The terrifying world of Torvik without preseason projections. You can drop the preseason projections out of Torvik by advancing the start date of your sample one day.

You should not do this, the stat persons say. Kenpom once asserted that his projections were more accurate if you left a little bit of preseason expectation in indefinitely, but people were very cross about this and the benefit was small so he didn't bother. We're not even to January, so the expectations carry an appropriately large amount of weight, especially because teams have played far fewer games than they usually do.

But let's do it anyway, because my face hasn't melted in at least a couple days. The Big Ten in this universe:

• #5 Wisconsin
• #8 Illinois
• #9 Iowa

Okay we're fine

• #15 Northwestern
• #21 Rutgers

tentacles emerge from the floor

• #25 Michigan
• #26 Indiana
• #27 Penn State

tentacles grow flowers

• #37 Purdue
• #46 Minnesota
• #50 Ohio State
• #51 Maryland

pink elephants emerge from the flowers

• #86 Michigan State
• #91 Nebraska

pink elephants explode, covering room in viscera, viscera sings "Sittin' on the Dock of the Bay."

It's not like the ranking systems are particularly enamored with MSU even when you leave expectations in. They're still second-to last on Torvik, but much closer to Maryland at #51 than Nebraska. Northwestern also gets penalized about the same amount—~30 spots—but they drop from fourth to 11th in the league as a result.

I sort of buy the projections' skepticism about Northwestern because they've experienced a lot of luck defending three-pointers. They give up a fair amount of them (181st) and opponents are hitting 27%. There are teams that can maintain opposition three-point shooing at low levels but they're usually dedicated zones or teams with incredible rim protection. Northwestern is neither.

As Ace detailed yesterday with an assist from Hoop Vision, there are reasons to believe that Northwestern's offensive turnaround is sustainable. They've mitigated much of their playmaking deficiency by using Pete Nance as a point-center in a five-out offense, a role that Nance is well-suited to. There is likely a reality check coming once teams scout the NW offense better, but Nance has proven to be a handful on the perimeter and should continue being one.

MSU… well, Ace just talked about them too but that was before last night's debacle.

[After the JUMP: a brief history of disaster factories]

Last night's debacle. MSU's defense has been talked about plenty, both in this space and others. Here's a bit more from Matt D:

Michigan State’s defensive issues are rooted in a lack of size/length in the frontcourt.

Izzo attempts to compensate by going small/quick with Hall at times but he’s getting exposed.

The Spartan Post Defense is very bad and I don’t anticipate it getting fixed. Simply too small. pic.twitter.com/jH3U9ng9D7

— LAbound (@LAbound2) December 29, 2020

This is not likely to be fixable since MSU's frontcourt isn't going to get bigger or more athletic unless Mady Sissoko has it click in the near future. Returns from the Minnesota game, when he got 12 minutes, weren't particularly encouraging.

This was sort of baked in and put a ceiling on Michigan State's potential this year. What has to be more alarming is MSU's ineptness on offense in Big Ten play. They got over 1 PPP against Wisconsin because they shot 8/11 from three; that anomaly is the only reason they've hit 1—a mediocre to bad number—in their three games so far.

MSU's two stars can barely find a two pointer worth taking in Big Ten play. Aaron Henry came off the bench and took seven heavily-contested midrange jumpers against Gabe Kalscheur.

He hit one. Watts was 1/5 in the midrange and found just one shot at the rim. Henry is shooting 41% from two in Big Ten play; Watts is shooting 16%(!). Collectively they are 14/59 from three on the season. Neither guy has the flair of Chase Audige or Both Gatch…

…but both are in contention for Disaster Factory POY based on numbers.

Meanwhile, Liam Robbins is legit. Minnesota scooped him up from Drake thanks to some connections on their coaching staff and he has translated extremely well. His rim protection against MSU resulted in 3 blocks, several more altered shots, and was a major part of MSU's 12/45 performance inside the arc. His block rate has dropped a bit from sixth nationally last year but he's still in the top 50.

On the offensive end he's maintained his rock-bottom-for-a-big TO rate and overwhelmed the MSU frontcourt. Despite being a transfer he seems like he's got a better hold of Minnesota's offense than some of his teammates. Like, yup, Gach. Here he appears to point out the backscreen his teammates need to run:

And-one layup with Foster Loyer looking around like Wile E Coyote after Minnesota gave him a diagram of what they were doing.

If Robbins's early surge in three point shooting is sustained, look out.

Way too early bracket items. I've been a little concerned that Michigan's nonconference schedule, which was heavy on good mid-majors but totally devoid of high-majors after the NC State game was cancelled, might end up hurting them in tournament selection and seeding. Early returns are encouraging in that department. Bart Torvik's "if the season ended today" tourney projection has Michigan as a four seed, which is at least two seed lines better than I thought was realistic given Michigan's schedule to date.

If the committee does put weight on their new NET rankings and appropriately judges the difficulty of going up against the tier of teams ranked around 10th, Michigan should be fine. Michigan ranks 13th in Torvik's pretty-good attempt to replicate the (proprietary and secret) NET and 12th in Wins Above Bubble. That "if" is in italics for a reason, and I kind of expect Michigan to end up slipping because the committee will look at the noncon superficially.

But we are likely talking about slipping a seed line and not being disastrously mis-judged.

Toledo: worth playing [Campredon]

BONUS SCHEDULING PHILOSOPHY ITEM: This nonconference schedule is an accidentally perfect window via which to judge the wisdom of avoiding buy games against the Houston Baptists of the world. Michigan's nonconference games are in a tight window between 90th and 154th on Kenpom with the exception of Oakland. Results:

• a series of comfortable victories, with the exception of Oakland
• reasonably good schedule strength despite the total lack of high-major opposition

That Oakland game was close largely because Michigan turned the ball over 20 times against Oakland's Bullshit Low Major Junk Zone™. More respectable outfits have run almost all man-to-man. In some ways games against BSLMJZs are like scheduling service academies in football: you can get a scare because the opposition is running something bizarre and playing that bizarre team does nothing to help you prep for the rest of the season. (See also: that Binghamton game from a couple years ago.)

If Michigan paired their buy game philosophy from this year with the usual number of high-major games they'd have one of the strongest nonconference schedules in the nation without upping their chance of a loss too much. If Michigan's able to level up here and recruit like they have in 2021 semi-consistently that only increases the argument for playing real teams in buy games.

GIMME GIMME GIMME THE BALL. A select group of teams have emerged as contenders for the Last First Dunk. They are:

• Tennessee-Martin (OVC)
• New Hampshire (America East)
• Presbyterian (Big South)
• Youngstown State (Horizon)
• Dixie State (WAC)
• Canisius (MAAC)

None of these teams have experienced the exhilarating thrill of sending it in, Jerome. Canisius almost did but 6'8" Scott Hitchon couldn't complete the deal against Monmouth. Previous Last First Dunk winner Elon is 2/2 on the season, FWIW.

On the other side of the ledger, only Coastal Carolina, Vermont, and Albany have not been posterized this season. New Hampshire might have some problems since the latter two teams are also in America East.

All* Time Disaster Factories. I got to thinking about where Isaiah Washington and Jacob Young stacked up in the annals of horrible high-usage players in Big Ten history, and punched in some numbers into Torvik's player stats page. Parameters: at least 12 MPG, usage over 20%. After it went beep boop a while it spat out Washington's sophomore year as the #9 and Young's junior year as the #14 worst ORTG season in the 12 years the Torvik database covers.

But wait! Some persons with terrible seasons are surrounded by terrible players and do not particularly stand out as overconfident because the rest of the team is just as bad as them. The all-time worst season by this reckoning belongs to Iowa point guard Jeff Peterson, who was a freshman on a 13-19 Todd Lickliter team that had the nation's #241 offense. None of the other starters had a 100+ ORTG, so while his 74 is indeed miserable there are many mitigating factors. Other guys at or near the very top of the list played for some of the worst teams in tempo-free Big Ten history—08 Northwestern (#191 in Kenpom), 09 Indiana(#209), 15 Rutgers (#198), 16 Rutgers(#279!)  etc.

Washington and Young are special because they turned in those seasons for teams that made the tourney (or would have, in Rutgers's case). Once you eliminate guys on teams that finished in triple digits in Kenpom offensive efficiency, our no-style-points Disaster Factory all* time top, uh, six:

1. Bill Edwards, Penn State 2010. Edwards was a freshman wing playing on one of those Penn State Guard Vs The World teams—in this case a junior Talor Battle—and managed to hit the 30/30/20 trifecta by turning in a 31 TO rate and shooting 39/29. When everyone on the other team was tripling Talor Battle. Edwards transferred to Miami (Not That Miami), where he couldn't get off the bench.
2. Kam Chatman, Michigan 2015. Ohhh nooooooo. Chatman got pressed into more service than he otherwise would have seen thanks to season-ending injuries to Caris LeVert and Derrick Walton; he shot 36/26. He got stuck on the deep bench as a sophomore, hit a game-winning three in the Big Ten tournament, and then transferred to Detroit, where he was the best player on a very, very bad team. Now he's playing in Belgium.
3. Kyle Rowley, Northwestern 2009. This was the Shurna/Coble Bill Carmody team that was a tourney near-miss at 17-14; they got an NIT bid. They had a top 50 offense; Rowley was a freshman post who checked in with an 80 ORTG because he hit 47% from two and had a TO rate of 28. Rowley transferred to St Mary's, where he couldn't get off the bench.
4. Tarrance Crump, Purdue 2008. This team had E'Twaun Moore and Robbie Hummel on it! Crump was a senior! He shot 47/15 with a 30 turnover rate! As a point guard! Purdue was 15-3 in league play!
5. Isaiah Washington, Minnesota 2019. The king sticks in the top five by shooting 36/21 on a team that finished just outside the top 50 in offensive efficiency.
6. Jacob Young, Rutgers 2020. The heir comes in next. His 26% usage is three points higher than anyone else on this list and his 44/25 shooting is even more staggering in that context. Young is the only one of these guys to evolve into a solid high-major player.

I do not remember any of the guys from the aughts. Given the context of the above guys I think Crump has the best chance of unseating our more recent champions based on style points, of which Washington and Young have scads. 

*[assuming that time started with the earliest season available on Torvik]