I think Walton eventually makes a roster. Seemed crazy two months ago but I think this Walton is who he is. That injury took a lot out of him and he's just now truly healthy. He didn't just all the sudden learn how to finish.
Anatomy of an Elite Offense
If you’re looking for Ace’s Northwestern Preview, it’s here.
With Michigan’s weekend win over Purdue, the Wolverines probably sealed an NCAA Tournament bid – but more importantly, they continued their excellent month of basketball: they’ve won five out of their last six with the only loss coming on the road in overtime to perhaps the hottest team in the conference (Minnesota). Michigan struggled early in conference play against an easy schedule because of their atrocious defense, but significant improvements on that end helped rack up some key wins during the relatively difficult stretch of Big Ten opponents.
As I pointed out last week, Michigan’s offense has been remarkably consistent. It ranks in the top ten nationally in Kenpom’s Adjusted Offensive Efficiency. Even though there isn’t a surefire NBA player on the roster (by any means), a balanced six-man core has produced one of Beilein’s best offenses in Ann Arbor. Four of the six main rotation players (Walton, Wagner, Wilson, and Robinson) are extremely efficient and five of the six (those four and Abdur-Rahkman) are shooting 38% or above from three. Derrick Walton has played especially well as of late, but Michigan’s offense has been good all season (with the notable exceptions of the South Carolina and Texas games). If Zak Irvin can play well in a complementary role, it can be a lethal – though quite slow-paced – attack. As we saw against Purdue, sometimes Michigan doesn’t even need him to be able to run a good team out of the gym.
Offense at the top of college basketball has improved in recent years. Each team in the top ten of Adjusted Offensive Efficiency has a mark greater than 120 – back in 2013, Michigan’s Final Four season, the Wolverines were the only team in the country over 120. As recently as 2010, the best offense in the country was national champion Duke’s 117.0. That number would rank 30th in the country in 2017. Recent rule changes are surely a big reason for the general improvement offensively, and the increased pace of play has helped make college hoops more appealing.
With an adequate defense – one that has been much better since opponent threes haven’t gone in at an absurd outlier rate – Michigan and its offense can think about improving its seeding or making a run in the Big Ten tournament instead of worrying about their position on the bubble. The Wolverines’ offensive firepower is the obvious strength of the team so I compared the 2017 Michigan offense against elite offenses from the last five years.
[Takeaways after the JUMP]
Good Shooting = Good Offense
Seems obvious, right? Effective field goal percentage has the strongest correlation to efficiency of any of the four factors by far, and the teams with the best offenses in college basketball are usually the best shooting teams. Sometimes elite offenses can be built with good-but-not-great percentages from the field (UNC ‘16, Wisconsin ‘14, Duke ‘14 and ‘16) but usually they require excellence from two of the other four factors.
Michigan’s eFG% of 56.2 is the highest in the Beilein era at Michigan. While the strong three-point shooting – at 38%, 45th-best in the country – is an asset, the Wolverines are actually better from two (55%, 17th-best). By taking so many threes, the floor is spaced well enough for cutters, players rolling off screens, or dribble-drivers to get to the basket easily. With a nearly ideal shot distribution, Michigan takes far more valuable shots than most teams – mostly just catch-and-shoot threes and layups.
TO% Matters (OR% and FTR, Not as Much)
A common theme for teams with an Adjusted Offensive Efficiency better than 120 is that few of those teams turned the ball over much. 2016 Indiana, despite having star PG Yogi Ferrell, is the notable outlier there, but they made up for it with a ridiculous eFG% and one of the best offensive rebounding front lines in the country.
Most other teams – Beilein's included – avoided turnovers well. Of course, Bo Ryan’s last two teams at Wisconsin were the best at preventing them; even though most of the other marks on the above chart are exceptional, the Badgers were a clear cut above. Michigan’s turnover avoidance this season ranks 5th in the country: having possessions result in shots instead of turnovers is critical for any team, but especially one that doesn’t generate many extra possessions by grabbing offensive rebounds.
Prioritizing transition defense at the expense of those extra opportunities is a conscious choice for many coaches, including John Beilein. While some elite offensive teams crash the glass with reckless abandon – and others parade to the free throw line after drawing fouls – neither of those factors is necessarily required for success. There’s far wider variance among elite offensive teams in terms of offensive rebounding and free throw rate than there is in effective field goal percentage and turnover rate.
Lots of Fringe NBA Players
Almost every team on that chart has some representation at the next level. While most of these guys are role players at best (and end-of-the-bench afterthoughts at worst), there clearly needs to be some NBA talent for teams to have elite college offenses.
- 2013 Michigan: Trey Burke, Tim Hardaway, Nik Stauskas, Glenn Robinson, Caris LeVert
- 2015 Duke: Jahlil Okafor, Justise Winslow, Tyus Jones, Marshall Plumlee
- 2014 Duke: Rodney Hood, Jabari Parker, Marshall Plumlee
- 2014 Michigan: Nik Stauskas, Glenn Robinson, Caris LeVert
- 2015 Notre Dame: Jerian Grant, Pat Connaughton, Demetrius Jackson
- 2016 Michigan State: Denzel Valentine, Deyonta Davis, Bryn Forbes
- 2014 Wisconsin: Frank Kaminsky, Sam Dekker
- 2015 Wisconsin: Frank Kaminsky, Sam Dekker
- 2016 Kentucky: Jamal Murray, Tyler Ulis
- 2016 Indiana: Yogi Ferrell, Troy Williams
- 2016 Duke: Brandon Ingram, Marshall Plumlee
- 2014 Creighton: Doug McDermott
- 2016 Iowa State: Georges Niang
- 2016 North Carolina: Brice Johnson
- 2015 Villanova: None [EDIT: Daniel Ochefu]
- 2016 Villanova: None [EDIT: Daniel Ochefu]
A few things:
- Regarding the Villanova teams, senior Josh Hart is an All-American candidate with a shot at making a roster, even though he might not get drafted in the first round. Other than that, every other team has a pro – even if they’re rookies who are barely clinging to an NBA contract. [EDIT: as a commenter helpfully pointed out, former Wildcat big man Daniel Ochefu is actually one of those players clinging to a contract (with the Washington Wizards)]
- Michigan’s 2013 team has the most NBA talent of any team on the list, even without Mitch McGary, who’s now out of the league after a brief injury- and suspension-filled career with the Oklahoma City Thunder. Oddly enough, Tim Hardaway is probably the best pro from that team, though Caris LeVert is certainly showing some enticing potential in his time back since the injury.
- I cannot believe Marshall Plumlee has a roster spot in the NBA. Phil Jackson… what are you doing, man?
- Trying to figure out the best player on the list was tough – I’m settling on Rodney Hood or Jabari Parker. Ironically, they were both a part of the Duke team that lost to 14-seed Mercer in the first round of the tournament a few years back.
- Perhaps Moritz Wagner or DJ Wilson will eventually make the league but at this point, it seems likelier than not that 2017 Michigan will not have NBA representation when it’s all said and done.
A Quality Defense Will Get You Far
Sorting the table by team defense provided some interesting results. With the exceptions of 2015 Villanova (who lost in a 1 vs. 8/9 game against North Carolina State) and 2016 Michigan State (LOLOLOLOL), six of the eight teams with the best defenses among the teams with elite offenses made it to the Final Four. Outside of 2014 Michigan and 2015 Notre Dame (who both lost heartbreakers to Kentucky in the Elite Eight, oddly enough), the teams with elite offenses and sketchy defenses rarely made significant tournament runs.
At this point, Michigan’s offense is a known quantity – and it’s quite good. Still, games will be won and lost because of the defense: the Wolverines are 18-1 when allowing 1.077 points per possession (with the only loss coming in the horrendous South Carolina game) or fewer and 1-9 when allowing more than that number. The Wolverines won’t be a good defensive team this season, but if they play well enough on that end, they can beat anyone.
He's been playing out of his mind. I can't believe there's not a single NBA team that won't give him a look. Whatever clicked to give him the confidence level he's at now, it's not going to turn off.
numbers are interesting. Beilein, it could be argued, was a ahead of his time when it comes to scheme. There's a reason pro coaches visit him.
offense put a lot of emphasis in spacing and shooting. If you can shoot 3 and make easy layups, your offense tends to be efficient. NBA caught up to the analytics that more and more teams are focusing 3 pt shooting, spacing and easy baskets.
but if Zak Irvin can just get going even a little bit, this team's offense would be absolutely humming. I also really like the change of pace X has shown off the bench in recent games.
It seems to me the main key for this team is keeping the bigs out of foul trouble. They can survive stretches without Wagner and Wilson, but they probably can't survive auto benches or even either of those two fouling out. Wagner, for the most part, did a nice job of not fouling Swanigan; maybe NCAA refs will like him better than Big Ten refs.
that five of the top eight defensive teams in this "elite offense" group made the championship game, and none of the others with lower defensive efficiency even made the final four.
I mean it kinda matches what you would expect. If you're really good at both offense and defense, you're more likely to make the final four than someone merely elite at one.
Daniel Ochefu was on Nova last year and is currently on the Wizards roster, although as a rookie barely hanging on.
Thanks for the correction. Post is updated.
I bet Zak and Walton both get roster spots. Walton's game is complete and Zak does everything but score. Obviously Zak is a long shot but Beilein really gives players the skills they need at the next level. I don't think either will be consistent players but they will bounce around for 5-7 years and average 5 pts a game until they decide to coach. Hoping for the best for both of them. Now about this tourney run! Get it!
Is there a way to see what our defensive rank is just for the month of February? It feels like we turned a corner and I'm curious how we compare just in the recent time frame.
Also very interesting about the OR%. More ammo to use when people complain about Beilein's rebounding strategy. Although the real correlation is whether a lower OR% can improve your Defensive rank. It doesn't appear that way from the data, but it's hard to infer how much worse a defense could be if they crashed the boards more.
More fodder for the Beilein as genious tome. Nice job Alex.
Right, because it has no basis in reality? Other coaches and commenetators consistently praise Beilein as an innovative offensive mind. Guess they are also in on this "tome". Care to enlighten the sheeple on this?
You can rightly knock his recruiting but that is it. The Beilein sucks tome is idiotic. The minute he got high caliber players he went to the tourny final (and was robbed of a NC) and then the elite 8 falling to an elite Kentucky squad on a buzzer beater. Guess what, the coaches that often make the final four always have elite players too. The coaches that make deep tourney runs consistently have elite players consistently. Beilein does more with less for sure, but he also did more with more (when he had it).
This team has two identities:
1: The one we saw beat Purdue, with everyone clicking on offense minus Irvin.
2: The one where Mo and/or DJ get into foul trouble early and often, the offense loses it's flow, and we look like the team that lost to Ohio State almost a month ago
hes going to be a 1st round draft pick next year
...when it does not predict championships? The Kenpom era tells us that no team with a defense worse than top 21 has ever won a title. Can it be done? Only time will tell. But why invest in offense when you can instead invest in defense? It's like having a shiny sports car that you can't drive when there's a thunderstorm. Ultimately pointless. No Beilein team has come close to top 21. Only one Michigan team in Kenpom era has done it and that was the 2004 Ammaker team that won the NIT. Clearly, defense alone cannot get it done (see Cincinnatii every year) but it is a prerequisite for winning it all. A prerequisite we never have.
Yes - either you are the national champion, or you are a pathetic loser. There is nothing in between.
Those banners hanging in Crisler commemorating the 2012, 2013 and 2014 seasons are telling a lie.
I never said we are pathetic losers. I'm saying it's possible we go into every season with a fundamentally flawed strategy. I'm calling attention to our process primarily. If we went into every season with a top 20 defensive team and had our same outcomes I would not be disappointed with those finishes. The problem is I don't think our defense is ever good enough to win it all, which again thus far seems to be a pre-requisite. We only have 15 seasons of data so it's possible this is all small sample size but why bet on THAT when we can just bet on defense?
Well considering the fact that Michigan went to a championship game and proceeded to lose a very close game I beleive it can be posited that a team outside the top 21 can win a championship. Same goes for 2015 Wisconsin that had almost identical AdjO and AdjD ranks to 2013 Michigan according to Kenpom.
This offense is as "elite" as Speight. 10 points in 11+ minutes to start tonight. Does this blog ever look beyond fancy stats?
can we rely on your eye test instead Please.
28 first half points. Great day to discuss being an elite offense
Thanks a lot Alex
good job, jinxie!
and then taking a simple average, M will either lose in the sweet 16 game, but likely win and lose in the elite 8 game. With an average of 4.125 where '4' is elite 8 and '5' is sweet 16.
I'd take that considering how this year looked like it was going to end in January.