By the numbers: What’s wrong with Michigan’s defense?

Dylan Burkhardt

Michigan’s defense is broken. At this point, no one is going to debate that simple fact. The Wolverines have fallen nearly 100 spots in KenPom adjusted defensive efficiency metric, checking in at No. 185 at the time of publishing.

That’s 90 spots behind where the Wolverines finished the season last year and they have the worst defense in the Big ten by almost .1 points per possession.

What’s wrong with the Michigan defense? Obviously there are a lot of issues, but we broke down some of the numbers to take a closer look at where the Wolverines are struggling.

Pick-and-roll defense

Michigan’s pick-and-roll defense has completely fallen apart. In the last six games, the Wolverines have allowed .986 points per possession (including pass outs) in the pick-and-roll game. Compared to seasonal numbers across Division I, that would rank 336th nationally.

Is there an answer to Michigan’s pick-and-roll woes? There are many different ways to defend the ball screen, but they all give up something. Good ball screen defenses usually adopt a strategy that can funnel ball screens into one of their strengths. For Michigan, I’m not sure what that would be.

Michigan’s base ball screen defense has been a standard hedge, although in actuality it’s more of a show. Against Nebraska, the Wolverines tried to ‘ICE’ more ball screens, forcing the ball handled to reject the pick and using the baseline as an additional defender.

Going back through and watching some film from the last few games, it seems like Michigan’s issues are much more executional than strategic.

Whatever the Wolverines choose to do, they are making mistakes. There have been several instances where the ball handler splits Michigan’s hedge, others where the big man recovers late, but the majority of the issues are that the other 3 players on the floor aren’t getting it done either. If the help is there, it struggles to recover to pass outs and easy straight line drives. Other times, the help isn’t there  at the rim and results in an easy layup.


Thanks to Krossover, we can look at Michigan’s defensive shot chart from Big Ten games only. The immediate obvious concern is that teams are making a ridiculous number of their jumpshots against the Wolverines.

Michigan opponents are shooting better than the Krossover average in every three-point zone, which shouldn’t come as a surprise given that Big Ten teams are shooting 55.3% from three-point range against U-M.

Teams have a 62 eFG% on all contested shots against Michigan and a 64 eFG% on uncontested jumpers — meaning it hasn’t really mattered whether the Wolverines are almost getting a hand up or not. Jump shots are just always falling against the Wolverines.

Ken Pomeroy has written pretty extensively about three-point defense and whether the percentage allowed could have at least some factor of luck. When the same thing happens six games in a row, it doesn’t feel like luck. There’s no denying that Michigan’s doing a poor job of contesting shots, opponents feel comfortable shooting against the Wolverines, and opponents are making their shots. But I still think there’s at least an element of luck that most of these teams wouldn’t shoot 50% from three in a walk through if they had the opportunity.

Is zone defense a fix?

It’s a natural intuition: If man-to-man defense isn’t working… try a zone! Michigan has tried zone defense, it has promptly been shot right out of it.

The Wolverines have played 23 possessions of zone defense in 2016-17, per Synergy, and they’ve allowed 27 points.

Michigan actually got four stops out of the zone — all in the 2-3 — against Nebraska, but before that the last missed shot that the Wolverines forced out of zone defense was back in November at South Carolina. They also still gave up five baskets out of zone looks against the Huskers.

If Michigan is going to play a zone, I think the 2-3 zone is the answer. The extended 1-3-1 zone isn’t helping anyone because the Wolverines don’t have the length to force turnovers or the quickness to keep people out of the paint. It has been far too easy for an opposing point guard to dribble at the middle of the zone and slice right at the big man to the hoop.

Rule changes

The other dynamic at play is the continued effect of rule changes on the college game. Both Tim Mile, John Beilein and Michigan point guard Derrick Walton mentioned defensive rule changes in their post-game press conference on Saturday afternoon.

Walton focused on the tendency of referees to call fouls on the perimeter and how that is causing the Wolverines to struggle against dribble penetration.

“The new rules,” Walton said when asked about the difficulty of playing perimeter defense in the college game. “It’s kind of hard to guard guys when they’re coming at you full speed because everything is a foul. It’s hard to keep your man in front of you.”

You can certainly see this ‘playing not to foul’ mentality at times in Michigan’s defense. Redshirt sophomore DJ Wilson was one of Michigan’s better defenders and rebounders early on this year, but he’s seemed to regress on that side of the ball. He’s struggling to keep drivers in front of him and his rim protection has regressed.

Wilson has only committed 3 fouls in the last four games and his shot blocking and rebounding numbers have both taken a step back in league play. I don’t mean to single Wilson out because he’s the only Wolverine who lacks aggression on the defensive end, but because he’s the one player on the roster with the ability to emerge as a legitimate rim protection option.

Is there a solution?

There are no easy fixes, otherwise we probably would have seen them already.

Michigan has physical limitations on the defensive side of the ball, but there’s also no reason that the defense should be this bad. The Wolverines need to figure out their ball screen coverages and improve their rotations if they want the season to go anywhere.

“We’ve got to find these little steps to try and get better,” John Beilein said on Saturday. “Since the four-day break, (Wisconsin will be our sixth game). It’s hard to change your philosophy when you’re just trying to get rest.”

While Michigan might not be able to make sweeping changes in the 2 or 3 days between Big Ten games, it does need to figure out how it can make incremental improvement to its pick-and-roll defense.

Through three weeks of conference play, Michigan has the league’s best offense and the worst defense. A few notable gains on the defensive end of the floor and things could really start to fall into place. A few more games where the defense struggles to get a stop and the season will start to feel like a season that wasted a spectacular offense.

This site is supported by donations

Like what you see? Click the button below to donate and access exclusive content.

  • gobluemd16

    Great piece, Dylan. Really appreciate this thorough breakdown.

  • Mattski

    The question still lurks: why a FALL in defense with a largely veteran squad when you brought in a defensive specialist and are placing more emphasis ON D? Beilein hinted that the players may be overthinking the changes, but he’s a very candid guy; wish someone would ask him in pointed fashion. It’s possible it’s an “all of the above” kind of answer–lots and lots of minutes, not a real quick squad, a fog of new ideas not being executed, bigs who are not real strong. But even if that’s the answer I would like to hear Beilein articulate it.

    Was there even the tiniest uptick against Nebraska? I guess forcing them into quite a lot of turnovers might count. . .

    • tom48160

      Nebraska didn’t hit every three. That’s progress.

      Seriously, I think the players needed to see some misses to know that they aren’t completely snakebitten. Would we rather see the guys rally and impose their will more? Sure. But they needed that win in the worst way.

  • AC1997

    Here are some of my general thoughts on this subject:
    — Walton and Irvin have never been good defenders. Walton has his moments, but overall has been below average. Perhaps it is his size, quickness, so many minutes, or being afraid of fouling. But that’s two of your starters right there.
    — Wilson and Wagner are new starters this season paying two critical defensive positions. Neither has played much basketball before this season. While we’re thrilled at their offensive progress, it isn’t a surprise to see them struggle some on D.
    — Irvin is playing a new position at SF instead of PF this year. While that’s good for the team overall, it is a change in the type of defense he’s being asked to play compared to the past two years. (You could argue that Wilson has mostly thought of himself as either a 5 or a 3 before this season too – not his natural 4.)
    — I think asking them to change their defensive strategy does take adjusting to and they might be thinking too much. But I don’t understand how it has gotten so much WORSE as the season has gone on.
    — I think that Michigan, more than most teams, would benefit from someone like Mo Bamba playing center for them. It doesn’t solve the fundamental problems with the defense, but it gives you a safety net around the rim for blocks, altering shots, rebounding, etc.
    — Michigan lacks depth in a big way. At the end of last season everyone looked at the depth chart and wondered how there would be enough minutes to go around. Then Dawkins and Chatman leave. Then we learn that Watson, Teske, and Davis are not good enough to contribute in any way as Freshmen. Then we realize that Simpson has a long way to go. So you’re back to playing 7.5 players in your rotation, all of whom were here last year except Simpson, and none of them have ever been known as a good defender.
    — I think next year, even without Bamba, there is some potential to improve as the recruits are more athletic and this year’s freshmen will have a year of experience. At least there will be options.

    • Champswest

      Lots of good points, especially the next to last. The easiest and quickest way to get some improvement would be to give the starters more rest. The rotation guys aren’t as good as the starters, but they may be as good as tired starters. Good defense is hard work and it is harder to do when you are tired.

    • Mattski

      Really good post. In response to Champswest (below) I do think that it’s been tough to play the bench when so many games have been squeakers. And–for better or worse–we have not seen huge promise yet in Teske or X. While I think Teske can be a difference-maker in time, I still have my doubts about X as a scorer. It’s a serious liability if the guy at the top of the key is not a threat to put it in the basket.

  • A2MIKE

    I do think that Beilein is reluctant to change his ball screen defense mid-game. You saw Tim Miles do it 3 or 4 times in an effort to find something that works. Michigan finally did switch to an ICE defense on the ball screen and I thought they were more effective there. To me the biggest problem is the help defense is sooooo poor that using an ICE defense and keeping the ballhandler out of the middle accomplishes 2 goals that are also part of Michigan’s weaknesses. It limits the options as far as passing to an open wing when Michigan over helps and it forces a finish at the rim to be funneled to one side which should help Wagner and Wilson with blocks and rebounding. Just my 2 cents.

  • AA7596

    Interesting quote from Walton. It’s strange: Michigan was an ultra-low-foul team before the rules tightening and still is, yet the first thing on at least one player’s mind when asked about the defense is concern about fouling.

    Besides the physical and tactical problems on defense, there’s a big philosophical one about aggressiveness that Michigan has to sort out. Our foul/contact avoidance seems to be becoming counterproductive.

    • mikey_mac

      Something needs to give, and it’s on JB to make these fixes. Auto-bench is still a thing, yet the team is still fouling at super low levels. The defense is accounting for the changes in foul calls by defending drives on skates, but the offense doesn’t seem to be taking those same advantages.
      I don’t see coherent leadership here at all.

      • I think Beilein has been a lot more flexible with auto bench. Several times this year he’s gone back to guys with 2 fouls late in the first half.

        • mikey_mac

          Yeah, I’ve noticed that, too, and it’s a change, but it’s a conservative adjustment that still doesn’t match the foul aversion. Perhaps JB was also anticipating higher foul numbers as part of their new defense tactics, that just haven’t happened as the players haven’t bought in.

  • Barth Applefeld

    Nice piece as always, and the comments are great as alway too. The defense is disappointing, and the huge drop since B10 play has started is concerning, but the difficulty of adapting to a new defensive philosophy in general is much higher than often acknowledged. When the adapting is to be done by two veteran players lacking in quickness & hops, Duncan Robinson who defensively lacks everything physical except height, and two first-year starters (without great defensive credentials coming in), it can’t be underrated. It’s true that some teams/players would have adapted more quickly, but it just takes longer for some. Unfortunately, with some it never comes, but let’s hope it does for Michigan this year.