D-I Women’s Bracketology: Feb. 9, 2006

Are the Wildcats a lock for No. 1 if they win out? What are Minnesota’s chances for taking home ice from Minnesota-Duluth? The second women’s bracketology column will focus on these issues and a little more.

So what is this week’s bracket projection?

Harvard at No. 1 New Hampshire
Mercyhurst at Minnesota-Duluth
Minnesota at St. Lawrence
Princeton at No. 2 Wisconsin

The only slightly tricky aspect of this bracket is swapping Mercyhurst and Minnesota, the No. 6 and No. 5 teams in the Pairwise Rankings, respectively. This is necessary to avoid an intraconference quarterfinal between Minnesota and UMD. The swap is made in the manner the best preserves bracket integrity and protects the higher seeds, so this is the clear choice.

Who controls its own destiny for the No. 1 seed? UNH or Wisconsin?

My intuition says UNH, but going strictly by the Pairwise Rankings, it’s still too close to call.

This question is tougher to answer than it looks. Right now UNH appears to have a sizeable lead, but the Wildcats play a weaker schedule than Wisconsin down the stretch, and Wisconsin’s RPI in USCHO’s calculation is presently distorted by two games against Vermont that will be dropped by the NCAA’s calculation. When UNH plays its last pair against Vermont, the top teams’ RPI in the USCHO calculation will be more comparable.

If both teams win out, the teams’ RPI project to be so close right now that which team finishes in front might well depend on the results of games in which neither team is participating.

The same goes for their records against teams with RPI greater than .500. Both teams project to have two losses and one tie in that criterion, but Wisconsin will have 18 wins to UNH’s 16 — 17 if Maine ascends into Hockey East’s No. 4 spot. A team like Maine or St. Cloud falling back below .500, or a team like Connecticut or Minnesota State moving above .500, could also affect that comparison.

To summarize the total comparison, UNH will have the head-to-head win over Wisconsin and a full game advantage over Wisconsin in the Last 16 criterion. The common opponents comparison is a wash. Wisconsin, at best, will have a slight edge over UNH in both record against teams above .500 and RPI — a slim advantage that would amount to less than a half game advantage in each category.

In such a scenario, the Pairwise Rankings would say Wisconsin is the No. 1 seed, because each team would have won two selection criteria, and RPI is the tiebreaker. My intuition, however, is that the committee would give UNH the top seed, because UNH’s advantages in two criteria would be decisive, while Wisconsin’s two criteria wins would be murky at best.

Such a decision would not be without precedent. It would be similar to the 2003 Frozen Four, when the Pairwise Rankings said Harvard should be No. 1 based on a mere half-game advantage in its Last 16 games, but the committee went with Minnesota-Duluth instead.

One last point — the same logic most likely applies if both teams lose the same number of games down the stretch. The one exception is if New Hampshire loses to Vermont, a common opponent of Wisconsin. The Wildcats’ would effectively be handing the Badgers the No. 1 seed with such a defeat.

What does Minnesota have to do to win home ice in the NCAA tournament?

This is another difficult question. My intuition says, given UMD wins out the regular season, Minnesota would need to reach the WCHA final (possibly beating UMD a third time to get there), and win at least one game against Wisconsin (whether it be this weekend or the WCHA final), and win out otherwise.

In this scenario, Minnesota would have a clear 3-2 edge in head-to-head play, and a clear game and a half edge in the Last 16 criterion. UMD would have a clear two-game edge in common opponents. In RPI, UMD would project to have a narrow edge, less than the equivalent of half a game. In record against teams with RPI above .500, UMD would still be 8-4-2. Minnesota would 11-7 if its opponent in the WCHA quarterfinals is below .500, putting the Gophers behind UMD by less than a game. Minnesota would be 13-7 if its opponent in the WCHA quarterfinals is above .500, putting the Gophers ahead of UMD by a half game.

Much like the Wisconsin-UNH scenario described above, the Pairwise Rankings might favor UMD by the narrowest of margins, but my intuition says the committee would pick Minnesota. Reason being, Minnesota would have the clear edge in two categories, UMD would have the clear edge in one category, and RPI and record against teams with RPI above .500 would both be about a wash.

A few follow-up points — if UMD loses its WCHA quarterfinal series and Minnesota reaches the WCHA final, the Gophers would have the clear RPI advantage and be well on their way to hosting. Lastly, if the Gophers were to go 2-1 or 1-1-1 against Wisconsin down the stretch and win out otherwise, it would avoid the touchy scenario I described above entirely.

Do you have anything else to say about the teams closer to the tournament bubble?

Not much — there are too many possible scenarios to break down any one.

One point to be made about Mercyhurst — the Lakers are still in good shape to hold off Princeton and the other ECACHL teams. The games against Wayne State are crucial because the Warriors are the last team with an RPI above .500 that Mercyhurst plays, and the Warriors are a common opponent of Princeton.

What are the biggest games of the weekend in terms of NCAA prospects?

The biggest games are fairly obvious. Harvard and Princeton are both squarely on the tournament bubble, so that matchup could ultimately make a huge difference down the stretch. As previously mentioned, the Minnesota-Wisconsin series is crucial for Minnesota’s chances for quarterfinal home ice and Wisconsin’s chances for the No. 1 seed. Sunday’s game against Providence is potentially the last game UNH has against a team with an RPI above .500 before the Hockey East tournament. The game also matters for Providence’s at-large chances, though the Friars’ easiest route to the tournament is through the autobid once again.