A big thanks to Keith for clarifying some things, such as how the crossover and re-seeding was done. Now some comments. * Rick Comley, as Keith relayed, said that the seedings were done largely by the computer. Note that the top 12 teams in RPICH (or RPI) all got in. CC was 13th, SCSU 14th. The difference between CC at 13th and MSU at 12th was .0019 (by RPICH). * Schedule strength killed CC. Look how they compared with the two WCHA teams that got in (OWP from Erik's RPICH): Minn .5236 Wisc .5136 CC .4721 How can this be since all three play in the same conference? Well, to start, CC played Air Force twice, Miami, and Kent as NC games. Minnesota played Michigan, MSU, Cornell, and LSSU. Wisconsin played Michigan, MSU, UAF, and NMU. Right there, CC had a weaker regular season NC schedule - but each team only had 4 NC games available, just 11% of their total regular season schedule. On top of that, because of the WCHA scheduling, CC only played Minnesota twice this year (WCHA teams play 7 teams 4x and 2 teams 2x). And on top of THAT, by finishing first, CC drew last place MTU and played them 3x. As Keith says, the fact that CC was upset forced them to lose out on the opportunity to play some better teams in the WCHA Final Five which would have boosted their schedule strength some - but the damage may have already been done. It certainly appears that the WCHA unbalanced schedule hurts teams, and also the rating works against teams which finish high in their conference but are upset early. On the one hand, it seems to make sense - you should have to win the games you should win - but on the other hand, it also appears that the combination of all the above factors dealing with CC's scheduling in 1993-94 did them in. I would be curious to see how the old way of calculating the RPI might have affected the ratings. As well, does the small number of NC games available to WCHA (and to a lesser extent, CCHA) teams hurt those teams? Now it is VERY important that you go out and schedule good NC opponents. It would be very bad if a team like CC were forced to drop their traditional rivalry with Air Force because of what happened this season. * The tv selection show was, as expected, mostly fluff. I would have liked to see more talk about why the seedings were done the way they were done, instead of a few seconds with every coach they could round up. I got a strong sense that Lowell's Bruce Crowder was not happy with his team being sent to East Lansing to face MSU even though Lowell was likely seeded several spots higher than MSU. Host Frank Mazzocco made a mistake by asking Bruce about Wisconsin, and Bruce had to correct him by reminding him Lowell was facing MSU. It happens, but Bruce's response seemed to indicate his unhappiness with the proceedings (IMO). Frank also erred significantly in saying that CC was the first regular season conference champ not to receive a bid. For starters, it happened to Harvard in 1992. It also happened to BGSU in 1983 when the tourney only consisted of 8 teams - maybe Frank meant since the expansion to 12 teams, but even if so, he was incorrect. An interesting point that Rick Comley brought up was the head-to-head record between teams. I believe Rick said that the coaches wanted each win counted as a point in a team's favor - i.e., UNH beat Lowell head to head 3-2-0, so instead of UNH getting one point in their favor, they'd get 3 and Lowell 2 (and for Minn-Wisc, Minn would get 4 and Wisc 0). But I couldn't tell if he was saying that it was done this way or that this was what was desired. * Keith said the seeds were originally done thusly by region: EAST BU, Harv, UNH, UML, RPI, NU WEST Mich, Minn, LSSU, Wisc, WMU, MSU I agree with all of these except UNH ahead of UML. Looking at the 5 factors: UNH UML EDGE TO RPICH .5543 .5491 UNH Head-Head 3-2-0 2-3-0 UNH TUC Win% 9-9-2 .500 7-7-3 .500 EVEN Last 20 Win% 8-10-2 .450 11-4-5 .675 UML CommOpp Win% 16-10-2 .603 15-6-6 .667 UML (TUC is teams under consideration, which I considered to be all of the teams that made the conf final 4/5/6 + CC & Maine for a total of 21 teams. It may have been slightly different as some teams like MTU were eliminated.) What was the deciding factor? Head to head? Comley said that all five factors were treated equally. UNH won its two factors just barely, while UML won Last20 and CommOpp pretty significantly. This proved to be a huge decision, since it was the difference between getting to stay in the East or going to East Lansing to face lower-seeded but host MSU. I think I would be pretty unhappy if I were Lowell. Looking at this another way, why was Lowell seeded ahead of LSSU when the crossovers were done? If Lowell and LSSU are swapped, then MSU hosts LSSU and Lowell plays Northeastern - on a neutral surface. Here is the Lowell-LSSU comparison: UML LSSU EDGE TO RPICH .5491 .5417 UML Head-Head NONE NONE NONE TUC Win% 7-7-3 .500 10-9-3 .523 LSSU Last20 Win% 11-4-5 .675 12-4-4 .700 LSSU CommOpp Win% 1-0-2 .667 1-0-0 1.000 LSSU LSSU wins 3-1, but boy was it close. A difference of a tie in Last20, two ties in CommOpp (only one CommOpp may make it not worth considering), and one win in TUC Win%. I'd consider the teams in a virtual deadlock. It may have been better to seed Lowell 4W and LSSU 3E to avoid this problem. * A factor that may prove to be big in Albany is that the Knick has the 200x100 ice surface. Harvard and RPI have just come off playing two games each on that size rink in Lake Placid. BU and UNH have not played on it this season. Wisconsin has played several times on it at UAA, Minn, and SCSU, and WMU has not played on it. But UNH coach Dick Umile said he was looking forward to it (and he'll have that size surface in his home rink in a year or two), so it should be interesting. * Finally, the seedings set up one first round meeting of teams who have already met this season. UNH defeated RPI on the road early in the season. There are three possible such matchups in the second round, including two possible rematches of conference championship games. RPI could meet Harvard again, who is 2-1-0 vs RPI this season. LSSU could meet Michigan; Michigan has won all 4 meetings. Also, MSU could meet Minnesota, which won the only meeting. And if UML and BU make it out of their regionals, they would meet in the semifinals in St Paul; BU won 3x to Lowell's 1 including defeating Lowell for the HE championship. --- --- Mike Machnik [log in to unmask] Cabletron Systems, Inc. *HMM* 11/13/93 Opinions expressed in this post are solely those of the poster.