oooooh real scary, guy whose team lost to maryland at home [Marc-Gregor Campredon]

Big Ten Reset: Beat-Up Buckeyes, Oscillating Terps, Indispensable Players Comment Count

Ace January 11th, 2021 at 3:33 PM

At least until last night, this week was largely about the top end of the conference separating itself from the rest of the league. Notable scores from last week (home team listed second):

  • Maryland 55, Indiana 63
  • Rutgers 45, MSU 68
  • Minnesota 57, Michigan 82
  • Iowa 89, Maryland 67
  • Indiana 73, Wisconsin 80 (2OT)
  • Illinois 81, Northwestern 56
  • Purdue 55, MSU 54
  • OSU 79, Rutgers 68
  • Minnesota 71, Iowa 86
  • Maryland 66, Illinois 63

Iowa went an emphatic 2-0, Michigan blew out Minnesota, Illinois demolished Northwestern, and Wisconsin survived a scare from Indiana. Northwestern, Rutgers, and Indiana all had rough weeks; the only win among them was the Hoosiers righting the ship after blowing a big lead against last-place Nebraska.

Then Maryland went on the road to upset the Illini last night despite missing Eric Ayala, their only rotation player who somewhat resembles a point guard. The top four may be separating from the pack, but that pack is still dangerous.

We had more COVID-related postponements this week: Wisconsin at Penn State, Nebraska at Purdue, Penn State at OSU, and Michigan at Penn State did not happen as scheduled. PSU paused their program last week and their game against Rutgers originally slated for tomorrow night is postponed.

The Nittany Lions are going to have at least four games to make up when they're able to retake the court; they'll either face a brutally packed schedule to fit them in or the Big Ten is going to have an unbalanced final schedule unless the NCAA pushes the NCAA Tournament back, which they're considering—but more on a one-week-later timeline than a "May Madness" timeline.

The Standings

  Record   KP/Torvik Avg   OFFENSE   DEFENSE
Team Overall Big Ten Nat Rk (change) Proj. B1G Rec. KP Torvik KP Torvik
U-M 10-0 5-0 6.5 (up 2.5) 14-5.5* 6th 6th 18th 22nd
IOWA 11-2 5-1 5.0 (up 2.5) 14.5-5.5 2nd 2nd 72nd 111th
WIS 10-2 4-1 4.0 (down 0.5) 13.5-6* 10th 8th 7th 6th
ILL 9-4 5-2 8.5 (down 2) 13-7 8th 9th 17th 24th
OSU 9-3 3-3 21.0 (up 4.5) 10-9.5* 9th 10th 55th 95th
IND 8-5 3-3 22.0 (--) 10-10 51st 57th 15th 11th
PUR 8-5 3-3 36.0 (up 3.5) 10-9.5* 46th 62nd 37th 30th
NW 6-4 3-3 54.0 (down 2) 8-12 64th 49th 59th 57th
MIN 10-4 3-4 34.5 (down 8) 10-10 27th 41st 47th 51st
RUT 7-4 3-4 35.5 (down 18) 9-10* 45th 37th 42nd 48th
MSU 8-4 2-4 48.0 (up 5.5) 7.5-12.5 49th 58th 48th 50th
UMD 7-6 2-5 46.5 (up 0.5) 8.5-11.5 17th 15th 77th 108th
PSU 3-4 0-3 41.5 (up 3) 7-10.5* 20th 14th 78th 79th
NEB 4-8 0-5 106.0 (up 3.5) 3.5-16* 134th 126th 114th 74th

*Torvik includes projections for games that have been postponed, KenPom does not.

Teams are mostly settling into their spots in the advanced rankings pecking order, with the four top-ten squads, a couple tiers in the middle, and Nebraska hanging just below 100th. The exception is Rutgers, which has seen their offense plummet after a hideous loss to MSU and a less-hideous one to OSU. After looking like they might move into the top 15, they're now threatening to drop out of the top 40.

[Hit THE JUMP for trying to pin down Maryland, OSU's injury issues, the league's most indispensable players, and more.]

What Are You, Maryland?


Darryl Morsell (11) was the hero in MD's upset of Illinois [Paul Sherman]

Bart Torvik's site boils down a team's performance in each game into a "game score" (a concept adopted from Bill James and baseball's sabermetrics movement) that boils it down to a nice, easy-to-understand number; the scale is from 0 (awful) to 100 (amazing) and the average is 50.

Michigan has posted game scores between 89 and 99 in every game save Oakland, when they escaped with a win despite a 37, which passes the sanity test given the Grizzlies rank 280th in the country. Outside of that zone-induced, early-season hiccup, Michigan has been remarkably consistent.

The same cannot be said for Maryland, which has:

  • beaten Wisconsin and Illinois on the road
  • lost all five of their other Big Ten games

The schedule can explain some of it. It's tough to face Michigan and Iowa anywhere, even at home. There's no shame in losing at Indiana and Purdue. (The 14-point home loss to Rutgers isn't looking so good this week.) But game scores are adjusted for opponent quality and the Terps are all over the damn place. Each dot represents a game; the red line is their moving average, the black line their five-game moving average:

They played like one of the worst teams in the country against Clemson, a below-average D-I team against Rutgers and Iowa, and a Final Four-caliber squad against Wisconsin and Illinois. Notably, both the Badgers and Illini have one primary perimeter threat—D'Mitrik Trice and Ayo Dosunmu, respectively—driving a lot of their offense.

When Maryland can put shutdown defender Darryl Morsell on the other team's top option, they have a chance to play above their heads; Morsell was the main reason Dosunmu needed 25 shooting possessions to score 23 points last night. Even then, though, they needed help; they were getting crushed by Kofi Cockburn (21 points on 13 possessions) before the Illini stopped giving him the ball down the stretch and Da'Monte Williams missed a pair of critical late free throws.

The Terps seem exceedingly matchup-reliant but if one falls into their wheelhouse they can evidently hang with top-ten teams. Thankfully, Michigan has already shown they're not such a matchup.

Point: [Blank]


injuries may force Duane Washington Jr. to man the point [Campredon]

Point guard hasn't been a position of strength for Ohio State. It's at least been a position they had, though, and now they're stuck choosing between a Maryland-like all-wings lineup and fielding some potentially overmatched players after injuries struck down starter CJ Walker and replacement Jimmy Sotos.

Walker was shut down indefinitely after last Sunday's loss at Minnesota when a lingering hand injury turned out to be too serious to play through at all effectively:

The day after Walker scored two points while missing all five shots from the field with two assists and one rebounds in 23 minutes of Ohio State 17-point loss to Minnesota, he got his hand looked at. From those tests, they learned of the torn ligaments and made the choice to shut him down for the time being.

Sotos, a Bucknell transfer who's been a poor combination of low-usage and mistake-prone so far at OSU, started the following game against Rutgers, hit a couple threes, had four assists against only one turnover—and injured his shoulder in a collision late in the game. Chris Holtmann just called him a game-time decision for Wednesday's tilt against Northwestern, which would be an impressive turnaround given Sotos seemed to indicate he'd dislocated his shoulder.

Even if Sotos comes back, the Buckeyes need other lead ballhandler options. Starting shooting guard Duane Washington Jr. isn't a traditional lead guard by any means but he can create his own shots and guard the position; he'll likely see a number of his minutes shift to the point. Cal transfer Justice Sueing, a 6'7 wing, has 13 assists over the last four games and spent a lot of time with the ball in his hands down the stretch against Rutgers. It's possible Holtmann can use his current rotation to patch the hole.

He might not have many other options. 6'1 PG Meechie Johnson is a very young freshman; the #123 overall recruit enrolled mid-year after reclassifying from 2021, and he missed his entire junior season with a knee injury. He played four minutes without recording a statistic in his season debut against RU and Holtmann is talking about a "slow build" for him.

It's an unfortunate confluence of events for the Buckeyes, which added depth at the point in the form of Utah State transfer Abel Porter only for their medical team to uncover a career-ending heart condition in the offseason. Playing with no true point guard and an undersized center is a hard way to live in this conference. OSU is fielding such a team while facing one of the most brutal schedules in the conference from this point forward; they still have two games apiece against Illinois and Iowa, a road date at Wisconsin, and a final stretch of Michigan-@MSU-Iowa-Illinois. Oof.

The Indispensables


Myles Johnson is one of the B1G's best rim protectors when here's there to protect the rim [Campredon]

This section was inspired by the OSU-Rutgers game, which turned—just like their previous matchup—when RU center Myles Johnson got into foul trouble. Rutgers had just gone on a 13-2 run to take a six-point lead when Johnson got tagged with his second foul of the first half. From that point into the early portions of the second half, the Buckeyes went on a 27-4 run, and they never looked back. Frustratingly, Johnson only picked up one more foul the rest of the way.

This season, Rutgers has outscored OSU 77-64 with Johnson on the floor and been outscored 95-59(!!!) when he's on the bench. This inspired me to look for players whose teams have the biggest gaps in all-around quality between when they're on and off the floor.

A stipulation to prevent this from being too easy and center-focused: players in the KenPom Player of the Year top ten are excluded. That leaves out Iowa's Luka Garza, Indiana's Trayce Jackson-Davis*, Michigan's Hunter Dickinson, and Purdue's Trevion Williams.

These three are only a little less obvious but we are discussing indispensable players here. Despite inspiring the list, Johnson doesn't make it; if you take out the OSU matchups, Rutgers has been a significantly better defensive team against good competition when he's not on the floor. Your opinion of Johnson may depend on which night you caught him playing as much as any player in the league.

Marcus Carr, PG, Minnesota. The Gophers have been extremely reliant on Carr to be their offense. They're at their best when they're able to get out and run—or, at least, when Carr is able to get out and run, as he scores in transition so much better than his teammates that it's only partially an indictment of said teammates:

  Halfcourt Transition
  Poss Points PPP Poss. Points PPP
Marcus Carr 235 217 0.92 52 72 1.39
Other Gophers 767 683 0.89 136 127 0.93

That halfcourt number undersells Carr's value there, too. A lot of that Minnesota halfcourt production is coming off of his passes, which aren't included in Carr's figures above. He's also tasked with taking a lot of tough, late-clock shots when nobody else is able to create something. He's second in the league in minutes after finishing first last season.

The other point guard option is Jamal "Hair Trigger" Mashburn Jr., who's in the Disaster Factory of the Year conversation. Center Liam Robbins has largely gone missing against the Big Ten's better centers. This team goes as far as Carr can (sorry) drive them.

EJ Liddell, F/C, Ohio State. This one has to be a little worrisome for the Buckeyes; one of the most irreplaceable players is 6'7 and getting a lot of his minutes at center. They haven't played a team with a dominant interior presence outside of Purdue, which has one on the offensive side in Williams—who had 16 points and 9 boards in only 27 minutes in their win over the Bucks.

So, we'll see if this holds when OSU faces the Garzas, Cockburns, and Dickinsons of the conference. That all said, right now, the Buckeyes badly need their undersized stretch big on the court. Via Hoop Lens (cupcakes removed):

OSU's offense can stay afloat without him but their defense goes from good to atrocious inside the arc, allowing opponents to hit nearly 60% of their two-pointers. The combination of Kyle Young, Zed Key, and Seth Towns—the other three players to rotate at the center or power forward spots unless OSU is throwing out a rare ultra-small lineup with Sueing at the four—is providing almost no rim protection or interior defense. Liddell may not be tall but he's holding up a heck of a lot better.

Micah Potter, F/C, Wisconsin. One big question facing Greg Gard this season was how he'd handle his two bigs, Potter and 6'11 senior Nate Reuvers. So far, they've played a similar number of minutes, with each getting a little over half the available time on the court because they share the floor for about 20% of Wisconsin's non-garbage possessions.

The issue is that those two-big lineups, while great defensively, destroy UW's spacing and bog down the offense—it's a niche lineup, not one to be used for the long-haul. Meanwhile, while Gard is slightly favoring Reuvers over Potter when he plays one center, the numbers indicate he should be leaning the opposite way, and pretty hard at that.

Personnel 2P% For 2P% Allowed Off. Poss. PPP Def. Poss. PPP Against Eff. Margin
Both On 45.3 42.9 111 0.87 112 0.74 0.13
Potter On/Reuvers Off 52.3 41.2 185 1.25 182 0.93 0.32
Reuvers On/Potter Off 43.5 47.7 218 1.03 213 0.97 0.06

Potter has been a significantly better offensive player, both individually and functioning within the team, and his individual defense is good even though he doesn't amass blocks like Reuvers. Wisconsin's worst defensive group, in fact, is when Reuvers is their only big man.

If there's an adjustment coming, it wasn't on the way last time out, when Potter appeared frustrated in the huddle and saw only 24 minutes in a double-overtime game—hey, maybe it wouldn't have reached that point if he'd played more—despite Potter going 5/7 from the field with four boards, one turnover, and only two fouls. I'm not sure what's going on between Gard and Potter but I hope it continues through Tuesday night.

[*I need to get this TJD on/off stat in here anyway. With him on the floor (cupcakes removed), IU scores 0.99 PPP with a 50.8 2P% while allowing 0.93 and 43.6% for a +0.06 efficiency margin; when he's off the floor they score 0.83 PPP (WOOF) with a 43.8 2P% while giving up 1.12 and 57.1% for a -0.29 EM. Holy hell!]

Updated Tiers

From last week:

Tier I: Wisconsin, Michigan, Illinois, Iowa
Tier II: Rutgers, Minnesota
Tier III: Indiana, Ohio State, Northwestern, Purdue, Maryland
Tier IV: Michigan State, Penn State
Tier V: Nebraska :(

Updated:

Tier I: Wisconsin, Michigan, Iowa, Illinois
Tier II: Ohio State, Purdue, Minnesota, Rutgers
Tier III: Indiana, Northwestern, Maryland
Tier IV: Michigan State, Penn State
Tier V: Nebraska :(

This Week's Games

All times Eastern.

Tuesday: Wisconsin at Michigan (7 pm, ESPN)
Wednesday: Northwestern at OSU (7, BTN), Illinois at Nebraska (9, BTN)
Thursday: Purdue at Indiana (7, FS1), MSU at Iowa (9, FS1)
Friday: Wisconsin at Rutgers (7, FS1)
Saturday: OSU at Illinois (noon, FOX), Nebraska at Maryland (noon, BTN), Michigan at Minnesota (2, ESPN2)
Sunday: Indiana at MSU (TBD, CBS), PSU at Purdue (TBD, BTN)

Comments

matty blue

January 11th, 2021 at 4:11 PM ^

as someone who bought bill james' very first mass-market baseball abstract in 1982 (shut up, you children) i love the notion of a game score for hoops...but this isn't quite as elegant as james' version.

his was crazy-simple and could be calculated in your head in about ten seconds using an old-timey box score (i.e. the agate-type version you'd find in a local newspaper - the muskegon chronicle, in my case). this version appears to need offensive and defensive efficiencies on a game-level basis, from which you calculate a pythagorean expectancy and oh god, where's my hp-30 calculator?  and, if i'm reading it right, a score can change over time, as the teams progress later in the season?

i mean, it's fun, and not totally un-useful, and i look forward to seeing more of it.  i don't mean to crap on it.  i just want to be able to figure it myself without a slide rule (not to the youths - a slide rule is a...oh, never mind).

Ali G Bomaye

January 11th, 2021 at 5:00 PM ^

One of the reasons baseball was the first sport to have sabermetrics is because baseball consists of a few largely discrete actions and outcomes, which makes it easier to measure what has an effect on scoring runs and winning. Basketball doesn't, which makes any all-in-one metric significantly more complicated.

matty blue

January 12th, 2021 at 7:51 AM ^

oh, absolutely, and i'm not a michael wilbon-type "all analytics are junk" guy.  on the other hand - and this kinda goes to your point - basketball fancystats, because they're newer, have (to my less-elastic-than-it-used-to-be brain) are less accessible at least in part because they have less context and history.  some - ORtg, win shares - are easy to grok.  others less so.  it took me a while to get my head around usage, for example.

which goes back to game scores.  the basketball version doesn't seem to follow normal distribution as much as i'd like - i know we're playing great right now, but all those games over 89 seems...put it this way - max scherzer has had some towering single-game performances over his career, and has a total of SIX with a game score over 89.  down the list, on june 12 2014, against the white sox, he threw a complete game, three-hit shutout with eight strikeouts and three walks.  a brilliant performance...with a game score of 84.

i understand that, again, this is just me learning how to process the number, and also me being "why can't this be like that" guy.  just get off my lawn and we'll be fine.

Sambojangles

January 12th, 2021 at 12:31 PM ^

Is it possible that the Bill James Game Score calculation is/was too simplistic, and should not be idealized? I think (without much evidence because I don't know all the details behind the calculations) that efficiency, opponent quality, and all the other additional factors, actually improve the utility. Like you said, it takes some time to get comfortable with newer stats, but that doesn't mean they're not improvements over stats created in a pre-computer world.

There are two other issues with comparing the Bill James pitcher game score with the Bart Torvik score: 1) the baseball one is only for one player on one side, while the basketball version relates to a team's performance on both offense and defense. 2) Baseball isn't timed, while basketball of course is, so it makes sense to normalize/adjust basketball stats for tempo and efficiency. Scoring 25 points means more in a game score in the 60s vs 80s. And scoring 30 points as a player doesn't necessarily help your team win if you're cannibalizing possessions and scoring opportunities from teammates. So basketball obviously has to account for that increased complexity vs baseball, which is a relatively straightforward "out-or-not" construction.

NittanyFan

January 13th, 2021 at 12:02 AM ^

You are not wrong.  The Bill James Game Score was simplistic.  James himself initially referred to it as a "garbage stat."

But --- it was both (1) accessible, (2) intuitive and (3) directionally informational.    

(1) accessible --- any 5th grader could calculate it in his head, using only a newspaper box score.

(2) intuitive --- 0 to 100 scale (though sub-zero and 100+ was theoretically possible).  Empirical data fit a normal distribution fairly well.  Jack Morris has a game score of 50 - well, he didn't really increase nor decrease the Tigers' chances of winning that particular game.  70 - the Tigers are looking good to win.  90 - one of the season's top pitching performances.  100 - one of history's top pitching performances.

(3) directionally informational --- definite correlation between better game scores and better advanced metrics.

I concur that advanced metrics are better, but I still think there is value in metrics with each of the 3 above attributes.