final lacrosse bracketology

Normally I would wait til the games have all finished up this afternoon, and then pushed out my last best guess, but I'm busy this PM and not sure I'll have the time.  Instead, here's what you get: two brackets.  (This is the power of the spreadsheet toy I made.  Not only can I predict the RPI and related rankings, I can put together two brackets in the time it used to take to do one.)

Here is the prediction if Harvard beats Penn:

And here is the prediction if Penn beats Harvard:

I am of course making the completely bonkers assumption that Duke will not lose to Boston U. this afternoon.  If that happens, ok, well, I guess Duke is the 2 seed or something.

Mostly these predictions are the same, as you'd expect.  Notably, the similarities are:

-- The play-in games
-- Albany at the #3 seed
-- The seed order of Maryland, ND, Cornell, UVA
-- The seed order of Duke and Cuse
-- UVA/Denver, Cornell/UNC

And the main differences:

-- Hofstra is out if Harvard wins.
-- Penn and Loyola swap seeds depending on whether Penn wins or loses.

Do I think the committee would rematch Hopkins and Loyola?  I bet they would, given the chance.  First-round rematches are not taboo and I'll take a stab at the thinking that a rematch so soon would be a little extra ratings juice.

Also note: there are strong arguments for the order of the 5-8 seeds, but almost equally strong ones for shuffling that around entirely.  UVA having beaten both Cuse and Loyola is one of them; the reason I have UVA down at the bottom of that pecking order is that those wins were months ago and recency does matter somewhat.  This is a tougher season than usual, though, for getting the seeds right.  Penn and Loyola aren't airtight 3 and 4 seeds, either.  Last year I got seeds 1 through 5 exactly right; this year, it wouldn't surprise me if I got all eight seeded teams right but the wrong order entirely.  Let's see what happens when the rubber hits the road.  Lot of bitching, probably; the question for me is whether or not I join in.