2025 Ratings and Ranks

For my projected end-of-season NCAA RPI ratings and ranks, incorporating the actual results of Week 1 games and using projected results for games not yet played, you can see my 2025 Article 14: RPI Report After Week 1 Games. Or, if you want the tables provided in that article in Excel workbook form, you can download it using this link: RPI Report After Week 1 Games. [NOTE: If you use the Excel workbook download link, you will see the workbook in a GoogleSheets format. Rather than using that, in the File menu click on Download and in the Download menu click on Excel. This will download the workbook as an Excel document which will be easier to use.]

My weekly reports this year are set up with an emphasis on the differences between teams', conferences, and regions' NCAA RPI ranks and their ranks as Strength of Schedule contributors within the NCAA RPI formula. This should give you a clear picture of which teams, conferences, and regions get hurt by or benefit from the way the formula computes Strength of Schedule. If you spend even a little time looking at the numbers, the discrimination patterns will be quite clear.

The reports also will include rankings using my Balanced RPI, which is a modification of the NCAA RPI that does not have the discriminatory patterns you will see for the NCAA RPI. In addition, the reports will include Massey's ranks, which are currently available since they incorporate past years' ratings, and the KPI ranks once they are available. This will let you see how the KPI ranks tend to be similar to the NCAA RPI ranks whereas the Massey ranks tend to be similar to the Balanced RPI ranks. The Women's Soccer Committee currently will be using the KPI ranks as supplemental to the NCAA RPI, having selected those ranks to use rather than either Massey or the Balanced RPI.

In addition, for regions the reports will show how each of the four geographic playing regions' games are distributed in terms of in-region and out-of-region games. This is important, since out-of-region games are what allows the NCAA RPI to properly rate teams from a region in relation to teams from other regions. This year, unfortunately, the numbers of out-of-region games are reduced from historic patterns by about 30%.

Finally, for regions the reports also will show the proportion of each region's games that are ties. The proportion of ties is a surrogate for in-region parity (and matches another measure of in-region parity which is the shape of the bell curve of in-region distributions of winning percentages). The amount of in-region parity is important since the way the NCAA RPI measures strength of schedule discriminates against regions with higher in-region parity and in favor of regions with lower in-region parity.

The early weekly reports are more for entertainment and also education about the NCAA formula's problems. With each passing week, as I replace more and more predicted results with actual results, the reports will become better and better at predicting where teams will end up come the end of the season.

 

For my projected end-of-season NCAA RPI ratings and ranks, incorporating the actual results of Week 2 games and using projected results for games not yet played, you can see my 2025 Article 15: RPI Report After Week 2 Games. Or, if you want the tables provided in that article in Excel workbook form, you can download it using this link: RPI Report After Week 2 Games. [NOTE: If you use the Excel workbook download link, you will see the workbook in a GoogleSheets format. Rather than using that, in the File menu click on Download and in the Download menu click on Excel. This will download the workbook as an Excel document which will be easier to use.]

In this week's 2025 Article 15, I discuss the Women's Soccer Committee's change this year in the NCAA RPI formula, devaluing ties in the Winning Percentage computation from 1/2 of a win to 1/3 of a win, and show that the change has slightly increased the NCAA's already existing discrimination against regions with higher in-region parity and in favor of regions with lower in-region parity. I also discuss a concern the Committee should have about this year's reduction in out-of-region travel and how that reduction may affect the NCAA RPI's ability to properly rate teams within a single national system.

 

For my projected end-of-season NCAA RPI ratings and ranks, incorporating the actual results of Week 3 games (including the Labor Day games) and using projected results for games not yet played, you can see my 2025 Article 16: RPI Report After Week 3 Games. Or, if you want the tables provided in that article in Excel workbook form, you can download my workbook using this link: RPI Report After Week 3 Games. [NOTE: If you use the Excel workbook download link, you will see the workbook in a GoogleSheets format. Rather than using that, in the GoogleSheet File menu click on Download and in the Download menu click on Excel. This will download the workbook as an Excel document which will be easier to use.]

Something I am following is, for each region, the percentage of in-region games that are ties. I follow this because the RPI's design, presumably unintentionally, discriminates against teams from regions with high levels of parity, one of the indicators of which is the precentage of in-region games that are ties. We'll see where the numbers end up at the end of the season, but as of now I am projecting the West and Middle regions will have extraordinarily high levels of in-region ties.

 

For my projected end-of-season NCAA RPI ratings and ranks, incorporating the actual results of Week 4 games and using projected results for games not yet played, you can see my 2025 Article 17: RPI Report After Week 4 Games. Or, if you want the tables provided in that article in Excel workbook form, you can download my workbook using this link: RPI Report After Week 4 Games. [NOTE: If you use the Excel workbook download link, you will see the workbook in a GoogleSheets format. Rather than using that, in the GoogleSheet File menu click on Download and in the Download menu click on Excel. This will download the workbook as an Excel document which will be easier to use.]

As has appeared to be the case once teams published their schedules, there is going to be a significant reduction in teams' out-of-geographic-region games this year. The most isolated region, the West, will see an almost 1/3 reduction.

In addition, so far it is looking like there will be an increase in the percentage of games that are ties. The West appears likely to have the highest percentage of in-region ties with the South having the lowest percentage. With the NCAA having changed the NCAA RPI's winning percentage calculation so that ties now count as a third of a win rather than a half of a win, it appears the change on average will hurt the ratings of the West's teams (and the North's teams this year), to the benefit of the Middle's and South's teams. For the West, the high percentage of tie games is not surprising since historically it has had greater in-region parity than the other regions.

Altogether, I see the Women's Soccer Committee having to deal with some pretty serious problems for the NCAA RPI this year. Before the change in the NCAA RPI's valuation of tie games, the NCAA RPI already discriminated against teams from the West -- on average, they have had better results in out-of-region games than their ratings said they should have had. The tie valuation change slightly increases the discrimination. And now, with reduced out-of-region travel, the NCAA RPI's ability to properly rate teams from a geographic region in relation to teams from other geographic regions is likely to be further weakened. Whether the Committee will be able to handle these problems well is a pretty big question.

 

For (1) the actual current NCAA RPI ranks based on games through Sunday, September 14, and more, and (2) projected end-of-season ranks, and more, if teams perform from here on out exactly in accord with their current NCAA RPI ratings, see my 2025 Article 18: RPI Reports After Week 5 Games.

 

For (1) the actual current NCAA RPI ranks based on games through Sunday, September 21, and more, and (2) projected end-of-season ranks, and more, if teams perform from here on out exactly in accord with their current NCAA RPI ratings, see my 2025 Article 19: RPI Reports After Week 5 Games.

Reminder: The linked article has links to Excel workbooks with the tables in the article. If you want to download the workbooks, which is how I recommend looking at the tables:

1. Click on the link, which will open a GoogleSheets workbook.

2. Click on File, at the upper left of the GoogleSheets page.

3. In the drop down menu, click on Download.

4. In the drop down menu, click on Microsoft Excel.

That will download the Excel workbook.

 

Correct!

 

For (1) the actual current NCAA RPI ranks based on games through Sunday, September 28, and more, and (2) projected end-of-season ranks, and more, if teams perform from here on out exactly in accord with their current NCAA RPI ratings, see my 2025 Article 20: RPI Reports After Week 7 Games.

At the beginning of the article, I discuss what I believe will be significant problems the Women's Soccer Committee will face this year due to:

1. A significant decline in out-of-region competition;

2. The regions having differing levels of in-region parity;

3. The difference between how the RPI rates/ranks teams overall and how it rates/ranks them, within the RPI formula, as strength of schedule contributors to their opponents.​

 

For (1) the actual current NCAA RPI ranks based on games through Sunday, October 5, and more, and (2) projected end-of-season ranks, and more, if teams perform from here on out exactly in accord with their current NCAA RPI ratings, see my 2025 Article 21: RPI Reports After Week 8 Games.

At the beginning of the article, I provide some summary data on the size of the differences between teams' NCAA RPI overall ranks and their ranks under the NCAA RPI formula as strength of schedule contributors to their opponents -- the differences are large -- and similar information for the Balanced RPI -- where the differences are minimal. This relates to why the NCAA RPI has discrimination problems when rating teams from a conference in relation to teams from other conferences and from a region in relation to teams from other regions -- and the Balanced RPI does not.

I've also slightly reconfigured the tables for the current ranks so you can easily compare how the NCAA RPI, KPI, Balanced RPI, and Massey ranks compare.

 

NCAA RPI #57 is the poorest ranked team to get into the Tournament. #27 is the best rank (using the formula then in effect) not to get into the Tournament. So the at large "bubble" is teams #27 through #57.

 

This week I've done a bonus article. It uses the Women's Soccer Committee's historic NCAA Tournament bracket formation patterns to predict, what the Committee's decisions will be this year as compared to what they would be if the Committee were using the Balanced RPi rather than the NCAA RPI. The prediction is based on the actual results of games played through October 5 and predicted results of games not yet played.

2025 Article 22: Predicted NCAA Tournament Bracket Based on the NCAA RPI as Compared to the Balanced RPI

 

Here's an interesting piece of data that illustrates the problem with how the NCAA computes the strength of schedule portion of the RPI, which accounts for 50% of the RPI's effective weight:

Fairfield currently ranks #13 in terms of its strength of schedule contribution to its opponents within the NCAA RPI formula. The formula thus considers them a stronger opponent than these who are next in line: West Virginia, Georgetown, Illinois, Wake Forest, Western Michigan (!!!), Colorado, South Carolina, Saint Louis, and North Carolina.

We are about two thirds of the way through the season, so you might think this will change by the end of the season. It will. But ... my full season projection, based on teams performing exactly in accord with their current NCAA RPI ratings for their remaining games, has Fairfield finishing as the #5 strength of schedule contributor.

Does anyone believe Fairfield is close to being the #13, much less #5, strongest opponent to play?

 

Right. The reason Fairfield improves as a strength of schedule contributor is that within the NCAA RPI formula, their winning percentage makes up 80% of the effective weight of their contribution whereas the teams against whom they achieved their winning percentage account for only 20% of the effective weight.

This is one of the main reasons the NCAA stopped using the RPI for basketball, which matters to the NCAA a whole lot more than the sports that still use the RPI. That the money from basketball matters is understandable,. Sadly, the result is that the other sports get second class treatment on issues like the rating system the NCAA uses.

 

For (1) the actual current NCAA RPI ranks based on games through Sunday, October 12, and more, and (2) projected end-of-season ranks, and more, if teams perform from here on out exactly in accord with their current NCAA RPI ratings, see my 2025 Article 23: RPI Reports After Week 9 Games.

The actual current ranks report includes new data that compares teams', conferences', and regions' actual results so far to what their current NCAA RPI ratings say the results should have been. This lets you see which teams,, conferences, and regions are performing better than or poorer than their NCAA RPI ratings indicate they should have performed.

 

For (1) the actual current NCAA RPI ranks based on games through Sunday, October 12, and more, and (2) projected end-of-season ranks, and more, if teams perform from here on out exactly in accord with their current NCAA RPI ratings, see my 2025 Article 24: RPI Reports After Week 10 Games.

The actual current ranks report includes new data that compares teams', conferences', and regions' actual results so far to what their current NCAA RPI ratings say the results should have been. This lets you see which teams,, conferences, and regions are performing better than or poorer than their NCAA RPI ratings indicate they should have performed.

In addition, the last part of the report shows my projected NCAA Tournament bracket based on the Women's Soccer Committee's historic patterns and also what the bracket would look like if the Committee were using the Balanced RPI rather than the NCAA RPI.

 

For (1) the actual current NCAA RPI ranks based on games through Sunday, October 26, and more, and (2) projected end-of-season ranks, and more, if teams perform from here on out exactly in accord with their current NCAA RPI ratings, see my 2025 Article 25: RPI Reports After Week 11 Games.

In addition, there's a bonus report this week, 2025 Article 26: NCAA Tournament Bracket Projections After Week 11 Games and In/Out Case Study. The bonus report shows my projected NCAA Tournament bracket based on the Women's Soccer Committee's historic patterns and also what the bracket would look like if the Committee were using the Balanced RPI rather than the NCAA RPI. It also shows Chris Henderson's current bracket prediction, so you can compare all three. And in a further addition, it has a case study of why under my current projections, the Committee using the NCAA RPI would give Saint Louis an at large position and not give one to Kansas State whereas if using the Balanced RPI it would give Kansas State the position and not give it to Saint Louis.

 

If Fairfield doesn't win the conference tournament, I don't think they get an at large position. They played no Top 50 opponents, thus have no Top 50 results much less Top 50 wins or ties. Their RPI rank won't be able to overcome that.

The reason for their high NCAA RPI rating is that they play in the weakest regional playing pool in the country, by a lot. And, they play no out-of-region games. And, regional isolation is greater this year than ever (inter-regional games are down about 30%). So their rank basically reflects where they are in their region, not where they are in the country.

Your choice of Fairfield to look at is a good one, as they illustrate a big problem the Committee is going to have this year. On the other hand, I think this is a very easy call for the Committee if Fairfield doesn't win their tournament.