I've wondered whether I have been too critical of the Committee's giving an at large selection to Rice rather than Minnesota or Cincinnati. Based on the Committee's past patterns, Rice wouldn't have gotten an at large selection, whereas Minnesota and Cincinnati were possibilities for a selection. There's nothing, however, that says the Committee has to follow past patterns, although in most cases it's probably fair to expect them to have at least a reasonable degree of consistency over time. So, I went through a different kind of analysis to test the Committee's decisions. This analysis is based on ranking the teams simply as among the three of them, in relation to each of the main factors the Committee considers. For each factor, I've indicated how I rate the three of them. You can do your own rankings. ARPI: 1. Rice #36 2. Minnesota #48 3. Cincinnati #52 ANCRPI: 1. Cincinnati #51 2. Minnesota #68 3. Rice #69 Note: It might be reasonable to treat Minnesota and Rice as a tie, so 2.5 position for each of them. Results Against Top Teams (best results): 1. Cincinnati T v #5, W v #15 2. Minnesota W v #11, T neutral v #14 3. Rice T v #34 Head to Head Results: None Common Opponent Results: 1. Rice and 2. Cincinnati: Houston: Rice W v Houston; Cincinnati L @ Houston Memphis: Rice L @ Memphis; Cincinnati T @ Memphis Minnesota had no common opponents with Rice or Cincinnati Note: I don't particularly like doing common opponent results this way, as it ignores too much information, but it may be how the Committee does it. Poor Results: 1. Rice L neutral v #82 2. Minnesota L v #136 3. Cincinnati L v #115, T v #184 Standing within Conference/Conference Rank Regular Season Standing/Conference Rank/Conference Average ARPI Note: All three lost in the quarterfinals of their conference tournaments, so that's why I'm using only their regular season standing. 1. Minnesota #2.5/4/0.5679 2. Cincinnati #3/6/0.5491 3. Rice #1/12/0.5017 Note: I included the conferences' average ratings for a reason. They show that the differences' in the conferences' strengths are large, in fact vary large especially for Rice. When I look at all of these, the main question I ask is how much weight to assign to the different factors. For example, are poor results as important as good results? How important is the ANCRPI? How important are common opponents results, especially when they are as ambiguous as those for Rice and Cincinnati? How important is conference standing and strength? How does the #1 team in the #12 conference compare to the tied for second team in the #4 conference? These all are questions the Committee members have to ask themselves, and the members are not required to assign equal weights to all the factors. For me, in the comparisons of these three teams I think the good results are especially important, and more important than the poor results. I think the good results show the level at which a team could compete in the NCAA Tournament: Cincinnati (tie with #5) has shown it has the potential to compete successfully at the quarter-final level; Minnesota (win over #11) has shown it has the potential to compete successfully at the third round level; and Rice (tie with #34) has shown it might have the potential to compete successfully in only the first round. Yes, Cincinnati and Minnesota have some poorer results than Rice, but for the NCAA Tournament, I'm more concerned with not excluding teams that have demonstrated they can be significant factors in the Tournament. On the other hand, looking at it in terms of the above comparisons, the Committee's selection of Rice doesn't seem as poor a decision as I previously had thought. But, I still think it was a poor one.
The NCAA today published its Final RPI rankings for the season, which include the NCAA Tournament games as part of the data base. Unfortunately, there are two errors in the NCAA's data base that make the rankings not right: They have the NCAA Tournament game of Northwestern v Butler listed as a Northwestern win, whereas it was a tie with Northwestern advancing on PKs. This error is causing a number of the rankings in the upper end of the rankings to be off. Then have the SWAC Tournament game of Alabama State v Grambling listed as a Grambling win, whereas it was a tie with Alabama State advancing on PKs. This error has no effect on the upper end of the rankings. The NCAA staff now is aware of these errors and hopefully will correct the data errors and substitute the correct rankings for the ones currently available on-line. In the meantime, here are the correct ratings/rankings for the top 100 teams: Stanford 0.7336 1 Duke 0.6964 2 SouthCarolinaU 0.6917 3 NorthCarolinaU 0.6904 4 UCLA 0.6873 5 Princeton 0.6697 6 TexasA&M 0.6640 7 PennState 0.6563 8 WestVirginiaU 0.6555 9 UCF 0.6486 10 Rutgers 0.6424 11 FloridaU 0.6366 12 FloridaState 0.6358 13 TexasU 0.6334 14 SouthernCalifornia 0.6276 15 TennesseeU 0.6271 16 NotreDame 0.6267 17 VirginiaU 0.6239 18 SouthFlorida 0.6224 19 Pepperdine 0.6220 20 SantaClara 0.6202 21 Auburn 0.6156 22 NCState 0.6138 23 OhioState 0.6129 24 Georgetown 0.6124 25 Baylor 0.6116 26 MurrayState 0.6089 27 ArizonaU 0.6074 28 WisconsinU 0.6071 29 CaliforniaU 0.6045 30 Hofstra 0.6035 31 Clemson 0.6030 32 Vanderbilt 0.5955 33 AlabamaU 0.5952 34 NorthwesternU 0.5951 35 OklahomaState 0.5929 36 Monmouth 0.5889 37 FloridaGulfCoast 0.5884 38 WashingtonState 0.5882 39 Rice 0.5837 40 WakeForest 0.5834 41 ColoradoU 0.5833 42 Butler 0.5801 43 MississippiState 0.5775 44 ArkansasU 0.5774 45 MississippiU 0.5771 46 MinnesotaU 0.5739 47 LaSalle 0.5730 48 Marquette 0.5702 49 BostonCollege 0.5688 50 Memphis 0.5673 51 SanJoseState 0.5666 52 TCU 0.5665 53 Cincinnati 0.5662 54 VirginiaTech 0.5662 55 Northeastern 0.5655 56 WashingtonU 0.5650 57 StLouis 0.5649 58 LSU 0.5625 59 Drexel 0.5617 60 Yale 0.5592 61 SouthAlabama 0.5590 62 Navy 0.5579 63 Purdue 0.5572 64 MichiganU 0.5569 65 Milwaukee 0.5567 66 SMU 0.5559 67 Drake 0.5558 68 NorthTexas 0.5541 69 Lamar 0.5538 70 LouisianaTech 0.5526 71 LoyolaMarymount 0.5521 72 LongBeachState 0.5491 73 Louisville 0.5487 74 Columbia 0.5486 75 NebraskaU 0.5486 76 SanDiegoState 0.5485 77 UCIrvine 0.5478 78 TexasTech 0.5476 79 StJosephs 0.5474 80 Bucknell 0.5444 81 EasternWashington 0.5385 82 Brown 0.5385 83 CalStateNorthridge 0.5373 84 IndianaU 0.5364 85 Providence 0.5356 86 MarylandU 0.5353 87 UNI 0.5349 88 EasternKentucky 0.5348 89 Charlotte 0.5345 90 VCU 0.5345 91 UNCGreensboro 0.5342 92 KansasU 0.5341 93 MissouriU 0.5338 94 Syracuse 0.5337 95 CalStateFullerton 0.5336 96 TennesseeTech 0.5334 97 HighPoint 0.5331 98 UAB 0.5321 99 Samford 0.5301 100
Just a note that the NCAA has corrected the two data errors I mentioned in the preceding post and revised the final rankings posted on-line accordingly. Kudos to the NCAA staff, they have a very difficult job and do it really well under the circumstances they have to deal with.