Simulating the 2010 Tournament (March 2 Edition)

The details are the same as last time. I've slightly tweaked the adjustment for recent results, but didn't make any big changes. The ratings I used are here, the four fields (just used Bracketology seeds this week) are here, and the full results are here. Now for the teams with a 1% chance or better of winning it all (including Duke!):

I'll start by saying that I think I'm underrating Kansas here. They've fallen behind Duke in both my ratings and the "Champ" column, mostly based in their disappointing performance in their last two games. While for most teams I think that recent results have a significant effect on future lines, I'm not sure that this is the case with the Jayhawks, since they are a known entity at this point, and we haven't seen much variance in their lines all season. This week (vs. Kansas St., at Mizzou) will clear things up on that front.

As a direct result, I think Duke is slightly too high, since they end up playing Kansas in the Final Four with some frequency. I wouldn’t be at all surprised if the numbers for Duke and KU are flip-flopped a week from now. Not that it makes a huge difference, since 15-1 is still quite good even at 17%.

There was value in Wisconsin last time, and their rating has gone from .945 to .959 (welcome back, Leuer), while they're all the way up to 60-1 at SBET, so I still like their odds despite a slight drop in seed (#3 to #5). I also think the Badgers have a good chance of making some noise in Indianapolis next week and moving above the fifth line.

Missouri's situation is essentially the same as last week, although they've moved up a seed in Bracketology (to #7), and have a slightly better chance of just making the tournament (Lunardi says 85%, for whatever that's worth). This simulation assumes each team is 100% to make the tournament, so to be fair on them you should multiply 1.7% by .85, which is 1.45%, which is still extremely good at 200-1.

I was surprised to see that BYU is still 200-1 at Sportsbook. I know they lost to the powerhouse that is New Mexico (BYU was favored by eight in that one), but they are a #4 seed right now, and even after adjusting their KenPom rating by the spreads I have them as the 17th best team in the country. They are 50-1 at SBET, which seems far more reasonable to me.

SBET and SIA were the only books offering traditional NC odds as of last night, but WSEX has opened up their 4-pay market, which is exciting. In their market, you can buy a team for anywhere between $2 and $99 (in theory), and the payout is based on how far they go in the tournament. If they win it all you get $100, if they lose in the final it's $50, and if your team reaches the Final Four and loses then the payout is $25.

Shorting teams is also an option, although there's always a substantial gap between the buy and sell prices. Combining the NC/runner up/F4 figures above with my sim, I came up with how much each team should cost, and compared that to the actual buy/sell prices:

Same thoughts on Kansas and Duke as above -- I'd definitely buy the Blue Devils, but I wouldn't want to sell Kansas since I think I'm temporarily underrating them.

There's nothing temporary about my "underrating" of Kentucky and Syracuse, since I've been consistent on them for awhile now. Kentucky is probably the third best team in the country, but there's just not a huge difference between them and someone like Kansas St.

I'm not sure what Texas is still doing at 8/11, the sixth highest value on the board. They've dropped to #12 in my ratings, and that's before you even consider that they're a #8 seed in Bracketology this week. The Longhorns also have a tough game on Saturday at Baylor, so they could continue to fall in the bracket. As a #1 seed I wouldn't want to play Texas in the second round, but their implied value at WSEX is still ridiculous.

Purdue is a decent value at 8 according to the table above, but since we're only one game into the post-Hummel era, I'm still overrating them. They were -3.5 against Michigan St. on Sunday, which is 2-2.5 points lower than they would have been with a healthy Hummel; we'll have to see if their poor showing in that one causes their lines this week (vs. Indiana, at PSU) to be even less encouraging.

This week's Bracketology has a potential (2) New Mexico vs. (7) Georgia Tech matchup in round two, and my ratings have the Yellow Jackets as favorites in that game. I'm really hoping the Lobos beat TCU on Wednesday and have a decent showing in the MWC tournament, because there's some real potential here; I went back 10 years and couldn’t find a #2 seed that was a second round underdog. The logistics of shorting a team at four are probably painful, but New Mexico is extremely overvalued right now.