Projecting the 2013 Big Ten Basketball Race: February 26

Last week, the race was pretty tight between Michigan State and Indiana for first place. Two difficult games for Michigan State both went the wrong way, however, and now it is quite unlikely that Indiana will not at least get a share of the title. Still, there are three more byes up for grabs (along with the fight for the 2 and 3 seeds to stay out of Indiana's half of the bracket) and an interesting fight among four teams for the 6 and 7 seeds (also presumably avoiding Indiana, though it's not as if the 2 or 3 seeds are going to be pushovers either).

The title race isn't quite simplified enough to go to game-by-game calculations for title and bye scenarios, so these will again be computed from 100,000 simulations. As always, ratings are here.

The Contenders

Basic Method

Team	Avg wins	10-8	11-7	12-6	13-5	14-4	15-3	16-2	Outright	Shared	Top 4
Indiana	14.69	N/A	N/A	0.42%	7.10%	31.93%	44.10%	16.46%	68.31%	93.95%	99.47%
Michigan	13.08	0.16%	3.11%	18.91%	43.95%	33.87%	N/A	N/A	3.33%	20.51%	91.60%
Michigan State	12.96	N/A	1.26%	20.79%	58.37%	19.58%	N/A	N/A	1.44%	9.22%	91.73%
Wisconsin	12.83	0.19%	3.73%	24.50%	55.87%	15.71%	N/A	N/A	0.54%	7.25%	88.97%
Ohio State	11.62	6.52%	34.43%	49.25%	9.80%	N/A	N/A	N/A	<0.01%	0.68%	28.24%

Indiana's remaining schedule is no cakewalk (as we'll see in a later table), but when you have a 1.5 game lead over your nearest competition and their schedule is nearly as difficult, it doesn't really matter. Ohio State, having to go to Indiana, is the least likely to earn a bye, with Michigan, MSU, and Wisconsin nearly equal.

Margin-Aware Method

Team	Avg wins	10-8	11-7	12-6	13-5	14-4	15-3	16-2	Outright	Shared	Top 4
Indiana	14.84	N/A	N/A	0.16%	4.14%	27.45%	47.97%	20.28%	73.21%	96.27%	99.60%
Michigan	13.26	0.04%	1.34%	12.84%	43.65%	42.13%	N/A	N/A	2.33%	20.42%	93.73%
Wisconsin	13.02	0.03%	1.22%	15.33%	63.32%	20.10%	N/A	N/A	0.29%	6.90%	92.19%
Michigan State	12.87	N/A	0.99%	23.22%	63.24%	12.54%	N/A	N/A	0.53%	4.50%	89.41%
Ohio State	11.76	2.90%	26.59%	61.89%	8.62%	N/A	N/A	N/A	<0.01%	0.33%	25.07%

Wisconsin moves up slightly with MSU's schedule difficulty even higher than in the no-margin ratings (and with their non-MSU games all being as easy as they get in the Big Ten this year), but again Ohio State is most likely to be the odd man out.

Remaining Schedule and Tiebreaker Situation

Tiebreakers are as follows (this is the 2012 release; couldn't find one for this year):

1. Head-to-head record
2. Winning percentage against the top team (or teams, if tied)
3. Winning percentage against the next team (or group of tied teams), and so on down the standings
4. Combined record of all D-I opponents
5. Coin toss

For the moment, we will assume that none of the 10-win teams lose out and/or none of the 8-loss teams win out, so we can ignore any ties between groups. Here are the remaining schedules and head-to-head results for the top 5: (projections are basic win percentage / margin-aware win percentage / average margin; head-to-head is listed as home/away results, with a dash indicating no game and a question mark indicating an upcoming game)

Team	This week	This weekend	Next week	Next weekend	vs. IND	vs. MICH	vs. MSU	vs. WISC	vs. OSU
Indiana	@MINN 59%/64%/+3	IOWA 91%/95%/+15.5	OSU 81%/87%/+10.5	@MICH 38%/39%/-2.5	N/A	W/?	W/W	L/-	?/W
Michigan	@PSU 91%/96%/+17	MSU 72%/82%/+8.5	@PUR 82%/87%/+10.5	IND 62%/61%/+2.5	?/L	N/A	?/L	-/L	W/L
Michigan State	BYE	@MICH 28%/18%/-8.5	WISC 75%/74%/+6	NW 93%/95%/+16.5	L/L	W/?	N/A	?/W	W/L
Wisconsin	NEB 88%/95%/+15.5	PUR 88%/93%/+14	@MSU 25%/26%/-6	@PSU 83%/89%/+11.5	-/W	W/-	L/?	N/A	W/L
Ohio State	@NW 72%/80%/+7.5	BYE	@IND 19%/13%/-10.5	ILL 72%/84%/+9	L/?	W/L	W/L	W/L	N/A

We can clinch the 2 seed with wins in the next two as long as Ohio State loses a game; even if that doesn't happen, a win over Northwestern or either Michigan or Wisconsin winning out (besides our game) to join the tie would give us the 2 on head-to-head. (If we then lost to Northwestern, Michigan and Wisconsin both dropped another game, and Ohio State won out, we would be the #3 seed behind Ohio State due to their win over Indiana.) Any two wins guarantees a bye; at worst we would be tied with Ohio State and whichever of Michigan and Wisconsin that we swept, with an advantage in head-to-head. (We could fall to the 4 seed if the team we sweep loses another, however.)

Everyone Else (Except Penn State)

Basic Method

Team	Avg wins	4-14	5-13	6-12	7-11	8-10	9-9	10-8
Minnesota	8.65	N/A	N/A	0.44%	8.76%	32.71%	41.98%	16.12%
Illinois	8.63	N/A	N/A	N/A	5.20%	38.16%	44.62%	12.02%
Iowa	8.13	N/A	N/A	2.41%	18.72%	44.83%	31.42%	2.62%
Purdue	6.86	N/A	N/A	36.92%	42.93%	17.17%	2.82%	0.16%
Northwestern	5.16	13.16%	59.60%	25.62%	1.62%	N/A	N/A	N/A
Nebraska	4.86	36.82%	43.25%	17.07%	2.72%	0.15%	N/A	N/A

Minnesota and Illinois are favorites to get the 6 and 7 seeds (and thus avoid Indiana's half of the bracket). Given their identical conference records to date and similar overall records (Minnesota's is just one game better), it's amazing that the Gophers are a near-lock despite their recent slump and Iowa, after their loss at Nebraska, is effectively in auto-bid-or-bust territory. It is once again time for my rant against the RPI: while Minnesota should certainly be rewarded for scheduling most of their non-conference games against the top 110 and Iowa penalized for having only three serious non-conference games (and managing to lose one of the gimmes on top of that), any metric that sees Minnesota as 74 spots better (#17 in the RPI entering Monday to Iowa's #91, per Crashing the Dance) is putting far too much emphasis on the relative quality of the bad games. For a tournament-level team, the difference between playing #1 and playing #50 might be half a win on average, while the difference between #300 and #347 is less than 1% of a win - yet the RPI considers a schedule of #1 + #347 to be about equal to #50 + #300. The no-margin system, which puts much less emphasis on the quality of your cupcakes and more on games against teams close to your own level, makes the gap only about half as large (Minnesota #32, Iowa #71) and suggests that Iowa should be at least a fringe bubble team instead of completely out of the discussion.

Margin-Aware Method

Team	Avg wins	4-14	5-13	6-12	7-11	8-10	9-9	10-8
Minnesota	8.81	N/A	N/A	0.11%	4.92%	27.90%	48.00%	19.07%
Iowa	8.48	N/A	N/A	0.46%	7.73%	38.39%	50.63%	2.79%
Illinois	8.41	N/A	N/A	N/A	5.11%	53.99%	35.90%	5.00%
Purdue	6.64	N/A	N/A	48.90%	39.59%	10.38%	1.09%	0.04%
Northwestern	5.11	10.99%	68.30%	19.90%	0.82%	N/A	N/A	N/A
Nebraska	4.48	59.40%	33.89%	6.24%	0.46%	0.01%	N/A	N/A

Iowa rates even higher here thanks to their many near-misses against the upper echelon; they even have the upper hand in the race for the 7 seed over Illinois.

Remaining Schedule and Tiebreaker Situation

Team	This week	This weekend	Next week	Next weekend	vs. MINN	vs. ILL	vs. IOWA	vs. PUR
Minnesota	IND 41%/36%/-3	PSU 94%/98%/+20	@NEB 64%/77%/+6.5	@PUR 66%/70%/+4.5	N/A	L/W	W/L	-/?
Illinois	BYE	NEB 86%/91%/+12.5	@IOWA 49%/34%/-3.5	@OSU 28%/16%/-9	L/W	N/A	-/?	W/L
Iowa	PUR 77%/86%/+10	@IND 9%/5%/-15.5	ILL 51%/66%/+3.5	NEB 76%/90%/+12	W/L	?/-	N/A	?/L
Purdue	@IOWA 23%/14%/-10	@WISC 12%/7%/-14	MICH 18%/13%/-10.5	MINN 34%/30%/-4.5	?/-	W/L	W/?	N/A

This chart should make it pretty clear why Purdue is likely to be #9: their remaining schedule is very nasty compared to the other three (who all play Nebraska, and Minnesota adds Penn State as well). I left Nebraska and Northwestern out because their chances of getting to #7 are pretty much nil. Should it come down to "record against top teams" for the tiebreaker (and that seems entirely plausible since the only teams to have won on the road within the group did it against each other), none have beaten Indiana or Michigan, Minnesota has beaten MSU, Illinois has beaten OSU, and Iowa has beaten Wisconsin.

The Quest for 0-18

System	Avg wins	0-18	1-17	2-16	3-15	4-14
Basic	0.51	57.12%	34.91%	7.34%	0.61%	0.02%
Margin-Aware	0.32	71.13%	25.73%	3.01%	0.13%	1 in 59,000

Penn State continues to put up a fight in their recent games, losing by just five at Illinois. As a result, the margin-aware rating keeps climbing, but the opportunities for a win continue to decrease. Their four remaining chances to avoid this ignominy:

• vs. Michigan: 9% / 4% / -17
• at Minnesota: 6% / 2% / -20
• at Northwestern: 20% / 14% / -9.5
• vs. Wisconsin: 17% / 11% / -11.5