Projecting the 2013 Big Ten Basketball Race: February 18

Last week, it appeared to be a three-way race for the title between Indiana, Michigan, and Michigan State. Since then, MSU put an emphatic beating on Michigan before taking care of business at Nebraska. Indiana answered with a pair of home smackdowns of conference bottom-feeders, while Michigan struggled early with Penn State at home. As always, the ratings are here and championship and bye odds are calculated from 100,000 simulations. By next week, scenarios might be simple enough to do those by brute force.

The Contenders

Basic Method

Team	Avg wins	8-10	9-9	10-8	11-7	12-6	13-5	14-4	15-3	16-2	Outright	Shared	Top 4
Michigan State	14.29	N/A	N/A	N/A	0.13%	2.87%	16.97%	37.62%	33.14%	9.27%	41.50%	69.52%	99.43%
Indiana	13.92	N/A	N/A	N/A	0.42%	6.12%	26.56%	39.43%	22.91%	4.56%	23.63%	49.39%	98.75%
Michigan	12.86	N/A	0.03%	0.78%	6.50%	24.46%	41.84%	26.39%	N/A	N/A	1.47%	16.91%	94.91%
Wisconsin	12.42	N/A	0.07%	1.57%	11.89%	37.28%	41.31%	7.88%	N/A	N/A	0.52%	5.75%	91.03%
Ohio State	10.57	2.00%	12.99%	31.62%	34.82%	16.32%	2.25%	N/A	N/A	N/A	<0.01%	0.16%	14.26%

Michigan State is now the favorite, with almost all of the gap coming from having home court in the Tuesday night game; Indiana, Michigan, and Michigan State are ranked #4, 5, and 6 respectively. After Tuesday, the remaining schedules for Michigan State and Indiana are eerily similar:

• a road trip to Michigan
• another road game against a top-half opponent (Indiana at Minnesota, MSU at Ohio State)
• a home game against a top-half opponent who they already beat on the road (Indiana vs. Ohio State, MSU vs. Wisconsin)
• and a home game against a bottom-half opponent (Indiana vs. Iowa, MSU vs. Northwestern)

Michigan and Wisconsin have easier remaining schedules (Michigan has their only two top-half games remaining at home; Wisconsin has the road trip to MSU but their other four games are against the bottom four), but they have to make up two games on the teams ahead of them. Michigan almost controls their own destiny for a share of the title (they need only another loss by whichever of MSU and Indiana wins their game), but Wisconsin needs a lot of help from other teams. (On a side note, I'm starting to believe the schedule-maker is a Wisconsin fan: no trip to Michigan and no visit from Indiana this year.) Ohio State, at three games back, has only a theoretical chance.

Margin-Aware Method

Team	Avg wins	8-10	9-9	10-8	11-7	12-6	13-5	14-4	15-3	16-2	Outright	Shared	Top 4
Indiana	14.29	N/A	N/A	N/A	0.10%	2.41%	17.23%	38.32%	32.71%	9.24%	40.35%	70.30%	99.14%
Michigan State	13.81	N/A	N/A	N/A	0.29%	6.69%	30.06%	40.36%	19.73%	2.88%	20.59%	47.74%	98.25%
Michigan	13.16	N/A	1 in 25,700	0.17%	2.55%	16.18%	43.29%	37.80%	N/A	N/A	1.67%	23.57%	96.51%
Wisconsin	12.67	N/A	0.01%	0.44%	6.02%	31.62%	50.24%	11.68%	N/A	N/A	0.42%	6.99%	92.02%
Ohio State	10.90	0.68%	6.82%	24.82%	39.78%	25.21%	2.69%	N/A	N/A	N/A	<0.01%	0.13%	13.59%

Here, Michigan State's tendency for close games knocks them down a notch in the ratings (to #11), making Indiana the favorite. About half of the swing here is from Tuesday night's game being nearly a coin toss instead of a MSU as a favorite.

Everyone Else (Non-Futility Division)

Basic Method

Team	Avg wins	3-15	4-14	5-13	6-12	7-11	8-10	9-9	10-8	11-7	Top 4
Minnesota	9.17	N/A	N/A	N/A	0.17%	4.04%	19.98%	38.14%	29.77%	7.90%	0.90%
Illinois	8.80	N/A	N/A	N/A	0.23%	5.39%	31.78%	41.95%	18.33%	2.32%	0.49%
Iowa	8.75	N/A	N/A	N/A	0.82%	8.41%	29.17%	39.61%	20.04%	1.95%	0.23%
Purdue	6.43	N/A	N/A	15.18%	40.79%	31.93%	10.49%	1.53%	0.08%	N/A	<0.01%
Northwestern	5.99	N/A	4.06%	27.00%	40.14%	23.26%	5.26%	0.27%	N/A	N/A	<0.01%
Nebraska	4.24	21.94%	41.87%	27.34%	7.81%	1.00%	0.05%	N/A	N/A	N/A	0%

Despite Minnesota's best efforts, they're not likely to be able to play their way out of the tournament; in fact, they're probably in better position than Ohio State despite the conference record being two games worse thanks to careful scheduling giving them an RPI boost - only four games outside the top 120 all season to date (although they have three more in their final five in conference), and as a result a much better RPI top 100 record (11-7 to 4-7). Illinois should also feel pretty safe with four RPI top 20 wins. Both should have no difficulty making the tournament with two more wins - and they both get Penn State and Nebraska in their last five.

Iowa is in considerably worse shape, despite the same record in conference now and a very similar projected finish. To overcome their abominable non-conference schedule (and soft, by Big Ten standards, conference schedule, with two of the top three as single-plays and two games each against all of the bottom four), Iowa probably has to get to 9-9 to even get a glance from the committee. 11-7 might be hard to ignore (especially since it would include a win at Indiana), but at 10-8 I suspect they will get the Colorado 2011/Penn State 2009/Virginia Tech every year treatment barring a major conference tournament run.

Margin-Aware Method

Team	Avg wins	3-15	4-14	5-13	6-12	7-11	8-10	9-9	10-8	11-7	Top 4
Minnesota	9.28	N/A	N/A	N/A	0.04%	2.20%	16.48%	40.69%	32.61%	7.98%	0.34%
Iowa	9.19	N/A	N/A	N/A	0.11%	2.38%	16.67%	42.24%	36.28%	2.32%	0.08%
Illinois	8.53	N/A	N/A	N/A	0.14%	5.29%	46.60%	38.15%	9.22%	0.60%	0.07%
Purdue	6.23	N/A	N/A	17.98%	47.95%	27.49%	6.00%	0.56%	0.02%	N/A	<0.01%
Northwestern	5.88	N/A	3.41%	32.18%	41.26%	19.48%	3.52%	0.14%	N/A	N/A	<0.01%
Nebraska	3.74	43.15%	41.63%	13.30%	1.80%	0.11%	1 in 42,500	N/A	N/A	N/A	0%

Similar story here. Interestingly, Illinois has a very polarized schedule remaining - two games at 92% or better to win, one at 21%, and one at 10%. The only one that wouldn't be a big surprise to go against the odds is at Iowa. (Iowa's is somewhat less extreme, with two games as 2:1 favorites to go with two as 88% or better and one at 6%.)

The Quest for 0-18

Method	Avg wins	0-18	1-17	2-16	3-15	4-14	5-13
Basic	0.55	55.03%	35.49%	8.49%	0.93%	0.05%	1 in 111,000
Margin-aware	0.32	71.24%	25.33%	3.24%	0.18%	1 in 22,000	1 in 2,460,000

Penn State's best remaining chance at a win (home vs. Iowa) went by the wayside in a two-point loss, and then they put up a surprisingly strong fight at Michigan but ultimately went down. Their rating in margin-aware came up noticeably with two closer-than-expected calls, but that just gets their win probability for their final two (at Northwestern, vs. Wisconsin) up to 11% each (in the basic method it's 17%).

Michigan State's Remaining Games

From the previous projection:

Michigan: won by 23 (projected 58% to win by basic method / 43% by margin-aware / -1.5 margin)
@ Nebraska: won by 9 (projected 78% / 81% / +8)

Remaining games, in schedule order:

Indiana: 65% / 49% / even (was 62% / 44% / -1)
@ Ohio State: 55% / 39% / -2.5 (was 49% / 31% / -4.5)
@ Michigan: 34% / 20% / -7.5 (was 28% / 13% / -10.5)
Wisconsin: 81% / 78% / +7 (was 78% / 77% / +6.5)
Northwestern: 93% / 95% / +16 (was 91% / 93% / +13.5)

Clobbering a top-5 team tends to swing the numbers in your favor; the only one to move less than a point is the Wisconsin game, and that's because they did some clobbering of their own (though "only" against a low-end top-25 team).

How important is the Indiana game? As with last week, I ran 100,000 simulations with the outcome fixed as a win and another 100,000 fixed as a loss. The results:

	Basic				Margin-Aware
	Outright with win	Share with win	Outright with loss	Share with loss	Outright with win	Share with win	Outright with loss	Share with loss
Michigan State	58.86%	88.58%	8.90%	33.23%	38.80%	79.03%	2.86%	17.47%
Indiana	55.24%	86.78%	6.99%	29.65%	67.11%	94.00%	12.71%	45.61%

The winner is very likely to claim a share of the title, at minimum. The loser's outright title hopes are faint, although a share is still plausible.