Here is my 2024 Report 14 with current actual RPI ratings, ranks, and other information, and teams currently in the candidate pools for NCAA Tournament seeds and at large selections, based on games played through Sunday, November 3. It also has predicted end-of-season ratings, ranks, and other information. And, it has predicted NCAA Tournament #1 through 8 seeds and at large selections. The predictions are based on the actual results of games played through November 3 and predicted results of games not yet played. The predicted results are based on teams' current NCAA RPI ratings from games played through November 3. At the end, the report also shows new teams predicted to get at large positions if the Committee were using the Balanced RPI
Here is my 2024 Report 15, with final NCAA Tournament seed and at large selection predictions if the Committee follows its historic patterns. It includes the last teams "in" and "out" for at large positions. I will not be surprised if the Committee does not give at large positions to a couple of the listed "in" teams and instead gives them to a couple of "out" teams.
Looks like seeding has some discrepancies at the top between your projections and Chris Henderson’s. Are your projections an effort to balance things from travel/geography perspective?
I do not understand how Liberty or Boise would ever get in over Cal, when Cal beat Stanford, Pitt, and tied Wake Forest. Liberty and Boise have no wins even close to that, and have worse losses. Cal also has a better RPI than Boise and nearly the same as Liberty.
I discovered some programming errors related to the NCAA RPI bonus and penalty adjustments for good and poor results. The linked report now has corrected predictions of what the Committee is likely to do if it follows its historic patterns.
If I'm Xavier and I am sitting at RPI 35 watching Colorado get into the tournament instead of me with an RPI 55, I am pretty dang mad.
I think this tweet sums it up well though. Xavier had 0 top 50 wins: Perfect match. Who'da thunk it? pic.twitter.com/A5H3WcWKJG— Chris Henderson (@chris_awk) November 11, 2024
As I look over the NCAA Tournament bracket and the Committee's historic decision patterns in relation to the NCAA criteria, I see very little to criticize about the Committee's decisions. Contrary to claims about bias, the main thing looking a little unusual to me is Ohio State getting a #3 seed. Past history suggests Ohio State would have been around a #6 seed. Western Michigan not getting a seed at all also is surprising, but this could be a function of the Committee having figured out that the MidAmerican Conference, whether intentionally or not, appears to have "gamed" the RPI into overrating its teams. For the at larges, although my program suggested Liberty, Pittsburgh, Boise State, and Xavier would get at large positions, they all had demerits making them borderline and Cal, Washington, Colorado, and LSU all were just at the border of being in. So the first four being out and the last four being in is reasonable from an historic perspective. In terms of where the seeded teams are placed in the bracket, my understanding is that the Committee "sub-seeds" the four #1s, meaning the Committee's order likely was Duke, Mississippi State, Southern California, Florida State. One might want to remember that the Committee looks at the entire season when making its decisions. The positioning is arguable, but to me is within a reasonable range of what the Committee could do. Once they get past the four #1 seeds, the Committee may or may not "sub-seed" the other seed levels. If it doesn't, the NCAA travel cost limitation program takes over and places teams in the bracket so as to minimize travel cost, subject to the limitation that two teams from the same conference cannot meet in the first two rounds (if it is possible to avoid it). Other than the first two round same conference avoidance rule, the NCAA policy seems to be that the balance of placements within the bracket will not take conference membership into consideration.
It shows Xavier had 0 top 50 wins. What is to get mad about? Every at large team that got in had top 50 wins.
Indeed Proud of this team (@XavierWSOC), a special group and one we’ll remember. 1 of only 12 in the nation (349 teams) with 15 or more wins. 10 regular season road games (8-1-1) - Tied for 2nd most road wins in the country. #29 Non-conference RPI (7–1-1). You deserved better.— Brian Hicks (@XUmktgguy) November 11, 2024 Feels a bit arbitrary, no? Why Top 50 - not 49, not 51, 50 - wins and not something else...like, say, RPI, which the NCAA literally published on its website as a ranking of teams? Maybe I'm just extra cognizant of the challenges that getting non-con matches against the P4 schools entails for the likes of Xavier.
Not arbitrary at all. The selection committee is instructed to prioritize this metric of Top50 results. Now, is it heavily skewed toward ACC teams, and does it judge the midmajor conferences harshly? Absolutely. Xavier's program has just come off a 2OT loss in last year's NCAA and should have gotten more respect than to be sent home with a 35RPI, IMO. We agree that the Top50 cutoff is bogus and punitive because it treats teams like UMass, whose marquee results are including a split with #53 Dayton and a ROAD win over #66 BC, as not qualifying. At least in MBB, these results would have been considered "Quad 1" wins because MBB's cutoff is #75, not #50. But in soccer, where there are just about as many teams across the board, those top results don't even get onto the first page of a team's qualifying resume. That's Grossly disproportionate. And it's never going to change as long as the ACC is owning the tournament and calling the shots. For their sake, they cannot let their NINTH team not make it in because heaven forbid either of the Big East (2-bid league), A10 (perennial 1-bid) or in some years, the Ivy instead of A10, would have another well-qualified bubble team make its bid.
This is woefully uninformed. The ACC is actually underrated by the RPI, not over rated. The SEC got more teams in than the ACC, so if you have a problem with “undeserving” teams from P4 conferences you can start there. Also it’s funny you use an ACC team as your example of a “good” UMass win. You know who else beat BC? A few ACC teams that didn’t make it into the tournament either.
Using UMass as an example, the reason Top 50 results matter is it is those results that show the level at which you are able to win games against teams likely to be in the Tournament. UMass played 3 Top 50 opponents, with one good a result -- a win against #33 Rutgers. That's a decent win, but all of the borderline teams that got in ahead of UMass had better wins: Cal v #13, Washington v #22, LSU v #23, Tennessee v #24, and Colorado v #27. Each of those teams also had additional good Top 50 results, which UMass did not. Of the other borderline teams that did not get in, none of them had wins better than those five teams. So, from that perspective, the borderline teams that got in ahead of UMass showed they are able to win games at a higher level of the Tournament field than UMass showed it is able to win. Those other teams did play more top 50 games than UMass, so it is true they had more opportunities to get good results against top teams than UMass had. The answer to that, for UMass, is to schedule more non-conference opponents that are likely to be highly rated. Although years ago, it was difficult for teams from mid-majors to schedule that way, these days if you are a mid-major that is expected to have a good overall record, you are a desirable opponent for teams likely to be in the NCAA Tournament to play. This is because it is in their RPI interest to play non-conference opponents that will have good winning records, especially ones from higher tier mid-major conferences. See Pepperdine's and Santa Clara's schedules as examples. On average, unseeded teams getting at large selections play about 8 Top 50 opponents. So, subtract the number of teams in the conference expected to be Top 50 and that gives you the number to try to schedule. And remember, you can't say, "Oh, if we schedule that many we'll get killed." If that's how you think, the implication is you're not NCAA Tournament material. The other factor affecting a team like UMass, however, is travel expenses. That definitely is a possible limiting factor for their NCAA Tournament prospects. It simply may be too expensive to schedule properly in relation to getting an NCAA Tournament at large position. There's no easy answer to that problem.
I'm sympathetic to the argument that the new super conferences have broken Top 50 wins as a criterion that any school can access just by deciding to schedule a hard non-con (and then winning, of course). See thread: Seems like there's a lot of misinterpretation willful or not about why LSU and Colorado got in the field yesterday.There seems to be this argument dictated towards the snubbed of "you have to play somebody!" in non-conference play. But...— Chris Henderson (@chris_awk) November 13, 2024
Georgetown actually is a really good example of how mid-major teams should schedule if they aspire to NCAA Tournament at large selections. Dave Nolan is about as knowledgeable and sanguine a mid-major coach as there is when it comes to this. As I mentioned in my previous post, Pepperdine and Santa Clara likewise are good examples. It in part will come down to travel expenses. If you consider Pepperdine's and Santa Clara's West Coast Conference, it is spread up and down the entire West Coast. Even for in-conference play, there are significant travel expenses. Then you add to that out-of-region travel to play likely highly ranked opponents and it gets even more expensive. The bottom line for mid-major schools may be that if you want to be in the NCAA Tournament you probably will need to be willing to have significant travel expenses and, as RusskyHoya says, you are going to have to win a good number of those travel games.
Yeah, Dave is quite adept at leveraging the advantages he has. While the university endowment has been a decades-long catch-up game relative to USNWR Top 25 peers, Georgetown is obviously better resourced then the vast majority of non-P4 programs, and so it can afford to fly cross-country to play at Pepperdine (2023) or Texas (2021) or Stanford (2017) if it wants to. DC is also a highly attractive destination for visiting teams for a number of reasons, including that it's centrally located along the Acela/I-95 corridor and makes for a memorable early-season road trip for a team. Most non-P4s, including those in the area like GW, AU, and GMU (Howard and Navy are obviously special cases), don't have that same suite of advantages. I should add that the prospect of having to play day games in August or September has definitely depressed demand somewhat. Fortunately, I learned earlier tonight that the university has finally filed its plans to add lights to Shaw Field, now that the new hospital helipad is active and there are no more Medevac aerial conflicts. I very much look forward to evening games finally happening during the warm weather months!
The following table may be of interest related to the discussion about scheduling and the NCAA Tournament. It is based on data from 2007 through 2023 (excluding Covid-affected 2020). The "Average Number of ..." columns refer to the average number of opponents in the column category.
Here are some numbers on how many non-Power conference teams have been in the NCAA Tournament over the last 10 tournaments (excluding the 2020 shortened field tournament). Throughout these years excet this year, there were 31 conferences with Automatic Qualifiers. Where at large teams were in not-Power conferences earlier but this year are in Power conferences (BYU, SMU, and UCF), I have not included them in the numbers. The first number is non-Power at larges and the second is both non-Power at larges and non-Power Automatic Qualifiers. 2024: 2 at large/28 teams 2023: 8/34 2022: 4/30 2021: 8/34 2019: 4/30 2018: 2/28 2017: 3/29 2016: 3/29 2015: 10/36 2014: 5/31 Question: Is 2024 simply a typical random variation or is it something more significant?
Is there an obvious difference in the consideration of "RPI Top-50 wins" this year versus prior years? Chris Henderson mentions that in a post on X (@chris_awk) as the reason "bubble" teams LSU and Colorado got in this year.