charles matthews

a master in chaos [Marc-Gregor Campredon]

Previously: Part One, Part Two. If you're looking for the Rutgers preview it's here.

You're definitely going to want to at least read part two of this series, which explains the stats I'm using below and details the 2009-14 seasons, before moving on to the rest of this post. Ideally, you'll read part one, as well.

Now that you're caught up, let's get to it.

2014-15: Bad Wheels

Team Stats: 27.7% pick-and-rolls + passes (#36 in country), 0.911 points per play (#62)

The Ballhandlers:

  P&R Plays (Own Offense) PPP on Own Offense (%ile) P&R Plays (Passes) PPP on Passes (%ile) Total P&R Plays Overall P&R PPP. (%ile) Keep %
Spike Albrecht 65 0.815 (70%) 98 1.276 (92%) 178 1.092 (93%) 36.5%
Caris LeVert 87 0.644 (35%) 58 0.862 (34%) 145 0.731 (28%) 60.0%
Derrick Walton 52 0.635 (33%) 61 0.967 (54%) 113 0.814 (47%) 46.0%
Zak Irvin 60 0.783 (63%) 43 1.395 (96%) 103 1.039 (90%) 58.3%
MAAR 39 0.872 (79%) 19 1.737 (100%) 58 1.155 (96%) 67.2%

The Screeners:

  Pop Plays (%) Pop PPP (%ile) Roll Plays Roll PPP (%ile) Slip Plays (%) Slip PPP (%ile) Overall Plays Overall PPP (%ile)
Max Bielfeldt 12 (36.4%) 1.167 (88%) 19 (57.6%) 1.000 (30%) 2 (6.1%) 2.000 (—) 33 1.121 (76%)
Ricky Doyle 1 (3.6%) 2.000 (—) 26 (92.9%) 1.308 (74%) 1 (3.6%) 0.000 (—) 28 1.286 (90%)
Zak Irvin 9 (69.2%) 1.222 (—) 4 (30.8%) 2.000 (—) 13 1.462 (96%)
Mark Donnal 1 (10%) 3.000 (—) 9 (90%) 1.556 (—) 10 1.700 (99%)

I almost didn't include this season or the next because of Michigan's injury issues, then decided it was useful to see what happens when a team's two best perimeter players get hurt in the same season.

While neither Caris LeVert nor Derrick Walton were producing particularly well in the pick-and-roll before their respective foot injuries, we saw later that these injuries delayed breakouts into effective players—Walton, in particular, eventually became a great P&R ballhandler.

The players that remained were effective but one-dimensional. Spike Albrecht drove to pass. Zak Irvin and Muhammad-Ali Abdur-Rahkman hunted shots off of screens. Irvin, defying reputation, struck the best balance between shooting and passing, and he was a very effective passer. Only MAAR was above-average at generating his own offense off of screens, though.

Derrick Walton's foot injury stunted a developing rapport with Ricky Doyle

Michigan was also working with a limited group of finishers. Ricky Doyle was the best roll man but was a roll man only. Max Bielfedlt(!) ended up with the most plays among screeners even though he was a 30th-percentile finisher on the roll; he salvaged decent efficiency with some pick-and-pop jumpers. If Zak Irvin was setting a screen, it was to pop or slip for a jump shot.

This marks the first season since 2008-09 that Michigan's pick-and-roll usage went down; they also slipped 40 spots in the efficiency rankings. This team was going to drop off with the departures of Nik Stauskas, Glenn Robinson III, and Jordan Morgan, then injuries made matters worse. Even if LeVert and Walton weren't high-level P&R ballhandlers at this point, their spot-up shooting could've helped.

Even with all that, Michigan's pick-and-roll offense ranked in the 83rd percentile by points per play. They weren't elite; they were still good. They just couldn't build the offense around it to the extent they had the previous year.

2015-16: Bad Wheels 2

Team Stats: 30.5% pick-and-rolls + passes (#22 in country), 0.923 points per play (#80)

The Ballhandlers:

  P&R Plays (Own Offense) PPP on Own Offense (%ile) P&R Plays (Passes) PPP on Passes (%ile) Total P&R Plays Overall P&R PPP. (%ile) Keep %
Derrick Walton 128 0.711 (44%) 120 1.000 (59%) 248 0.851 (51%) 51.6%
Zak Irvin 149 0.826 (68%) 98 1.306 (93%) 247 1.016 (86%) 60.3%
Caris LeVert 57 0.877 (77%) 62 0.855 (32%) 119 0.866 (54%) 47.9%
MAAR 67 0.910 (82%) 41 0.805 (24%) 108 0.870 (55%) 62.0%
Duncan Robinson 19 0.632 (29%) 17 0.647 (9%) 36 0.639 (14%) 52.8%

The Screeners:

  Pop Plays (%) Pop PPP (%ile) Roll Plays Roll PPP (%ile) Slip Plays (%) Slip PPP (%ile) Overall Plays Overall PPP (%ile)
Mark Donnal 12 (21.8%) 0.500 (12%) 40 (72.7%) 1.250 (60%) 3 (5.5%) 0.667 (—) 55 1.055 (60%)
Ricky Doyle 1 (2.9%) 2.000 (—) 30 (88.2%) 1.200 (54%) 3 (8.8%) 0.333 (—) 34 1.147 (73%)
Moe Wagner 3 (15.8%) 1.667 (—) 16 (84.2%) 1.375 (77%) 19 1.421 (95%)
DJ Wilson 9 (64.3%) 0.556 (—) 4 (28.6%) 1.500 (—) 1 (7.1%) 0.000 (—) 14 0.786 (24%)
Zak Irvin 9 (81.8%) 1.000 (—) 2 (18.2%) 0.000 (—) 11 0.818 (27%)

An unfortunate repeat, as Walton's previous foot injury sapped his ability to finish at the rim and LeVert—who'd improved considerably as a scorer off the high screen—again lost most of the season to a bad wheel.

Beilein increased the volume past where it had been in 2013-14 and the team's PPP slightly increased, though they came out worse compared to the rest of the country. Irvin was easily the team's best P&R ballhandler, continuing to pass at a high level while making enough pull-up jumpers to be relatively effective as a scorer.

some of those jumpers were rather important

MAAR pulled off a tough feat, averaging more PPP using his own offense than when he passed; that's very much a good news/bad news situation.

The roll men remained limited. This was the year Ricky Doyle seemingly lost the ability to catch and finish, so Mark Donnal ended up as the primary screener. Neither graded out particularly well. The center who did: enigmatic freshman Moe Wagner, who scored well as a roll man and flashed the ability to pop out and hit jumpers.

[Hit THE JUMP for Michigan exploring that a bit more.]

[Marc-Gregor Campredon]

This hit the feed today and caused a spiral of google searching and squinting at dense technical stats articles:

Naturally, you are thinking "what the hell is that?" if you didn't immediately log off. If this stat (BARAPM) adds any more components it's going to sound like you're firing a machine gun whenever you mention it.

[AFTER THE JUMP: wonkery and then an Isaiah Livers takeaway]

[Marc-Gregor Campredon]

3/28/2019 – Purdue 99, Tennessee 94 (OT) –

what

i mean did you see that game

look at this dude pointing at it

32548491047_198d48ee59_k

[Campredon]

it was really fun!

oh fine

3/28/2019 – Michigan 44, Texas Tech 63 – 30-7, 15-5 Big Ten, season over

I'm torn. On the one hand, walk-on CJ Baird crotching in a three in the desultory final moments of a 20-point game was somehow fitting. On the other, they showed a stat that it had been 261 games since Michigan had managed to go 40 minutes of basketball without hitting a single three, and that 0-fer in the box score would have been an even more powerful indicator of what happened than 1/19.

A collective mania set in as this was happening as the horrible results overwhelmed anyone's ability to process what happened before them. Four different threes rimmed out in the first half. A fifth was Michigan's first attempt, which was a blindingly wide open shot from Brazdeikis that barely grazed the front of the rim. Brazdeikis entered the game a 41% shooter from three, and did that with an uncontested catch and shoot look.

S9vfRzSVLUbJt7mlJuC3ZKTNO_IYQqtkwnqlRSV08l0

collective fanbase response to aforementioned shot

Michigan is a limited offensive team full of guys with major holes in their toolsets. Simpson can't shoot. Iggy can't pass. Poole can't stop oscillating wildly. Teske can't create his own shots. Matthews has some variety of all these issues. All these guys have assets that they managed to cobble together a top-25 offense out of, but there was a hard stop. So when they run up against a defense just as good as theirs they have a limited set of responses. When they run up against a team that's comfortable switching everything those responses narrow further.

When you are in this situation and literally do not hit a shot outside of the paint in the first 30 minutes you get run out of the building.

Last year's title game was against a very different team but was the same story. Michigan's offense did reasonably well inside the line, given the context (66% against Villanova, 50% against Texas Tech) and then had horrific, historically bad shooting from three (3/23 and 1/19) on looks that were more or less what you'd expect the opponent to give up. Tech did contest threes well; there were a couple of ugly stepbacks mixed in. But when Isaiah Livers rises up for a barely contested look in the corner where he's ~50% from and it rattles out to continue your 0-fer streak deep into the second half, there's nothing to say except "shit."

In a world where basketball consists of a million copies of every shot and you're awarded the average of your million trials, Texas Tech probably still wins this game. But, hell, hand Michigan 5 of their 18 pre-Baird attempts (28%, worse than Tech's season average allowed) and delete the banked-in prayer from Mooney and Baird isn't on the floor because it's a four-point game. The difference between a hard-fought game against an elite foe that aw-shucks you lost and last night's debacle is just shots going down or not.

Is there a reason that Michigan's last two seasons have ended in a flurry of bricks? To some extent, sure. I would kill and skin an entire herd of caribou for a shooter like Ryan Cline or Davide Moretti. The composition of this team leaves them vulnerable to nights where they can't hit anything. Versions of this column have popped up from time to time through the season. Michigan got in a 2 point game with Minnesota after going 3/22; there was nearly mass seppuku after a 3/19 night against Holy Cross.

32548471027_c8711862ad_k

nope [Campredon]

But also not really. Maybe there are reasons you go 25% from three. There are no reasons when you go 13% and 0%. Just frustration, and an offseason a little more sudden than hoped for. Michigan's started about five minutes into the second half. And while that sucks, Michigan basketball has never been in a better place. Don't let some bricks get that clouded.

[After THE JUMP: looking to the future]

man bad at crime 

Charles Matthews

All-other things

Charles Matthew and Michigan met halfway, and both are better for it.

in which the Galactic Points Bestower turns Michigan off for ten minutes 

HOOK ALL THE THINGS

in which John Beilein dons a hat 

if the dudes ever show up at the same place at the same time it'll be quite a crew 

a balanced performance holds off Rutgers, which is Legitimately Feisty 

Michigan doesn't have a backup 5, and that's bad 

they booed until they could boo no more