NCAA sports agree. But it is a clearly defined and anyone can calculate the results - no voodoo needed.
     
  18. SBZipfan

    SBZipfan Member

    Nov 19, 2014
    Santa Barbara, CA
    I'm not aware of any NCAA sport that uses any number other than 25% for a team's own record in RPI. Hockey changes the other two percentages (from 50 & 25 to 21 & 54), but not the 25 for the team itself. AFAIK, all other sports use the standard 25-50-25.

    RPI is an extremely primitive measure. You'd think the national collegiate athletic association could do better.
     
  19. espola

    espola Member+

    Feb 12, 2006
    Ice hockey only uses RPI as one of the measures in the pair-wise comparison chart which is the "final" method for selecting tournament teams. The other measures have changed from time to time often as well as the percentages used to calculate the RPI.
     
  20. SBZipfan

    SBZipfan Member

    Nov 19, 2014
    Santa Barbara, CA
    RPI is by far the most important measure in hockey's pairwise comparison (as it stands today). If Team A has a lower RPI than Team B, then the only way for Team A to win the comparison is to have both a winning record head-to-head against Team B, and a better record against common opponents. If either of those are tied, Team B wins the comparison. The number of comparisons that are flipped on this basis is very small in practice.

    But you still haven't backed up your claim that there are other NCAA sports that use an RPI that assigns a higher weight than 25% to a team's own record.
     

Share This Page