obligatory [Marc-Gregor Campredon]

Make A Damn Shot Or The Puppy Gets It Comment Count

Brian January 27th, 2020 at 1:57 PM

1/25/2020 – Michigan 62, Illinois 64 – 11-8, 2-6 Big Ten

Ha ha no not really. I would never threaten a puppy just because my favorite basketball team has been hit by a frying pan over and over again. I find this all funny, actually. I want you to know that.

I am personally unaffected by Michigan basketball's month long run towards a football held by a Lucy so diabolical that it put Franz Wagner, to date an 87% free throw shooter, at the line late. Wagner clanged both free throws to cap an 0/5 run on FTs. Ayo Dosunmu canned a tough runner over Simpson, his 10th make on 4 attempts in the midrange. Loss.

Dosunmu, a 29% three point shooter entering the game, hit two of three with both makes off the dribble.

I am not collecting these facts so I can print out a brick worthy of brandishing at the sky. I do not plan to rend my garments and moan "whyyyyyyyyyyy?" as I brandish the brick of outrageous items (Trent Frazier hit two threes from thirty feet! Michigan had two turnovers and lost!) at the sky. I think this is all funny.

Michigan had ten offensive rebounds and two turnovers and 21 free throws and did not score a point per possession. I am not looking up proper usage of the word "yeet," which appears to be something that addled teens use to beg powerful entities to erase them from existence. All of this is part of life's rich pageant. One must take the good with the bad.

Isaiah Livers lasted 20 minutes before appearing to re-injure his groin in a potentially season-crushing moment. I am… AAAARGH FUUUUUUUUU-

[After THE *UMP: still no letter that comes after i]

----------------------------------------------

Back in the day I noted that in football turnovers have close to no year to year correlation. This has become a predictive tool across the section of college football writing. Phil Steele notes it ominously for teams that did very well a year ago and positively for teams that were on the wrong end of the spectrum. S&P+ has a whole section of its box scores dedicated to turnover luck.

So I kept saying this for Rodriguez-era season previews, and Rich Rodriguez teams kept finishing the year double digits in the red. Michigan was -12 in 2009 and -10 in 2010. Both marks were dead last in the league. Brady Hoke came in with a horseshoe up his butt and finished +7 in 2011. Change. Hoke practices this stuff. Yeah.

In 2012 Michigan was right back down at –9. Yeet me into the sea turnover daddy(?).

In a similar vein this year, Michigan running backs built to a crescendo of miserable fumbling that saw Hassan Haskins emerge from the nether regions of the depth chart largely because he hadn't accidentally punted a baby out of a third floor window. The Michigan discourse was littered with despair and rage at how the coaches could let this happen, and then it stopped. Michigan running backs did not fumble after the Illinois game.

Weird shit happens all the time and we are very badly programmed to cope with it. NBA person Tim Hardaway the Younger's career three point shooting marks at Michigan: 37%, 28%, 37%. What the crap.

So yeah I still maintain that this stretch is arm-wavingly terrible luck and not much else. Since Big Ten play resumed Michigan is shooting 26% from three while their opponents shoot 40%. They rank 332nd in the country in both those metrics. Exactly one team has a bigger gap from behind the line in that time period: Incarnate Word.

Some of both can be attributed to Michigan breakdowns, or the fact that Michigan got six attempts from Jon Teske against Illinois and seven from Zavier Simpson against PSU. But Michigan is not the 349th best team at generating three point scoring differential or anywhere close. And closing that gap to something crappy but not absurd puts Michigan at .500 in Big Ten play.

This isn't a great team but the reason they've tailspun onto the bubble instead of treaded water is they can't make a damn shot. Also one member of the opposition is inhabited by the vengeful ghost of Reggie Miller, who has read all our comments about him during the NCAA tournament and can now turn into a ghost on command. At least this time it was Ayo and not a random scrub that the even opposing teams fans are asking "who is that?" about.

It is in these ways Michigan has collapsed.

Make a damn shot!

49439871028_f2985ae389_k

Are you ready to see several of these guys against Nebraska? [Campredon]

NOT GREAT BOB. The official site may have accidentally spilled the beans on the Livers injury:

That would take him out for a big chunk of time. The official word is that it's something else and not serious, but when it comes to athletic departments discussing injuries the paranoid thought is usually the right one.

ALSO NOT GREAT BOB. Simpson has been suspended for the Nebraska game. U release:

"While we are disappointed with what has transpired with Zavier, we know there are always lessons to learn and grow from. We take these matters and consequences very seriously," said Howard. "Moving forward, we will continue to handle this matter appropriately within our program and basketball family."

With Livers presumably also out that would make Adrian Nunez Michigan's only backup 1-4 unless Cole Bajema gets unearthed.

49440446281_77dcb6593b_k

doing the things he wasn't and not doing the things he was [Campredon]

Speaking of walking into the sea. Franz Wagner seemed like he'd found his shooting touch. After a rough stretch to start post wrist injury he went 14 of 32. Since: 4 of 22. He's back under 30% for the year.

This is one of the season's bigger disappointments. Wagner's data from last year isn't a huge sample size but this almost has to be an outlier. He was 48/125 from three across all competitions, a 38% hit rate from the FIBA line that college basketball just implemented, and hit 89% of his 65 FTs. (FT% is fairly predictive of 3P%.)

Wagner is a shooter and has to keep shooting. Hopefully he'll bounce back up.

The good news, such as it is, is that Wagner's flashing the ability to do more stuff. He got eight shot equivalents inside the line and would have been extremely efficient there if he'd shot free throws like he usually does. He generated most of those shots himself; he had a couple of offensive rebounds. And he continues racking up defensive stats. Ace pointed this out before but there's a ton of NBA on Wagner's list of Kenpom comparables despite his indifferent shooting; we thinks that's a combination of FT% and Wagner's block and steal rates implying a good shooter with freaky long arms.

49440687782_5c7f2e3827_k

[Campredon]

Post defense! Kofi Cockburn went 2/9 and Giorgi didn't hit a shot from the floor; Giorgi also got worked by Teske in the post. Cockburn was probably unfortunate to have a couple of good looks rim out. It was still a win on the interior especially since Michigan limited Illinois to 5 OREBs. Major turnaround from the first game.

Putting Austin Davis in when Illinois went dual posts late didn't make a ton of sense to me but with Livers out again Michigan's options were limited. Brandon Johns was in ghost mode in this game; 18 minutes, 0/1 from the floor, 3 rebounds, 1 block. I want Johns to be clearly a better idea than Austin Davis against Bezhanishvili but I'm not sure that was the case.

49439964058_5c70a12ca7_k (1)

Prince style [Campredon]

The other thing. Per Torvik, midrange shooting from Michigan opponents in 2020:

  • Illinois: 10/23
  • PSU: 10/26
  • Iowa: 5/13
  • Purdue: 14/40
  • Minnesota: 4/11
  • MSU: 11/21

That's 40.4 percent. This does include a number of possessions that aren't long jumpers—Trevion Williams was mostly taking other twos—but that's a number that would be in the 300s if it was a season-long one. Other two percentage appears to be a stat heavily segregated into high and low major tiers: only one high major team (Oregon State) is worse than that on the season.

A two for one isn't always the best thing in the world. Some controversy about the end of the game: Michigan got the ball in a tie game with 50 seconds left and did not get a two for one. I was fine with it since Michigan's plan was a good one: post Teske on Bezhanishvili. That matchup had been a series of bunnies for Teske and Teske again backed his man into the restricted circle for a three foot hook shot that goes down way more than half the time.

Getting there took a repost and a lot of time but the EV of the shot they got was worth giving up short possession IMO. It's at least close enough that it's not obviously wrong.

Comments

mgobaran

January 27th, 2020 at 2:51 PM ^

When things are flying high we tend to ignore the things we tell ourselves when things aren't so good. 

"It'll level out"

At the end of the Gonzaga game we were hitting 42.4% of our 3-point attempts on the year. That number would be good for #1 in the country right now, by a full percentage point. The team who is number one (South Dakota) shot 33.9% from three last year. 

"It'll level out"

Since the Gonzaga game where we shot 52.2% on 23 threes, we've shot under 30% six times, and under 20% three times! In 12 games!

"Maybe this is us leveling out"

On the year we are at 34.4% from three, ranking 116th out of 353 schools, 2 slots behind MSU, in the top third of all of basketball. 

Since the B1G started, we've sucked, sure. But we've been pretty good at generating open 3s. Seems like we are just paying for the hot shooting we had before the conference schedule started. 

MGoStretch

January 27th, 2020 at 3:09 PM ^

This has also been my question throughout all this "leveling out".  At what point do we start to view the strong shooting as the aberration?  And the recent displays of bricklaying as reflective of reality?  Similarly with other teams having multiple players going bonkers against us and having career nights.  At a certain point, people playing out of their mind against our defense starts to become a trend that points to a systemic issue and not just bad luck.

TrueBlue2003

January 27th, 2020 at 4:04 PM ^

The strong shooting was absolutely an aberration. It was obvious at the time.  The question then and now was: where does this team level out?

Unfortunately I think it's closer to the bad than the good for a couple reasons:

1. Livers is out.  Not hard to do the math on losing by far the best three point shooter on the team and the floor spacing that comes with it.

2. Teams are now scouting and prepping a lot more than they were/did during the bahamas when data was thin and prep time was non-existent.  They're pushing more usage towards Teske and Wagner and Johns and away from Brooks and DDJ (at the three point line at least).  We sit here and be like argh those are open threes but other teams are giving Teske open shots because those are bad shots for the offense. (I still think Wagner threes are good shots but until he proves he can hit them at a higher rate, I don't think opponents are too concerned about him). 

mgobaran

January 27th, 2020 at 4:21 PM ^

As all things, it's definitely more than one factor. It's more than "JUST MAKE SHOTS" like Brian's headline points to. The roster isn't set up to be an offensive juggernaut, and the defensive strength of this team has been nullified by a few factors. Transition cost between systems. Loss of one of the best defensive minds in basketball. Loss of Matthews & Livers. Offensive usage (energy expended) increased for Teske & Simpson.

It's a transition year. It would have been under Beilein, replacing 3 starters. It's even more so with the coaching changes. It's okay to be grumpy about it, but I don't think it's fair to pin this on Howard, or say it's a reflection of what his program will be. (Not saying you're saying that either, just wish the sky is falling crowd would stick to football). 

MGoStretch

January 27th, 2020 at 5:08 PM ^

Agree on all accounts, hopefully my post didn't come across as a Juwan hatin' or sky fallin' sort of deal.  It was more a critique of the, "Welp, statistics, what are you gonna do? We're just unlucky shooting and defending" analysis in the front page post.

michgoblue

January 27th, 2020 at 2:34 PM ^

Sometimes, a team is just not very good.  We have been spoiled by the last decade under Beilein, but unless you are one of a handful of teams, every so often, you are going to have a season like this.  

Like ours, usually that will come as a result of injury to an already thin team.  Often, it comes after a coaching change.  

The unfortunate reality is that this roster is not exactly built for a great season.  Teske (recent regression aside) is a great center, but other than him, we are relying on a true freshman from Germany who is going to need time to acclimate to the college game, a senior point guard who provides zero threat from the outside, and some pretty raw younger guys who all look like they are a year away.  And our best scorer has been out for almost the entire downturn.  Add to that the existing players learning a new system, and this is just going to be a down year.

The good news is that if this year's class stays intact, we will be well set up for next year.  Livers will almost definitely be back, an a bunch of the younger guys like DDJ, Johns and even Brooks will get some valuable in-game experience.

Unlike football, in which it takes a few years to turn a program around, because of the prevalence of one-and-done (or two and done), a basketball program can get on track relatively quickly if recruiting is goinf well.

mgobaran

January 27th, 2020 at 3:03 PM ^

Michigan 2012/13 - Lost National Title Game
Michigan 2013/14 - Lost in Elite 8 to a great team
Michigan 2014/15 - 2 Starters get drafted in the NBA, Best remaining player plagued by injury all year, missed Tournament

Michigan 2017/18 - Lost National Title Game
Michigan 2018/19 - Lost in Sweet Sixteen to a great team
Michigan 2019/20 - 2 Starters get drafted in the NBA, Best remaining player plagued by injury all year, Tournament status: Bubble

This is still a Beilein talented team, at a natural lull in the cycle of his roster construction/development. 

michgoblue

January 27th, 2020 at 3:38 PM ^

Exactly. Beilein teams have talent, but when they talent goes to the NBA, there isn’t always a plug and play replacement. In an era of players consistently declaring early (and before many fans think is optimal because there is still more development to be had on college for said player), rosters made up largely of high-level juniors and seniors are increasingly rare.  Generally, the college  shelf life of a great player is 1 season, and for a very good player it’s 2. If Juwan’s early recruiting is an indication of how he will recruit year after year, then we will be better set up to avoid these ups and downs. 

bronxblue

January 27th, 2020 at 2:43 PM ^

This was brought up during the podcast and I thought of it as well, but the problem with averages is that they aren't necessarily evenly distributed.  Michigan's a 36% 3pt shooting team this year but it came in bunches; they shot 47% against UNC, Iowa St, and Gonzaga, 16% against (EDIT:  Louisville), 42% and 34% against Iowa, 43% (!) against Oregon, and 20% the last two games.  My guess is they'll get "hot" again and rattle, I don't know, Rutgers for 42% from 3 and then shoot 9% against Wisconsin or something.  I don't really know how, as a coach and team, you can smooth out those deviations.  But it's weird watching this team struggle to hit a 3 against PSU or Illinois (on decent looks) and remember only a week or two ago they were hitting 36% against Minnesota on the road.

Part of me wonders if this team will be like a lot of those Beilein teams that sort of make a run at the end of the year, sneak in at the bottom of the bubble, and look really good on their way to a second-round loss in the tourney.  Because the shots are there and, like the fumbles with football, likely will start going back to the mean sooner rather than later.  

Zenogias

January 27th, 2020 at 3:00 PM ^

But that's the thing with randomness: it's not at all evenly distributed. And when it isn't, our brain immediately latches on to the perceived pattern and turns it into a story.

Could the explanation be something other than just randomness? Of course. The problem is that we'll never get a sample size large enough from one season of basketball to substantiate it statistically, and any non-statistical explanations are all colored by the fact that our brains are just dying to attach a story to the noise. The brain loves just-so stories. The amount of rigor necessary to find and support a story without statistics and without the bias our brains want to introduce is gargantuan.

And so in the absence of a statistical story or a deep, convincing, hard-to-produce, unbiased, non-statistical story, all we can do is fall back on the evidence we have about basketball in general, and that evidence is that what we've seen is almost certainly unsustainable. That's the good news.

The bad news is that, as good Bayesians, we also do need to lower our expectation of how good a shooting team this actually is. Of course, these are small incremental updates, not drastic ones, but we do need to slowly be walking back our expectations.

This feels horrible, of course, because the human brain is wired to find explanations for problems. We all want answers! Alas, we are unlikely to find any for the distribution of our three point shooting percentages.

And anyway, it's not as if the team doesn't have a lot of other stuff to work on. For me, it's almost not even worth worrying about our three point percentage, because if we really *are* actually as bad at three point shooting as we've been recently, nothing else really matters. Might as well work on the stuff that's highly likely not to be just random and hope the probably really random thing sorts itself out.

bronxblue

January 27th, 2020 at 3:18 PM ^

I won't argue that randomness doesn't play a role here, only that Michigan's shooting profile doesn't look demonstrably different than in past years (their 3% variance is very consistent with the past couple of seasons), but the big difference is that they don't seem to have ways to generate offense in other ways.  They don't shoot a ton of FTs (outside of Iowa fouling them 16 times in the last 3 minutes in their first game, they only cracked 20 FT attempts 2 times against major opponents during regulation) and while their 2% is solid it's not other-worldly.  

I guess what I'm trying to say is that I'm cautiously optimistic, even with Livers hurt, that they'll figure out their shooting a bit.  They won't be amazing at it by any means, but shooting 34-37% from 3 as a team certainly should be expected, which is in line with the past couple of seasons.

Zenogias

January 27th, 2020 at 3:29 PM ^

OK, then we're both basically on the same page. I also am cautiously optimistic about our shooting, even though we can't expect it to be great.

I would also note that there's more to the randomness than the exact shooting percentage and its distribution. The same distribution can yield different results depending on a bunch of other factors. That's why it's been so galling that we've paired these two horrendously bad shooting efforts with two otherwise winnable games, both featuring better than expected shooting performances from the opposition. Shoot poorly at Louisville or at Michigan State and you just lose a game by more than you otherwise would have lost it; shoot poorly at home on the same night your opposition is banging home 30-footers and you lose games you otherwise would have won. These were two pretty bad games to go ice cold.

trueblueintexas

January 27th, 2020 at 3:40 PM ^

The announcer said it best on that last shot, "the only thing you could do to guard that better is grow a few inches".

Michigan has three guys (Simpson, Brooks, DDJ) which any opposing guard over 6' 3" loves playing against. I know Simpson has the reputation of being a tough defender, but if you know you can rise up and over a defender at any time it really increases your confidence. Michigan needs 2 of these 3, and sometimes all 3, on the floor at all times. This is why it seems some opposing guard is having a career day every game. They are. And this will continue because someone on the opposing team will always have a significant size advantage this season as long as Livers is out. 

I would have put Wagner on Dosomu (sp?) on that last play because it would have forced him to drive into traffic (which he had proven he could do), but it still would have forced him to make a decision or involve other people. Instead, Simpson played his great D but could do little to actually stop the shot other than foul. That said, even putting Wagner on Dosomu would have left at least one other mis-match available to find and Illinois had plenty of time to run a play to do so.

harmon98

January 27th, 2020 at 3:50 PM ^

That picture of Livers grimacing whilst clutching his failed groin is the antithesis of those motivational posters thumb-tacked in so many managers' offices.

mrjblock24

January 27th, 2020 at 3:59 PM ^

Going to go ahead and strongly disagree with your take on the 2 for 1. It’s an obvious decision the gives your team at least two chances instead of only one... and you get the last shot knowing how many points you need to either tie or take the lead

username03

January 27th, 2020 at 5:02 PM ^

It's only an obvious decision if the combined chances of one of the two shots going in is better than the chance of the 1 shot going in, this is entirely dependent on the quality of the shots in question. One dunk/layup is better than two half court shots. Unless you have access to an alternate universe, its impossible to know the answer in this specific situation.

username03

January 27th, 2020 at 9:14 PM ^

Yes, I picked an extreme scenario to make my point. I'm confused though because you go on to confirm my point when you say two quality shots. With very few players on this team that can create their own shot, one of which can't shoot, I don't see how it's a given that they would get two quality shots in 40 seconds. If Trey Burke was still our point guard, absolutely go two for one. With Xavier, it's at the very least debatable.

footballguy

January 27th, 2020 at 9:17 PM ^

I just don't understand the logic behind thinking that a 1 for 1 could give Michigan a better look than both shots on a 2 for 1. 

All they did was toss it down into the post after doing nothing for the first 15 seconds of the clock. 

It was a bad call and bad situational awareness. I get your point, but it just doesn't really apply here.

 

username03

January 27th, 2020 at 10:00 PM ^

I think that is because you aren't thinking realistically about what those two shots would look like. Who is going to create and take two quality shots in a short time frame?  Most likely you're getting two non-layups from Simpson and I'm just not sure that is a higher percentage play than the three footer we got from Teske. You really don't think it's at least debatable?

MHWolverine

January 27th, 2020 at 4:27 PM ^

This season kinda reminds me of the 2014-15 season when Caris was hurt and lost for the season.

That team could never recover from that and did not even make the tournament which I'm sure you are all aware of.

They can still turn it around if they could just hit some open shots!!!!!

jbrandimore

January 27th, 2020 at 6:04 PM ^

The thing I do not understand about basketball analytics is the fact it seems to rely entirely on the proposition that all ten players on the court have exactly equivalent skill at everything they do.

This leads to the (self evidently false to me) theory that you can't do anything about 3pt percentages on either end of the court.

Yet, when teams make Teske and Simpson take our 3 point attempts, magically we suck as a team making them.

And analytics seems to suggest this is entirely random and nothing to be concerned about as it will "regress to the mean."

Color me a skeptic on that.

And one might ask if analytics is right, and it doesn't matter who takes your 3 pt attempts, why no one ever covers Simpson or Teske when they are in an area to take them?

footballguy

January 27th, 2020 at 9:22 PM ^

Their general view about the shooting is wrong. They keep saying it's luck, and it's just not. 

The looks given to DDJ, Brooks, and Franz are so much worse now than the beginning of the year. And zavier, Teske, and Johns are seemingly wide open on their takes.

The shots aren't in a vacuum. We're getting good looks from bad shooters, and bad looks from good shooters. And magically we're not shooting well any more.

jbrandimore

January 27th, 2020 at 10:19 PM ^

I can understand a thought that shooting - both for and against - might have a mean to revert to, but that doesn’t mean the analytical view still isn’t completely wrong.

A valid analytic might be “percentage of 3 pt shots taken by the opponents worst 2 shooters on the floor at any given time” as a plus, and “3 pt shots taken by the opponents best 3 point shooter” as a minus. Reverse this for offense.

ca_prophet

January 27th, 2020 at 7:45 PM ^

As always, much of this is random, because sports:

https://xkcd.com/904/

ReqQuibble:  Football turnovers aren't strictly random.  They are correlated with personnel, and in the NFL the relationship can be seen more clearly:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5969004/

http://harvardsportsanalysis.org/2014/10/how-random-are-turnovers/

On the other hand, fumble *recoveries* are almost entirely random.

In college that relationship is less true, because personnel change much more frequently and there's a wide variation in talent anyway, so turnovers correlate less well from year to year.

The problem with football turnovers is that, even if they are about half skill and half luck, is that turnover differential explains about 40% of W-L records in the NFL:

http://harvardsportsanalysis.org/2014/10/how-random-are-turnovers/

Put another way, when talent is (more) equalized, 1/5 or so of your season will come down to blind, dumb luck.  

MadMatt

January 28th, 2020 at 9:21 AM ^

Which is why our abominable record against OSU in the Game since 2004 is so whacky. Even with the Schmuckeyes being clearly superior in most of the games, the fact the breaks have added up for a win only once is simply remarkable. Problem is the frequency of games is so small, we can be waiting a VERY LONG TIME for regression to the mean to help us out. [Sigh]

Go Blue 80

January 28th, 2020 at 7:19 AM ^

The problem with the bad luck theory that wasn't mentioned is the psychological deflation that can occur for a team and can lead to ongoing confidence issues and poor play.

93Grad

January 28th, 2020 at 9:51 AM ^

Sorry but this just isn't a good shooting team without Livers.  So having them continuing to shoot 3's in hopes they make them seems like a bad plan.  

michgoblue

January 28th, 2020 at 12:47 PM ^

While I agree that they are not a good shooting team, the problem is that the current roster is not good at any other aspect of basketball offense. 

There are 4 primary ways to score points:  (1) driving; (2) 3 point shooting; (3) inside post game; and (4) midrange shooting.

Let's agree that this team is bad at 3-pointers.  Unfortunately, they are also bad at driving to the basket.  While having a 7-foot plus center would seem to make option 3 preferable, the problem with this team is that the lack of perimeter shooters allows teams to sag off of the outside shooters and play closer to the rim (keeping the inside more crowded and allowing them to double (either with an actual double or by just having various other players applying pressure) Teske down low.  This is why the loss of Livers really hurts.  His outside shooting and ability to drive forces the defenders to play up on him and somewhat relieves Teske. 

That leave us with midrange shooting which (a) is a terrible strategy for winning games with the way the game is played; and (b) again, we are not very good at.

As a long-time basketball coach, I get the urge to look at what we are currently doing and say, "this is not working - do something that we will be better at," but given the current roster construction, there is no easy answer.  Some teams are just not that good.

 

 

Michigan Is Su…

January 28th, 2020 at 5:34 PM ^

On the Michigan basketball insider podcast with Tim McCormick and Sam Webb . Tim was saying that he scouted a practice and during the shoot around he saw one player ..wagner..actually working on proper shot form and trying to duplicate his shot in a repeatable manner..John Beilein was great at teaching fundamentals and usually had a team that shot good ...I love the juwan hire but maybe that could be an improvement in his coaching approach..Howard Eisely was a good shooter