Numbers are your friend.
I don't know if anyone else has noticed, but lately I have begun to finally make a concerted effort to embrace and understand advanced metrics in basketball. I played enough basketball through my life that I had a pretty solid understanding of the game, and with that came a basic understanding of the kinds of advanced stats bits and pieces that play a big role in our modern understanding of basketball. I know what tempo free means and what the four factors are. However, it was time to take things to the next level.
I'll be writing a few more advanced stats pieces through the rest of basketball, as well as including more and more of it in my previews and postgame recaps. Hopefully my efforts to embraced advanced stats will benefit all of us.
Today, I want to take a look at three very interesting bits and pieces of statistical research that tell a whole lot about Michigan's team and give a deeper understanding of of what we've been seeing on the court this year.
The McGary effect
A big reason that I have finally started to try and dig deeper into the advanced stats is the work of Brian and Ace at MGoBlog. Both write incredibly detailed and statistically smart pieces on basketball, and on some level my motivations were as much about keeping up when I'm reading along with their as they were about catching up as a writer.
Well, Ace did it yet again with a short post on lineup specific stats. Most of the guard and wing rotation is sporadic so it doesn't lead to A) much difference in production and B) much of a sample size that is any use when it isn't Burke-Hardaway-Stuaskas-Robinson on the floor at the 1-4 spots. But, with those four playing the majority of minutes and a pretty even split between playing time for Jordan Morgan and Mitch McGary, there is quite a bit you can get out of that breakdown. The results are quite surprising.
Morgan is still Michigan's number one option at the five, but the numbers show that it is actually McGary being on the court that has led to better offensive and defensive numbers. Both lineups have plenty to look at in terms of sample size (199.1 offensive possessions for McGary and co., 180.6 for Morgan; Horford has only played 43.3 offensive possessions. Defensive possessions fall across the same line 202.6/179.2/43.8 respectively).
Offensively, Michigan performs much more efficiently with McGary on the floor, a difference of almost 15 in Kenpom's offensive efficiency rating. When McGary' is on the floor, the team has an Off Eff rating of 126.1 (Michigan as a team sits at 124.4 and is one of only four teams with a rating north of 120). Meanwhile, Morgan's unit comes in at 109.1, which would still be good for 12th nationally when compared to the rest of the country. Furthermore, Michigan's defensive efficiency is better with McGary on the floor (90.3 to 102.1) as are the overall plus/minus numbers (+49 to +11).
Now, as Ace goes on to point out, a lot of the raw data is close. Michigan's 2pt shooting percentage is close for both players, so are offensive turnovers and offensive rebounds by Michigan. The big differences are A) Michigan forces more turnovers with McGary on the floor and B) Michigan shoots better from outside. Ace comes to the conclusion that these aren't so unrelated:
There's a possible explanation for that, however, in the defensive numbers. The Wolverines force more turnovers with McGary on the court (19.2% to 15.6%), and of late many of Michigan's best looks from three have come off their transition game. That probably doesn't account for a 12-percent difference, but even if that's normalized there's still a gap in offensive production between the two; I consider McGary the better passer, a factor that may also contribute.
The difference between the two defensively is easier to figure out. McGary's activity defensively helps the team force more turnovers, while his excellence on the glass leads to a better rebounding rage. While McGary fouls more often than Morgan, the team fouls so rarely as a whole that the foul rate isn't affected greatly.
The last takeaway from all of this? Michigan is actually in pretty good hands no matter which of the three centers play. While Horford doesn't have as much time as his counterparts, his numbers fall right in line as far as effectiveness goes. Off Eff of 124.7 and Def Eff of 89.0 are McGary-like numbers that while being probably a little unsustainable over a larger sample size, are also good signs that Michigan is still in capable hands.
Feel free to go back to thinking about this team entirely within the phrase: embarrassment of riches.
The D: good or good enough?
Along with Michigan's startlingly efficient offense, it has a pretty solid defense. The Wolverines have been fluctuating somewhere around 40th in the nation in Kenpom's defensive efficiency rating. John Gasaway looks into this a bit more in response to a point made by Luke Winn that no team with a defense at the level of Michigan's has won a national title in the past decade.
The question, as Gasaway poses it, isn't whether Michgian's defense is good enough to win a national title, it is whether Michigan's defense is good enough to win a national title paired with the Wolverine's exceptional offense. Michigan's defense may not be the lock down unit that last year's Kentucky team was, but Michigan has a few defensive things going for it. The Wolverines don't allow a lot of offensive rebounds (12th in the nation in OR%, they don't foul much (2nd in FTA/FGA), and they don't turn the ball over much when on offense (5th in TO%). So while Michigan isn't great at forcing turnovers (Michigan is 19.3%, and below the national average which is 20.5%) the Wolverines do a lot of other things to limit scoring opportunities: namely, Michigan doesn't give up as many second chance points, free shots from the line, and points off turnovers.
Michigan may be at a slight disadvantage in the tournament when going up against other elite teams -- especially when one considers that Michigan's one loss came at the hands of a defensively minded Ohio State team -- but given the offense's level of production, it isn't hard to imagine Michigan surviving. After tonight, the next four games will be a very good indication of how good Michigan's defense is, and how good it will need to be when Michigan's offense is faced with a stiffer challenge.
Stauskas: momentum killer?
The last stop is Burgeoning Wolverine Star for a quick check in with Chris, who looks into what seems like Nik Stauskas's proclivity toward trying to match opponent three point makes. I'll let Chris set it up for you himself:
When watching Stauskas, it always struck me how aggressive he became after the opposition hit a three pointer, often resulting in an irresponsible shot--though occasionally flashing some of that "swag" the kids are so excited by. What I wanted to look at was Stauskas' typical usage rate versus how he responds to an opponent hitting a three pointer. Distilling all of the available data and filtering it properly--what 3-pointers was Stauskas actually on the floor for?--without watching every minute of every game again was difficult, but I found an easy workaround that stands up to a sanity test.
Chris than went about coming up with the numbers to test his theory. He dug deeply into game footage and came up with quite an impressive chart that breaks down how often an opponent three was followed by a Staukas shot. Obvious sample size concerns abound, but there is a little bit of something there. Once again, Chris:
According to Kenpom, Stauskas uses 16.7% of Michigan's possessions this season. While Stauskas is on the floor following an opponent made three pointer, that usage jumps to 27.5% (or 25% if you remove the two possessions ending in free throws). In a season where Stauskas is taking 1 of every 7 shots when he's on the floor, that number jumps to 1 of every 4 following a made three pointer by the opposition. He's also shooting 9% worse on FG attempts overall in these situations versus his season average (40% vs. 49%) and 11% worse on 3FG attempts (38% vs. 49%), indicating potentially worse shot selection, though there are concerns about sample size...
The data, although sparse, does coincide with what Chris has noticed -- that Stauskas has a bit of that killer instinct in him that makes him want to steal back momentum after a big shot from the opposition.
Not only is this an interesting line of thought, but if this is what passes from problems that need to be deeply investigated on this team, I think we can all agree that things are going pretty well.