Here's why I think sending three Western teams to the East regional this year, presumably to avoid *any* possible intraconference games in the regionals, indicates that the selection committee's priorities have changed in the past two years. The last time there were seven Western and six Eastern teams in the regionals was in 1997. As a reminder, these were the actual tournament brackets produced by the committee (see http://www.slack.net/~whelan/cgi-bin/tbrw.cgi?seeds$1997 for a full description of what was done that year): 5W Mich State (C) 6E Denver U (W) 4W Minnesota (W) 3E Vermont (E) 1W Michigan (C) --+--2E Boston Univ (H) | 2W North Dakota (W) --+--1E Clarkson (E) 3W Miami (C) 4E New Hampshire (H) 6W Cornell (E) 5E CO College (W) Notice that there are two Western teams in the East regional and five in the West, making the Michigan-Michigan State matchup in the second round unavoidable. Here were the regions before teams were swapped that year (Michigan State is considered an honorary Eastern team to balance out the field): Team lPWR RPI Comps | Team lPWR RPI Comps 1 Michigan (C) 1 .628 ND | 1 Clarkson (E) 1 .600 BU 2 North Dakota (W) 0 .588 | 2 Boston Univ (H) 0 .577 3 Minnesota (W) 3 .581 MmCCDU | 3 New Hampshire (H) 2 .591 __CrMS 4 Miami (C) 1 .577 CC__ | 4 Vermont (E) 2 .579 NH __MS 5 CO College (W) 1 .560 __ DU | 5 Cornell (E) 2 .570 __Vt MS 6 Denver U (W) 1 .554 Mm__ | 6 Mich State (C) 0 .547 ______ The committee simply swapped the bottom two teams from each region to get the actual regionals. But if they had done the same thing that year as this, and switched the bottom three teams, they would have ended up with the following regions: Team lPWR RPI Comps | Team lPWR RPI Comps 1 Michigan (C) 1 .628 ND | 1 Clarkson (E) 1 .600 BU 2 North Dakota (W) 0 .588 | 2 Boston Univ (H) 0 .577 3 Minnesota (W) 3 .581 CrVtMS | 3 New Hampshire (H) 3 .591 MmCCDU 4 Cornell (E) 2 .570 VtMS | 4 Miami (C) 1 .577 CC__ 5 Vermont (E) 1 .579 __ MS | 5 CO College (W) 1 .560 __ DU 6 Mich State (C) 0 .547 ____ | 6 Denver U (W) 1 .554 Mm__ There are two possible second-round games if we seed naturally, so the committee would have reversed Cornell and Minnesota in the West and UNH and Miami in the East to avoid this, also switching CC with DU in the East to preserve first-round pairings, but leaving MSU and UVM where they are to avoid a Michigan-MSU matchup, giving the following brackets: 5W Vermont (E) 6E CO College (W) 4W Minnesota (W) 3E Miami (C) 1W Michigan (C) --+--2E Boston Univ (H) | 2W North Dakota (W) --+--1E Clarkson (E) 3W Cornell (E) 4E New Hampshire (H) 6W Mich State (C) 5E Denver U (W) which, like this season's, have no possible intraconference matchups and three Western teams in the East regional. I can conceive of four reasons why they did this: 1) The relative importance of the different priorities (in this case, keeping at least four teams in their own regional versus avoiding intraconference matchups) has changed; 2) Nobody thought of swapping an extra team before this year; 3) The priorities are the same, but some subtle difference between this year's and 1997's field made the triple-switch appropriate now and not then (I'm sure Adam Wodon will tell us what that is ;-)); 4) The relative importance of the priorities is not predefined, and each year's committee picks whichever of the competing possibilites it feels like. My feeling is that #4 (possibly with a bit of #2: "Let's try something new this year") is the case. I have a problem with this, since it opens up the possibility that the committee will, possibly subconsciously, bias its decisions based on the teams actually involved come tournament time. (I wonder if that wasn't done this year; Boston College was kept in the East last year on the basis of attendance and pairwise comparisons, even though an intraconference matchup--which did occur--would have been avoided by sending them West instead of Yale. Perhaps shipping them in the third swap, despite obvious attendance considerations, was something of a "makeup call".) The NCAA's avoidance of opportunities to make the selection process more objective and algorithmic puzzles me. In fact this attitude seems to be even stronger in other sports, judging by comments like "none of us who edit RPIs here at the national office would ever expect or want a situation where a computer is picking teams for a tournament" in the article <http://www.ncaa.org/news/19990215/active/3604n35.html> pointed out by Bob Stagat last week. It's as if a conference decided that instead of having a uniquely defined tie-breaking algorithm to seed teams when their final records were the same (or even similar), they should have a committee meet the day after the season ended and decide how the teams should be seeded for the playoffs based on a smorgasboard of statistical measures and their own personal judgement. (The ECAC did something like this back when they had 17 teams and a unbalanced schedule, right? That also amazes me.) The current ECAC tiebreaker system is complicated, and you could program a computer to resolve it, but I don't see anyone calling for a "human touch" in seeding the league playoffs. (This argument works even better for the NFL, where the schedule is quite unbalanced, ties in the standings are more common, and the tiebreaker system is deliciously arcane.) John Whelan, Cornell '91 [log in to unmask] http://www.amurgsval.org/joe/ HOCKEY-L is for discussion of college ice hockey; send information to [log in to unmask], The College Hockey Information List.