clock menu more-arrow no yes

Filed under:

College basketball rankings: Big Ten advanced stats, predictions through 3 games

New, 7 comments

Friday Facts are back, and better than ever.

Mike McGinnis/Getty Images

I have massively upgraded my spreadsheet for this college basketball season, and the result is that the Friday Facts are no longer just a Big Ten season thing. They're year-round.

A couple changes, though. First, instead of using Kenpom efficiency ratings as the base for comparison, I will be using the T-Rank Power ratings instead. Why? Because they're mine, and they're fun.

Second, the "adjusted efficiency margin" that was at the core of the Friday Facts last year is a little bit changed. Last year, the margin was adjusted to show how each team was playing compared to the average Big Ten team. This year, the margin is tied to how each Big Ten team is playing compared to the average Division I team. In other words, the adjusted efficiency margins will be higher, in absolute terms. But they will still reveal a relative rank within the Big Ten.

The Friday Facts

Rank Team Name W-L Adj OE Adj DE Barthag B1G Proj. EM+/G
3 Wisconsin 3-0 116.1 85.6 0.9711 15-3 42
12 Ohio State 2-0 110.9 90.9 0.9077 11-7 27
19 Michigan 3-0 114.9 95.8 0.8893 11-7 20
21 Maryland 3-0 108.0 90.4 0.8856 11-7 29
22 Iowa 2-1 111.8 93.9 0.8823 10-8 23
24 Purdue 3-0 105.5 89.2 0.8731 10-8 28
26 Michigan State 1-1 110.8 93.9 0.8697 10-8 6
31 Illinois 2-0 108.1 92.7 0.8539 9-9 25
32 Indiana 3-0 111.7 95.9 0.8526 9-9 22
36 Minnesota 2-1 108.3 93.5 0.8438 9-9 17
48 Nebraska 2-0 106.8 94.0 0.8128 7-11 12
78 Penn State 2-1 103.9 96.0 0.7120 6-12 5
105 Northwestern 3-0 90.7 86.1 0.6436 4-14 1
141 Rutgers 1-1 99.8 98.3 0.5435 3-15 -3

The left column shows each team's overall rank in the T-Rank Power ratings. The right-most column shows how each team has actually performed, compared to DI average, in terms of adjusted efficiency margin. Unsurprisingly, Wisconsin leads the way in both overall rank and adjusted efficiency margin per game, after three impressive wins. Michigan State has been the most disappointing team, based on its narrow victory over lowly Navy. The relatively close loss to Duke on a neutral court was nothing to be ashamed of.*

Of course, it's too early to draw any big conclusions. Most of the Big Ten teams have taken care of business by blowing out terrible teams. Iowa, Michigan State and Minnesota all have losses, but each was to a team in the T-Rank top 10. We still have much to learn about these teams.

But one thing I'm pretty sure of: the Badgers are who we thought they were. "Summer Camp Sam" Dekker has shown up in games, and he is a sight to behold. Frank Kaminsky is oozing confidence and resolve and skill. Nigel Hayes shows signs of becoming an all-time great. And, oh yeah, two senior guards, etc. I don't think the Badgers will actually win the Big Ten by four games -- one of those other teams will separate from the pack and make a run -- but another 30-win season is a reasonable goal for this squad.

*At this point, the overall rank diverges from adjusted efficiency margins because preseason projections are still accounted for in the rank. Over time, the adjusted efficiency margin will reflect the overall rank as well.

This week's Big Ten games

Here are the top 10 scheduled Big Ten games for this week, ordered by Torvik Thrill Quotient.

Matchup T-Rank Prediction TTQ
22 Iowa vs. 20 Syracuse Syracuse, 69-68  (50%) 82
24 Purdue vs. 41 Kansas St. Purdue, 70-69 (57%) 77
36 Minnesota @ 67 St. John's St. John's, 69-68 (51%) 76
48 Nebraska @ 94 Rhode Island Nebraska, 69-68 (52%) 71
19 Michigan vs. 53 Oregon Michigan, 76-71 (67%) 69
61 Arizona St. vs. 21 Maryland Maryland, 69-64 (69%) 65
141 Rutgers vs. 123 Vanderbilt Vanderbilt, 66-64 (54%) 58
78 Penn St. @ 178 Bucknell Penn St., 68-65 (61%) 55
81 Boise St. @ 3 Wisconsin Wisconsin, 78-59 (96%) 52
108 Akron @ 78 Penn St. Penn St., 69-63 (73%) 50

(Yes, T-Rank does predictions.)

The big caveat here, of course, is that this is holiday tournament time. For example, Wisconsin will have a game against either Florida or Georgetown next Thursday, and a big game against the likes of North Carolina, UCLA or Oklahoma next Friday. So it will be a more thrilling week than this chart reflects.

The week gets off to a great start, with an interesting matchup between Iowa and Syracuse, both coming off disappointing losses (to Texas and Cal, respectively). The loser of that game might find itself on the bubble.

Weekly selection show?

The buzz this week is that the NCAA selection committee might take a cue from the football selection committee and do a weekly show. Most college hoops fans reacted with disgust at this idea, but there were a few strong voices against the orthodoxy.

For example, John Gasaway tweeted that weekly updates might prevent travesties such as N.C. State (Kenpom No. 66) getting an at-large bid over SMU last year. And Thomas Beindt over at BT Powerhouse thinks that the weekly selection show would be good for college basketball because it would grab headlines and keep people talking about the sport.

These are reasonable arguments, but I'm not persuaded.

First, the core fact here is that the NCAA selection process is not broken. It doesn't need fixing. So any proposed changes should be considered warily.

Second, I think there' s a real danger that weekly seeding process will make the end result significantly worse. The worry here is that the committee will be forced to form opinions about teams in January, without much good information about how good the teams really are. Then those opinions will set and become resistant to contrary information. This is unfortunately a fact about human nature: we have a strong urge to defend our opinions, no matter how stupid they are. And recent science suggests that being confronted with contrary evidence makes us less likely to change our minds.

This is why judges instruct juries not to discuss the case with one another until all the evidence is in and the trial is over. We want jurors to withhold judgment as long as possible so they don't get any stupid opinions into their heads. It should be the same thing for the selection committee -- delaying judgment as long as possible will create a better end result. Forcing them to make early judgments will make the end result worse.

Third, in light of this danger, I don't see why I should care whether people are talking about college basketball. Of course a weekly selection show would be good for sports media -- it would fuel talk radio, create clicks, etc. But good for the sport? Not a chance. College basketball is plenty popular enough. There are a dozen games on my TV every night that I don't have time to watch. In the last 10 years, I've seen basically every Badger game on TV even though I haven't lived in Wisconsin. The sport is doing fine. It doesn't need any phony buzz.

StatWatch™

The Badgers have been stuffing the stat sheet so far, so there is a lot to choose from here. But so far, the most intriguing Wisconsin stat is this: the man leading the team in minutes played is ... Nigel Hayes, with 93.

Will it last? I doubt it. But it's a sign that Mr. Hayes has already become quite possibly the most indispensable Badger.