RPI and Bracketology for D1 Women's Soccer Blogspace

Tuesday, October 13, 2026

2026 ARTICLE 18: ACTUAL CURRENT AND SIMULATED END-OF-SEASON RANKINGS AND OTHER DATA AFTER WEEK 9

This week, I'm combining my two regular reports into one. Later in the week, I'll follow this up with a case study of the Big 10 and how its scheduling this year, as compared to the scheduling of the other Autonomy (Power 4) conferences, appears to be affecting its NCAA Tournament prospects.

Below, I have provide links to two Exel workbooks. To download a workbook, use the following steps:

1. Click on the workbook link.
2. The link takes you to a Google spreadsheet, which you will not want to use. On the Google spreadsheet, to the left, click on File, which will bring up a drop down menu.
3. In the menu, click on Download, which will bring up another drop down menu.
4. In the menu, click on Microsoft Excel (.xlsx). This will download the Excel workbook.

ACTUAL CURRENT RANKINGS AND OTHER DATA AFTER WEEK 9

The first workbook provides current NCAA RPI, KPI, Balanced RPI, and Massey rankings and other data based on the actual results of games played through Week 9 (through October 11) of the season, for teams, conferences, and regions. Here is a link to the workbook:

2026 RPI Report Actual Results Only as of 10.11.2026

The first page of the workbook is for Teams, the second for Conferences, and the third for Regions. The data include almost all the non-conference games for the season and a portion of the conference games, about two-thirds of all games that teams will play. There still is room for significant changes in where teams finally will end up.

Here is what the Teams page shows, so far as potential NCAA Tournament seeds and at large selections are concerned:

SIMULATED END-OF-SEASON RANKINGS AND NCAA TOURNAMENT BRACKETS FOLLOWING WEEK 9

The second workbook is based on the actual results of games played through Week 9 (ending October 11) and simulated results of games not yet played, including conference tournament games. It projects where teams will end up at the end of the season, but before the NCAA Tournament, and provides other important information related to teams, conferences, and regions. Here is a link to an the workbook:

RPI Report as of 10.11.2026

2026 Article 8 has an explanation of the method I use to generate the simulated ranks and of the key information in the workbook.

For purposes of evaluating the usefulness of the ratings the simulation currently uses to predict the results of games not yet played:

The table's reference to Team 1 is to the home team or, if a neutral site game, the team whose name comes first alphabetically.

The first data row in the table, 2026 Team 1 Actual Results, shows Team 1's actual win, loss, and tie percentages in games played through October 11. The second data row, 2026 Team 1 Predicted Results, Simulated NCAA RPI, applies teams' simulated end-of-season RPI ratings to games played through October 11, to determine what Team 1's win, loss, and tie percentages should be if those ratings are correct. The third data row -- 2026 Actual Less Predicted Results -- shows the difference between the actual percentages and the predicted percentages. Thus, for example, the actual winning percentage for all Team 1s is 47.9%. According to teams' end-of-season simulated ratings, the winning percentage should be 49.0%. The Actual Less Predicted Results row shows the difference between predicted and actual is -1.148%.

The last two data rows are similar, but instead of using teams' simulated end-of-season RPI ratings, it uses teams' actual current NCAA RPI ratings to determine what the percentages should be if the actual current NCAA RPI ratings are correct. Here, the Actual Less Predicted Results row shows that predicted results based on actual current NCAA RPI ratings miss the actual results by 1.341%.

In other words, the simulation's current ratings are a better match with actual results than are the actual current NCAA RPI ratings. Because of this, for now I will continue using the simulation's ratings to predict future game results rather than converting to using the NCAA RPI's current ratings. In the upcoming weeks, I will continue doing this until the NCAA RPI's current ratings perform better than the simulation's current ratings.

The following table is from the workbook. If teams end up where the simulation currently says they will end up, it shows the teams -- if historic patterns hold -- that will be candidates for each seed level and for at large selections. It also shows, in the At Large Bubble column, which teams are at ranking levels that historically always have gotten at large selections (if not Automatic Qualifiers) -- the Top 26 in the NCAA RPI rankings. Thus the at large bubble teams are #27 through #57 (if not AQs). The table also shows which teams the simulation currently projects as Automatic Qualifiers.

SIMULATED NCAA TOURNAMENT BRACKETS USING THE NCAA RPI AND USING THE BALANCED RPI

Here are currently simulated NCAA Tournament brackets, first based on the NCAA RPI and then based on the Balanced RPI. The simulations use the Women's Soccer Committee's historic decision patterns and likely are very close to what the Committee would decide based on the game results to date and simulated results of games not yet played, depending on which rating system the Committee is using.

Following the two brackets is a table showing how the two compare. (NOTE: For conference tournaments, I have used simulated conference tournaments based on the NCAA RPI.)

Here is a key to the tables:

The first table shows the projected NCAA Tournament bracket based on the Committee's using the NCAA RPI:

The next table shows the projected bracket based on the Committee's using the Balanced RPI:

The final table compares the projected brackets. It shows the teams sorted first by region and within regions by conference so you can see how a change to the Balanced RPI would affect regions' and conferences' teams. Salmon highlighting means the team gets a poorer treatment by the rating system and green means the team gets better treatment. No highlighting means the treatment is the same.

NOTE: My weekly Comparison tables are preliminary and evolve throughout the season, It is not the details that are important but rather the overall impression in terms of the difference between using the NCAA RPI as compared to the Balanced RPI.

Looking at the table, you can see that the ACC and, especially, the Big 10 are hurt by the NCAA RPI relative to the Balanced RPI. In contrast, the non-Autonomy (i.e., non-Power 4) conferences are helped by the NCAA RPI relative to the Balanced RPI. This is not surprising since, as I have shown elsewhere, the NCAA RPI discriminates against stronger conferences and regions and in favor of weaker ones whereas the Balanced RPI does not.

It remains to be seen over the rest of the season whether the pattern shown by the comparison table will remain as clear as it is now. I suspect it will moderate some, but time will tell.

Wednesday, October 7, 2026

2026 ARTICLE 17: ACTUAL CURRENT RANKINGS AND OTHER DATA AFTER WEEK 8

This Article 17 provides current NCAA RPI, KPI, Balanced RPI, and Massey rankings based on the actual results of games played through Week 8 (through October 4) of the season, for teams, conferences, and regions.

Here is a link to an Excel workbook that contains all the information:

2026 RPI Report Actual Results Only as of 10.4.2026

To download the workbook, use the following steps:

1. Click on the workbook link.
2. The link takes you to a Google spreadsheet, which you will not want to use. On the Google spreadsheet, to the left, click on File, which will bring up a drop down menu.
3. In the menu, click on Download, which will bring up another drop down menu.
4. In the drop down menu, click on Microsoft Excel (.xlsx). This will download the Excel workbook.

The first page of the workbook is for Teams, the second for Conferences, and the third for Regions. The data so far are based mostly on non-conference games, but also on the first group of in-conference games. I consider these data preliminary and expect significant changes over the remainder of the season.

Here is what the Teams page shows, so far as potential NCAA Tournament seeds and at large selections are concerned:

This week, here are some data taken from the linked workbook's Conferences page:

This table shows how teams' actual winning percentages so far in non-conference games compare to what one would have expected their winning percentages to be based on their current actual NCAA RPI ratings. This is an indicator of how well or poorly the current NCAA RPI ratings do at rating teams from each conference in relation to teams from other conferences. A positive number means the conference's teams have performed better so far in non-conference games than their NCAA RPI ratings say they should have performed (i.e., are underrated) and a negative number means they have performed more poorly (are overrated).

As you can see, the Ivy League's teams currently appear to be significantly underrated. Next come three of the conferences from the West region mixed together with the Autonomy conferences. It will be interesting to see if, and how, these numbers change over the remainder of the season.

2026 ARTICLE 16: SIMULATED END-OF-SEASON RANKINGS AND NCAA TOURNAMENT BRACKETS FOLLOWING WEEK 8

This article is based on the actual results of games played through Week 8 (ending October 4) and simulated results of games not yet played, including conference tournament games. It projects where teams will end up at the end of the season, but before the NCAA Tournament.

Here is a link to an Excel workbook with the simulated full-season NCAA RPI and Balanced RPI ranks -- and other important information -- for teams, conferences, and regions.

RPI Report as of 10.4.2026

2026 Article 8 has an explanation of the method I use to generate the simulated ranks and of the key information in the workbook. Article 8 also has instructions on the best way to download the workbook.

For purposes of evaluating the usefulness of the ratings the simulation currently uses to predict the results of games not yet played:

The table's reference to Team 1 is to the home team or, if a neutral site game, the team whose name comes first alphabetically.

The first data row in the table, 2026 Team 1 Actual Results, shows Team 1's actual results. The second data row, 2026 Team 1 Predicted Results, applies teams' simulated end-of-season RPI ratings to games actually played through September 27, to determine what Team 1's win, loss, and tie percentages should be if those ratings are correct. The third data row -- 2026 Actual Less Predicted Results -- shows the difference between the predicted percentages and the actual percentages. Thus, for example, the actual winning percentage for all Team 1s was 48.1%. According to teams' end-of-season simulated ratings, the winning percentage should have been 49.4%. The Actual Less Predicted Results row shows the difference between predicted and actual was -1.311%.

The last two data rows are similar, but instead of using teams' simulated end-of-season RPI ratings, it uses teams' actual current NCAA RPI ratings to determine what the percentages should be if those ratings are correct. Here, the Actual Less Predicted Results row shows that predicted results based on actual current NCAA RPI ratings miss the actual results by 1.678%.

The following table is from the workbook. If teams end up where the simulation currently says they will end up, it shows the teams -- if historic patterns hold -- that will be candidates for each seed level and for at large selections. It also shows, in the At Large Bubble column, which teams are at ranking levels that historically always have gotten at large selections (if not Automatic Qualifiers), -- the Top 26 in the NCAA RPI rankings. Thus the at large bubble teams are #27 through #57.

Following up on my promise last week to highlight some data from the workbook each week, here is a table showing some of the data on the linked workbook's Conferences page:

The key columns in this table are the two salmon highlighted ones on the right. The first of those columns is for the NCAA RPI. As I have discussed extensively elsewhere, a major problem with the NCAA RPI is that teams' overall RPI ranks can be very different than their ranks as contributors to the strength of schedule portion of their opponents' RPI formulas. In the table, I've arranged the conferences in order of those whose teams' opponents have their strength of schedule contributions the most underrated at the top to those whose teams' opponents' strengths of schedule contributions are the most overrated at the bottom. As you can see, the Big 10's opponents (which includes both conference and non-conference opponents) are underrated as strength of schedule contributors by 32.7 rank positions. At the other end of the spectrum, SWAC's opponents' strengths of schedule contributions are the most overrated, by 32.8 rank positions.

The column on the right is for the Balanced RPI, showing that its overall ranks of teams and its ranks of them as strength of schedule contributors are essentially the same. That is not surprising since I designed the Balanced RPI with the purpose of having these two ranks the same.

The yellow highlighted column on the left shows the conferences' ranks when using the NCAA RPI. The green highlighted column shows the ranks after fixing the NCAA RPI's strength of schedule problem by using the Balanced RPI.

Although the details of the numbers will vary from year to year, the table paints a good general picture of the NCAA RPI's discriminatory patterns due to how it calculates strength of schedule.

SIMULATED NCAA TOURNAMENT BRACKETS USING THE NCAA RPI AND USING THE BALANCED RPI

Following the two brackets is a table showing how the two compare. (NOTE: For conference tournaments, I have used simulated conference tournaments based on the NCAA RPI.)

Here is a key to the tables:

The first table shows the projected NCAA Tournament bracket based on the Committee's using the NCAA RPI:

The next table shows the projected bracket based on the Committee's using the Balanced RPI:

NOTE: My weekly Comparison tables are preliminary and evolve throughout the season, so it is not the details that are important but rather the overall impression in terms of the difference between using the NCAA RPI as compared to the Balanced RPI.

This week, I want to draw your attention to the Big 10. The comparison indicates that using the NCAA RPI, the current prediction is that the Big 10 will have 5 teams in the NCAA Tournament. But if the Committee were using the Balanced RPI, it would have 10. That is a huge difference. Below the table are some thoughts about the difference.

So, why would the Big 10 have only 5 teams in the bracket if the Committee uses the NCAA RPI but 10 if it uses the Balanced RPI?

One major reason shows up in the table in the first part of this report, showing the difference between conference teams' opponents' NCAA RPI ranks and their ranks as strength of schedule contributors. As the table shows, the Big 10's NCAA RPI rank to strength of schedule contributor rank difference is -32.7 positions. Looking at the conferences whose teams are bumped out of at large positions using the Balanced RPI, they are the Big East with a difference of -14.4 positions, the Ivy at -13.7, the American at -9.7 positions, the Pac 12 at -4.7 positions, and the Sun Belt at +0.5 positions. In other words, the Balanced RPI, by eliminating the NCAA RPI's strength of schedule contributor discrimination problem, has moved up Big 10 teams' ranks relative to the ranks of teams from those other conferences.

I believe another reason is the Big 10 playing a 12-game in-conference season:

The effective weights of the three components of the NCAA RPI are 50% Winning Percentage, 40% Opponents Winning Percentage, and 10% Opponents Opponents Winning Percentage. Thus your opponents' winning percentages are very important (and against whom they achieved those winning percentages is not very important).
If teams were to play only games within their conferences, and assuming each conference played a full round robin, each team's opponents' winning percentage would be 0.5000. This is because if I play conference opponents A and B, they will have played each other with one winning and the other losing or both having ties.
The way conferences are distinguished from each other thus is not through their in-conference games but rather through their non-conference games -- and most important, by their winning percentages in non-conference games since these winning percentages are by far the primary determiner of the conference teams' values to their conference opponents as strength of schedule contributors.

The following table shows teams' non-conference winning percentages, using the NCAA RPI's Winning Percentage formula:

Next, consider this: The Big 10 plays a 12-game in-conference schedule. The only other conference to play that many in-conference games is the Metro Atlantic. Remember, from a strength of schedule perspective, each conference's teams strength of schedule from in-conference play is roughly 0.5000 (but not exactly if the conference doesn't play a full in-conference round robin). It is the non-conference winning percentages that pull the teams' values as strength of schedule contributors above 0.5000.
The Big 10, by playing a 12-game conference schedule, has reduced its capacity to play non-conference games relative to the capacity of teams from other conferences seeking NCAA Tournament at large positions. This gives the Big 10 teams fewer non-conference games through which to pull its overall opponents' winning percentages above 0.5000.
The Big 10 could compensate for this by scheduling only very easy non-conference opponents so that its teams' non-conference winning percentages would be higher than those of all the other conferences that play fewer in-conference games. This, however, would mean foregoing opportunities for "signature" non-conference results, which also are important in the NCAA Tournament seeding and at large selection process. In other words, the Big 10, by playing a 12-game schedule, has put itself in a disadvantaged position relative to other conferences when it comes to formation of the NCAA Tournament bracket, with no good way out of that position.

Tuesday, September 29, 2026

2026 ARTICLE 15: ACTUAL CURRENT RANKINGS AND OTHER DATA AFTER WEEK 7

This Article 15 provides current NCAA RPI, KPI, Balanced RPI, and Massey rankings based on the actual results of games played through Week 7 (through September 27) of the season, for teams, conferences, and regions.

Here is a link to an Excel workbook that contains all the information:

2026 RPI Report Actual Results Only as of 9.27.2026

To download the workbook, use the following steps:

1. Click on the workbook link.
2. The link takes you to a Google spreadsheet, which you will not want to use. On the Google spreadsheet, to the left, click on File, which will bring up a drop down menu.
3. In the menu, click on Download, which will bring up another drop down menu.
4. In the drop down menu, click on Microsoft Excel (.xlsx). This will download the Excel workbook.

The first page of the workbook is for Teams, the second for Conferences, and the third for Regions. The data so far are based mostly on non-conference games, as we're now early in the conference season. I consider these data very preliminary and expect significant changes over the remainder of the season, particularly in reation to how conferences and regions compare to each other.

As with Article 14 on the full season simulated rankings, starting this week, each week I'll highlight some data from the "actual games only" workbook that may be helpful to those interested in how the NCAA RPI works. For this week, I'll cover some data related to regions.

As indicated in Article 13, I have organized the conferences into regions based on their historic non-conference scheduling patterns:

Autonomy (i.e., Power 4): ACC, Big Ten, Big Twelve, SEC

Middle: Horizon, MidAmerican, Missouri Valley, Ohio Valley, Summit
MidEast: Atlantic Ten, Big East, Colonial (now Coastal), Ivy
North: America East, Metro Atlantic, Northeast, Patriot
South: American, Southland, Southwestern, Sun Belt
Southeast: Atlantic Sun, Big South, Conference USA, Southern, United
West: Big Sky, Big West, Mountain West, Pac Twelve, West Coast

The following table is from the linked workbook's Regions page:

As you can see, to date teams from the Mideast region, in games against other teams from the region, have played a very high proportion of ties. On the other hand, teams from the North have played a very low number. In relation to the 2024 change in the NCAA RPI's Winning Percentage formula, going from treating a tie as half a win to a third of a win, this means the change has affected teams from the Mideast region the most negatively and teams from the North region the most positively. If you compare this table to the similar table for the full season simulation in Article 14, you will see that the simulation projects that the spread in the above table will narrow. That will be something to watch for to see if it does.

2026 ARTICLE 14: SIMULATED END-OF-SEASON RANKINGS AND NCAA TOURNAMENT BRACKETS FOLLOWING WEEK 7

This article is based on the actual results of games played through Week 7 (ending September 27) and simulated results of games not yet played, including conference tournament games. It projects where teams will end up at the end of the season, but before the NCAA Tournament.

Here is a link to an Excel workbook with the simulated full-season NCAA RPI and Balanced RPI ranks -- and other important information -- for teams, conferences, and regions.

RPI Report as of 9.27.2026

For purposes of evaluating the usefulness of the ratings the simulation currently uses to predict the results of games not yet played:

The table's reference to Team 1 is to the home team or, if a neutral site game, the team whose name comes first alphabetically.

The first data row in the table, 2026 Team 1 Actual Results, shows Team 1's actual results. The second data row, 2026 Team 1 Predicted Results, applies teams' simulated end-of-season RPI ratings to games actually played through September 27, to determine what Team 1's win, loss, and tie percentages should be if those ratings are correct. The third data row then shows the difference between those predicted percentages and the actual percentages. Thus, for example, the actual winning percentage for all Team 1s was 48.8%. According to teams' end-of-season simulated ratings, the winning percentage should have been 50.0%. The Actual Less Predicted Results row shows the difference between predicted and actual was -1.187%.

The last three data rows are similar, but instead of using teams' simulated end-of-season RPI ratings, it uses teams' actual current NCAA RPI ratings to determine what the percentages should be if those ratings are correct.

As you can see, the simulation's current ratings are a better match with actual results than are the actual current NCAA RPI ratings. Because of this, for now I will continue using the simulation's ratings to predict future game results rather than converting to using the NCAA RPI's current ratings. In the upcoming weeks, I will continue doing this until the NCAA RPI's current ratings perform better than the simulation's current ratings.

Starting this week, each week I'll highlight here some data from the workbook that may be helpful to those interested in how the NCAA RPI works. For this week, I'll cover some data related to regions.

I assign conferences to regions based on their scheduling patterns, i.e., Against which other conferences, over the years, have their teams tended to play their non-conference games? This has led me to organize the conferences into the following regions:

Autonomy (i.e., Power 4): ACC, Big Ten, Big Twelve, SEC

Middle: Horizon, MidAmerican, Missouri Valley, Ohio Valley, Summit
MidEast: Atlantic Ten, Big East, Colonial (now Coastal), Ivy
North: America East, Metro Atlantic, Northeast, Patriot
South: American, Southland, Southwestern, Sun Belt
Southeast: Atlantic Sun, Big South, Conference USA, Southern, United
West: Big Sky, Big West, Mountain West, Pac Twelve, West Coast

The following table is from the linked workbook's Regions page:

This table shows the proportion of games each region's teams play against opponents from their own region and also against teams from each other region. Among other things, this allows you to see how much "linkage" there is between regions. Looking at the West region, as an extreme example, its teams play 85.1% of their games "in region" and 7.7% of their games against Autonomy teams, for a total of 92.8% of their games within those 2 regions. That leaves 7.2% of their games spread among teams from the other regions. That means each of the West teams on average plays a little over 1 game per year against opponents from all the other non-Autonomy regions combined. In other words, the West region plays almost in isolation from the other regions except for the Autonomy region.

The next table shows the proportions of in-region ties:

Prior to 2024, the NCAA RPI formula for a team's Winning Percentage counted a tie as half a win. In 2024, the Women's Soccer Committee changed this to a third of a win, thus reducing the value of a tie. This table shows, for this year, who is likely to be hurt by and who benefits from this change. As you can see, the MidEast teams are the most likely to be hurt and the North teams are most likely to benefit.

The next table is about the NCAA RPI's basic defect, from a regional perspective:

As I have shown elsewhere, how the NCAA RPI evaluates a team overall is different than how it evaluates that team when it comes to that team's contribution to an opponent's strength of schedule. This leads to teams' NCAA RPI ranks being different than their NCAA RPI ranks as strength of schedule contributors. In the above table, the first highlighted column shows the region's teams' NCAA RPI ranks less their NCAA RPI ranks as strength of schedule contributors. A negative number means the NCAA RPI strength of schedule contributor rank is poorer than the overall NCAA RPI rank. A postive number means it is better.

Using the Autonomy region as an example, its teams' NCAA RPI strength of schedule contributor ranks on average are 29.4 positions poorer than their actual NCAA RPI ranks. In other words, if you play an Autonomy opponent, on average your own RPI is going to get credit for playing a signifcantly weaker opponent than the NCAA RPI itself says you should be getting. Since Autonomy teams mostly play other Autonomy teams, this means that the NCAA formula on average undervalues those teams' strengths of schedule.

The table thus shows which regions benefit from and which are hurt by this NCAA formula defect.

The purpose of the Balanced RPI is to fix this NCAA RPI formula defect. As the third column in the table shows, for all regions, there is almost no difference between how the Balanced RPI ranks regions' teams overall and how it ranks them as strength of schedule contributors. Thus the Balanced RPI data in the workbook's Regions page show what happens when you eliminate the NCAA RPI's defect. The following table shows what happens, from a regions perspective:

As you can see, Autonomy teams' average ranks suffer the most from the NCAA RPI defect, averaging 24 positions poorer than they should be as indicated by the Balanced RPI. Next come teams from the West, averaging 18 positions poorer; and then the MidEast averaging 3 positions poorer. All the other regions' teams, on average, are ranked better than they should be.

SIMULATED NCAA TOURNAMENT BRACKETS USING THE NCAA RPI AND USING THE BALANCED RPI

Here are currently simulated NCAA Tournament brackets, first based on the NCAA RPI and then based on the Balanced RPI. Following those is a table showing how the two compare. (NOTE: For conference tournaments, I have used simulated conference tournaments based on the NCAA RPI.) The information above, particularly the table just above, provides context for understanding what you will see in the comparison table.

Here is a key to the tables:

The first table shows the projected NCAA Tournament bracket based on the Committee's using the NCAA RPI:

The next table shows the projected bracket based on the Committee's using the Balanced RPI:

NOTE: My weekly Comparison tables are preliminary and will evolve throughout the season, so it is not the details that are important but rather the overall impression in terms of the difference between using the NCAA RPI as compared to the Balanced RPI. In relation to the regions data in the first part of this report, the comparison shows the effects of the NCAA RPI's defect in relation to strength of schedule and what happens if you eliminate the defect.

Wednesday, September 23, 2026

2026 ARTICLE 13: ACTUAL CURRENT RANKINGS AND OTHER DATA AFTER WEEK 6

This Article 13 provides current NCAA RPI, KPI, Balanced RPI, and Massey rankings based on the actual results of games played through Week 6 (through September 20) of the season, for teams, conferences, and regions.

Here is a link to an Excel workbook that contains all the information:

2026 RPI Report Actual Results Only as of 9.20.2026

To download the workbook, use the following steps:

1. Click on the workbook link.
2. The link takes you to a Google spreadsheet, which you will not want to use. On the Google spreadsheet, to the left, click on File, which will bring up a drop down menu.
3. In the menu, click on Download, which will bring up another drop down menu.
4. In the drop down menu, click on Microsoft Excel (.xlsx). This will download the Excel workbook.

The first page of the workbook is for Teams, the second for Conferences, and the third for Regions. The data so far are based almost entirely on non-conference games. A way to think about the data at this point in the season is that they are most reliable in showing how the regions stack up against each other, are reasonably reliable in showing how the conferences stack up, and are not very reliable in showing how the individual teams stack up. How the individual teams stack up will be sorted out during conference play.

On the Teams page, on the left, are color coded columns. These columns are based on past history. They show which teams, based on their current ratings, are potential seeds (by seed level) and at large selections for the NCAA Tournament and which already appear assured of getting at least at large selections. As the color coding shows, teams with NCAA RPI ranks of #65 or better are potential #1 or #2 seeds. Teams ranked #192 or better are potential candidates for all other seeds and for at large selections. Only the top 5 teams appear assured of at large selections, which means the current at large bubble is teams ranked #6 through #192. These numbers show how unreliable the current rankings are, so far as how the NCAA Tournament bracket will end up is concerned.

Mostly for illustration purposes, here are the current NCAA RPI Top 65 teams -- i.e., the potential #1 and #2 NCAA Tournament seeds, with some comments below:

Comment 1: You may notice that some of the Balanced RPI ranks for teams are significantly different than the other systems' ranks. The NCAA RPI formula, at this point in the season, weights teams' winning percentages at 42.9% and their opponents' strength of schedule at 57.1%. This is different than what its weights will be at the end of the season, which will be approximately 50.0% and 50.0%. The Balanced RPI, on the other hand, currently weights teams' winning percentages at 50.0% and their strength of schedule at 50.0% and will do so throughout the season. (The systems calculate strength of schedule differently.) Most of the significant differences are accounted for by the NCAA RPI's current de-emphasis on teams' winning perentages.

Comment 2: Massey's rating formulas take into consideration teams' ratings over a number of past years. As the season progresses, the weight of prior years' ratings decreases. This accounts for some of the differences between his ratings and the other systems' ratings.

Comment 3: Although there will be significant differences among the rating systems at the end of the season, particularly between the NCAA RPI and KPI systems on the one hand and the Balanced RPI and Massey on the other, the differences are likely to be significantly narrower than what you see today.

Actual Less Predicted Winning Percentages

This year, I have added data for each of teams, conferences, and regions, showing their actual winning percentages as compared to their expected winning percentages based on their current NCAA RPI ratings. For Teams, these data are in the far right columns.

Using currently #1 rated Michigan State as an example, its Actual Winning Percentage so far is 74.1%. Based on its current NCAA RPI rating and the current ratings of its opponents, however, its Expected Winning Percentage Based on NCAA RPI Rating is 80.8%. The difference is -6.7%.

Compare this to North Carolina. Its Actual Winning Percentage is 90.0%, its Expected Winning Percentage Based on its NCAA RPI Rating is 75.9%, and its difference is 14.1%.

These numbers suggest that currently, the NCAA RPI is overrating Michigan State and underrating North Carolina.

For Conferences, the Actual less Expected numbers are in column H and how conferences rank in terms of this difference is in column I. For example, using the ACC as an example, the difference between its teams' Actual and Expected winning percentages is 4.5%, meaning the conference overall is underrated. And, it is the 6th most underrated conference.

For Regions, the numbers likewise are in columns H and I. Using the Autonomy region (all Power 4 teams) as an example, the difference between its teams' Actual and Expected winning percentages is 7.7% and it is the most underrated region.

Percentage of In-Region Ties

Of particular interest this week, column Q on the Regions page shows the proportion of each region's in-region games that are ties. This likely is a measure of in-region parity. Most notably, the MidEast region has had 30.2% of its in-region games to date end as ties. That is extremely high, especially in relation to an historic norm around 21%. This is significant because the NCAA in 2024 changed its RPI Winning Percentage formula to treat ties as only 1/3 of a win rather than the 1/2 of a win it had used in previous years. As the percentage of in-region ties numbers show, this is having a disproportionate negative impact on MidEast region teams this year.

The MidEast region consists of teams from the Atlantic Ten, Big East, and Colonial (now Coastal) conferences and the Ivy League.

2026 ARTICLE 12: SIMULATED END-OF-SEASON RANKINGS AND NCAA TOURNAMENT BRACKETS FOLLOWING WEEK 6

This week I begin posting two articles a week. This Article 12 is based on my simulated end-of-season rankings. Article 13 is based only on teams' actual games played through Week 6 and includes current NCAA RPI, Balanced RPI, KPI, and Massey ranks and other information.

------------

Here is a link to an Excel workbook with simulated full-season NCAA RPI and Balanced RPI ranks -- and other important information -- for teams, conferences, and regions. The ranks are based on the actual results of games played through Week 6 (ending Sunday, September 20) and simulated results of games not yet played.

RPI Report as of 9.20.2026

For purposes of evaluating the usefulness of the ratings the simulation currently uses to predict the results of games not yet played:

The table's reference to Team 1 is to the home team or, if a neutral site game, the team whose name comes first alphabetically.

The second data row in the table, 2026 Team 1 Predicted Results, applies teams' simulated end-of-season RPI ratings to games actually played through September 20, to determine what Team 1's win, loss, and tie percentages should be if those ratings are correct. The third data row then shows the difference between those predicted percentages and the actual percentages. Thus, for example, the actual winning percentage for all Team 1s was 49.1%. According to teams' end-of-season simulated ratings, the winning percentage should have been 50.3%. The Actual Less Predicted Results row shows the difference between predicted and actual was -1.272%.

As you can see, the simulation's current ratings are a better match with actual results than are the actual current NCAA RPI ratings. Because of this, for now I will continue using the simulation's ratings to predict future game results rather than converting to use of the NCAA RPI's current ratings. In the upcoming weeks, I will continue doing this until the NCAA RPI's current ratings perform better than the simulation's current ratings.

SIMULATED NCAA TOURNAMENT BRACKETS USING THE NCAA RPI AND USING THE BALANCED RPI

Here is a key to the tables:

The first table shows the projected NCAA Tournament bracket based on the Committee's using the NCAA RPI:

The next table shows the projected bracket based on the Committee's using the Balanced RPI:

The final table compares the projected brackets. It shows the teams in region order and in conference order so you can see how a change to the Balanced RPI would affect regions' and conferences' teams. Salmon highlighting means the team gets a poorer treatment by the rating system and green means the team gets better treatment. No highlighting means the treatment is the same.

NOTE: The Comparison table is preliminary, so it is not the details that are important but rather the overall impression in terms of the difference between using the NCAA RPI as compared to the Balanced RPI. If you review carefully the data in the above linked RPI report, you will see that the NCAA RPI has discriminatory patterns in relation to regions and conferences, whereas the Balanced RPI eliminates these patterns. Thus the table below shows the effect of eliminating the NCAA RPI's discriminatory patterns.