I just have published a series of posts below this one. If you scroll down through them, they'll hopefully be in a logical order. Here's what they are, in the order in which you'll see them as you go down through them:
Weekly RPI Report: Games Through September 25
This report shows teams' actual ARPI ratings and ranks. I've included only the top 148 teams, for a simple reason: Based on the last 9 years' experience, no team ranked poorer than #148 has gotten an at large selection or seed in the NCAA Tournament. At the RPI for Division I Women's Soccer website, however, I've posted a full RPI report covering all teams and games through September 25, so if you want more detail or all the teams, you can go there to see the full report. It's at the bottom of the RPI: This Year's Reports page, in the form of a downloadable Excel file.
For this week's RPI report, based on the last 9 years' experience with the NCAA Tournament, teams have the following tournament possibilities based on their rankings:
#1 Seed possible: #28 or better
#2 Seed possible: #42 or better
#3 Seed possible: #85 or better
#4 Seed possible: #85 or better
At Large Selection bubble: #17 through #148 (Better than #17 should get selections; poorer than #148 should not get selections.)
My ratings and ranks are slightly different than the NCAA's. This is because my data include the 9/18 game of UNC Asheville (H) 4 v Alabama A&M 1, whereas the NCAA's data, at this point, do not include that game. In addition, my data include the 8/23 game of USC Upstate (H) 0 v Winthrop 1, whereas the NCAA's RPI data incorrectly identify the score as 0-0. The NCAA is aware of these issues, so perhaps they will make corrections.
Notwithstanding these differences, I am confident that the NCAA has not made any changes in its adjustment bonus and penalty system since last year. I satisfied myself on this by temporarily changing the data I mentioned in the preceding paragraph to match the NCAA's and then re-computing teams ratings and rankings. I then matched these with the NCAA's ratings published at the NCAA's RPI Archive and found that my ratings, using last year's adjustment bonus and penalty system, match the NCAA's current ratings. This means that the NCAA is using last year's system.
2016 Season Simulation: Week 6 Update
This report is a continuation of the simulations I've published previously this year. This week's update is a hybrid of all games actually played through September 25 plus all games not yet played with simulated results. It shows teams' simulated end-of-season (including conference tournaments) ARPI ratings, rankings, and win-loss-tie records.
2016 NCAA Bracket Simulation: Week 6 Update
This report is my results after plugging the Season Simulation numbers into my system for identifying what the Women's Soccer Committee will do, if it makes decisions consistent with the ones it's made over the last 9 years. In looking at this report, the numbers in the left hand column represent the simulated decisions:
1 = #1 seed
2 = #2 seed
3 = #3 seed
4 = #4 seed
5 = unseeded automatically qualifying conference champions
6 = unseeded at large selections
7 = next teams in line
It's important to note that no team ranked more poorly than #57 has gotten an at large selection over the last 9 years, so my simulation abides by that.
2016 Season Simulation: How Teams Have Progressed, Through September 25
This report shows teams' weekly simulated rankings, starting with their pre-season simulated rankings, followed by their weekly updated simulated rankings as the season has progressed. In the right-hand column of the report, you'll see the cumulative change from a team's pre-season simulated ranking to its currently simulated ranking. This column shows which teams have performed as expected and which teams have performed better or worse than expected.
In addition, by reviewing a team's simulated rank as it's moved through the season, you may be able to guess whether a team's rank appears to have stabilized after some initial change or whether it appears still to be on the move.