Anyway, I don't toot my own horn much, but this is one small area where I'll play that trumpet fortissimo. In three years of doing this I've only missed two teams - both in 2010 when the system was in its infancy, and one of those misses was because I had to publish the thing before the last conference tournaments were finished up, and Mount St. Mary's upset Siena in the MAAC tourney at the last minute. Two years in a row I've been 16-for-16. The teams don't always go in the right places, but that's to be expected; nothing's perfect. But then again, sometimes they do. I was also 8-for-8 on seeded teams last year, and perfectly nailed two of the eight matchups. I'll put this system up against anything you see anywhere else, be it on LaxPower or IL or what have you.
Some of the things to keep in mind as you read:
-- This isn't a prediction; as always, it's "how do things stand right now." That's much more useful than a prediction, of course: it gives you an idea of what a team has to do to hang on to their spot, how safe they are, or how far they have to climb. A prediction would just be me foghorning about my opinion. Which would be far less likely to be accurate.
-- Conference autobids are, for now and until they're actually earned, handed out to the team currently leading the conference. 2-0 is better than 1-0. Ties have to be broken somehow, so I use the LaxPower computer rankings, since those are semi-predictive of future success and make it less likely that I'll have to change it week-to-week.
-- One more autobid this year, so the field is now split evenly between autobids and at-larges at eight apiece.
-- Technically, the tournament committee seeds only the top eight teams, and then scatters the other eight based on travel considerations alone. We all know that's not really the case; the #1 and #2 seeds generally always get the really low autobids, like from the MAAC or something. I make an effort to take travel into consideration, but not a very strong effort at this point in the year. Right now it's more about where teams are in the pecking order, so the nonseeded teams hew a little closer to a seeding order than they might when its actually go time.
-- I have a numerical system that spits out a ranking of teams from 1 to whatever, which I then fudge because the system doesn't know things like who beat who. Like if Team A and Team B are pretty close, and Team B beat Team A, you know the committee cares about that. So the ranking is not gospel. However, for the first four and next four out - the bottom side of the bubble - I don't bother fudging, I just put the teams in the order I was given them. Because why spend the time? Plus then you know who's closest and who's furthest.
Enough talk. Here it is.
Time to get into your head and figure out what issues you have with me before you say them:
-- You're such a damn homer. How is UVA in this thing? Don't blame me, blame the RPI. I don't care what screeds you have against the RPI, the committee uses it (a lot) and as it happens UVA is 11th in the RPI and does well in the other metrics, which are also based off the RPI.
-- Wait a minute, we just played OSU. Well, yes, and we also played #7 Syracuse and will have played #6 Duke, too. The committee isn't totally averse to rematches and I think shipping an at-large to Denver when there are a lot of crappy autobid teams who could go is the sort of thing that makes them bend on rematches. They're more concerned in not having conference-mates play each other (and remember, we've seen that they don't consider the ACC a conference, so Duke in fact would be a possibility.)
-- What the shit happened to Hopkins? They haven't beaten anyone. At all. The teams they've beaten are so bad they're killing the Hop's metrics. Like UVA, they'll have chances to rectify this. Unlike UVA, at this point they don't even have a Drexel or a Stony Brook to prop them up.
-- Duke is 5-4, why are they even in, let alone seeded? Their four losses are four of the top five seeds and they've beaten Loyola and Carolina.
Some other notes:
-- Hofstra is strong enough right now to have an at-large bid if they needed it. If they get through their season without too many trip-ups they can potentially insulate themselves from a loss in the CAA tourney.
-- Quinnipiac is the better team, in my opinion, in the NEC, over RMU. But rules are rules and RMU has just the tiniest sliver of a margin over Quinnipiac in those computer rankings.
-- Marist is quite a little surprise, but - and this is rare for a MAAC team - they would actually be in the "next four out" column, under Hopkins, if they didn't have the autobid. It probably means nothing in the grand scheme because they don't have the schedule necessary to pull off an at-large if they falter in their conference. I just find it interesting that someone other than Siena is the class of the MAAC right now.
Here are the games to watch this coming week from a bracketology standpoint:
-- Hofstra at St. John's. St. John's is lurking, and having Hofstra at home (not that it's a real long trip for the
-- Notre Dame at Ohio State. Notre Dame only stands to lose seeding, but if OSU could get a win here, they'd set it up so they'd have to do a major choke job down the stretch in order to miss the tourney.
-- Yale at Princeton. These two teams are like right next to each other on the bubble. The loser will have a major uphill climb to the tourney.
-- North Carolina at Maryland. UNC has work to do, but this is going to be a toughy for them.
-- Penn at Cornell. The Quakers already have a pretty strong resume with wins over Princeton and Duke, but this would really be a kicker for them. Could even vault them to #1, depending on the rest of the week's action.
-- Drexel at Hofstra. Big CAA matchup, one of the conference's biggest of the year.
-- Virginia at Johns Hopkins. No need to tell you about the stakes.
No comments:
Post a Comment