I’m not going to lie, I want 1 NCAA Tournament W. Coming into the season, getting that monkey off the backs of both Shaka Smart and Marquette, which have each gone winless since 2013 in the NCAA Tournament, was all I hoped for.
Of course, a fair bit has happened since then, the foremost being Marquette turning into an offensive juggernaut and becoming a top-15 team in both the AP poll and the advanced metrics, not to mention winning the Big East regular season title.
So it’s ok to say expectations have been lifted. With those results has come the high seed projections, which currently pit MU right on the cusp of a 2/3 seed. Using TRank’s data, 2 and 3 seeds have made the Sweet 16 about 91% of the time since 2000 (160 of 176 teams). So that would probably be a decent updated barometer.
Thus you can probably guess my reaction when Will Warren, a fantastic analyst/writer that usually focuses on Tennessee but also writes about D1 at large for his Substack as well as The Field of 68, published an article that listed Marquette’s defensive liabilities and historic comparisons as reasons to not be surprised at an early exit.
The key stat: Teams most like Marquette: 2012 Duke (2 seed, R64), 2013 Michigan (4 seed, runner-up), 2015 Iowa State (3 seed, R64)
Dammit.
But, being the nerd I am, and having just taken a stroll through the defensive numbers this week, I wanted to take Will’s amazing analysis a step further (please go and read it while it’s free). It felt all season like Marquette’s defense played down to its opposition, letting up on the gas after amassing 15+ point leads against Butler, DePaul and St. John’s just in the last month alone. So what if we just looked at its games against the best team?
To take this a step further, I took all 4 of the extreme O v D teams Marquette was grouped with in Will’s initial analysis (Gonzaga, Baylor, Xavier and Arizona) and then added the 3 most recent examples in LSU 2019, Xavier 2018 and Iowa State 2015 as well as the other 2 teams identified in Michigan 2013 and Duke 2012 (regular season numbers only).
I then used TRank splits to find defensive efficiency numbers for games in quadrants 1 and 2 for each team, as well as Qs 3 and 4 trying to see if Marquette was unique in their splits or it was a pattern shown by other, similarly rated teams.
Using the Adj D Efficiency number, Marquette’s variance was the quite noticeable. In the toughest 16 games, it averaged a D efficiency of 96.1, while that number ballooned to 100.3 in the “easier” games, a positive 4.2 point variance. Of the 10 teams in this sample, only 2 other teams had a positive variance over 2: Gonzaga this season and Michigan in 2013 (which made the Final 4).
Again, what I was trying to see with this breakout was how unique this split was. You expect teams to be better against “worse” teams so to see the Marquette is in fairly rare company does help to assuage some of those fears that the defense is a liability in the way it has been for other recent examples.
And although I don’t have a concrete answer, I do think Marquette’s relatively thin and very inexperienced bench is partly the reason here.
Using Hoop-Explorer, we can map out the exact number of possessions each player played against certain quality of opposition. So using the top 120 as a basic divider, you can see what was obvious to anyone that watched the team year round. Shaka was much about 12% more liberal with his bench usage in the “easier” games, which included games against
Don’t get me wrong, this isn’t to say that all of Marquette’s defensive woes can be tied to Tyler, Kam and Oso playing less. The aggressive trapping style and lack of a shot eraser is still a problem no matter who is out there. It’s not an elite defense and it won’t turn into one magically the next few weeks.
But there is a clear difference between the starters and the bench. And there is a clear difference between defensive performance in Q1/2 and Q3/4 goals that isn’t attributable to timing (early in the year vs late).
And have I mentioned that a comprehensive project by NYBuckets found bench play being cut fairly drastically the further into the tournament you got?
So as we approach the time when the rotation tightens and every minute is played with full intensity, I think I am more confident in this defense than the overarching data says I should be.
That’s not to say Marquette will make the Final 4 and you can book your Houston tickets now, but the flaws in the current squad aren’t quite as detrimental as they may appear at first glance.
Paint Touches 
Leave a Reply