2015-16 Louisville Cardinals: A rough road, on the road
The Cards were much better in home games than on the road during the season. How did the guard rotation and shooting at the rim drive this? Read my analysis to find out...
This is the 2nd in a series of 4 articles about the 2015-16 Louisville Cardinals. Part 3 is coming on Mon Nov 9, and Part 4 on Thurs Nov 12. Check out my archive for my previous articles and others in the Hoops Hindsight series.
As I mentioned in my introductory article for the 2015-16 Cards, their roster had very little major conference college experience. One notable way this manifested was that the Cards were significantly better in home games than on the road, even after accounting for opponent strength and typical home court advantage. UofL had an adjusted margin of +35 points per 100 possessions at home, but only +17 per 100 on the road.
In the 2015-16 season, Villanova had the best adjusted margin in the country at +32 pts/100 poss. Louisville at home was effectively the best team in the nation. But Louisville on the road was maybe the 30th best team, just a mid-level NCAA Tournament team. What was the big difference?
Not just a function of opponent quality
Louisville’s home schedule was much softer than their road schedule. The home/road splits could all be caused by the tougher schedule. We can look at Louisville’s home and road splits just against top-100 teams, and see if there’s a similar difference in adjusted margin. So, here are UofL’s stats in home vs road, counting only games against top-100 teams (per KenPom):
Not only due to the general patterns hold, they’re even magnified in many ways. Louisville’s adjusted margin was excellent in home games against top-100 teams but awful on the road. The Cards actually were better offensively on the road in these games, but were so much worse defensively that it was all for naught. For the rest of this analysis, I’ll focus on these games against top-100 teams to avoid any complications due to poor opponent quality.
Donovan Mitchell was a key to overcoming defensive issues
The biggest difference can be traced to whether Donovan Mitchell was in or out of the game. When Donovan Mitchell was in the games, Louisville’s defense got worse on the road but their offense also got better, and the adjusted margin was actually a bit better on the road.
When Mitchell was out, the defense again had problems on the road, but this time the offense didn’t pick up the slack.
The data indicates that Mitchell’s helped UofL’s offense rise to the challenge in these road games. How was he making a difference?
Trouble at the rim
When Mitchell was out, the Cards shot below average from 2 point range at home (46%) but covered that up by hitting 40% of their 3 pointers. But on the road, the Cards hit only 41% of their twos and 34% of their threes.
When Mitchell was in, the Cards were solid on 2 pointers at home (50%) but a horrendous 15% on 3 pointers. On the road, the Cards improved to hit 54% of their twos and 34% of their threes.
Louisville struggled to hit two pointers against quality opponents when Donovan Mitchell was out. Given how infrequently the team shot threes, this was a big deal and clearly set back the offense.
We can categorize two pointers as either “2 point jumpers” or “At the rim”. With Mitchell in the game, Louisville hit 68% of their shots at the rim at home and 58% on the road. With him out, they again hit 68% at home but only 44% on the road. 44% shooting from the rim is flat-out disastrous; so what was the problem?
The wrong mix of shot-takers
The issue boils down to who was taking most of Louisville’s attempts at the rim. The Cardinals had 8 players take at least 40 shots at the rim on the season. Here are their shooting percentages there, from best to worst:
Ray Spalding, 75% (75 attempts)
Chinanu Onuaku, 74% (105 attempts)
Donovan Mitchell, 67% (55 attempts)
Damion Lee, 64% (96 attempts)
Jaylen Johnson, 59% (85 attempts)
Trey Lewis, 54% (90 attempts)
Anas Mahmoud, 53% (40 attempts)
Quentin Snider, 44% (64 attempts)
When Mitchell was in the game, more shots at the rim were taken by players who hit them. Lee, Mitchell, and Onuaku took the most shots there, and Snider and Mahmoud took the least. Louisville hit 61% of their shots at the rim with Mitchell in.
When Mitchell was out, the three most common shooters at the rim were Onuaku, Trey Lewis, and Lee. Despite Lewis’ high volume, he converted only 13 of 32 (41%). It was even worse in road games without Mitchell: Lewis took 25 shots at the rim (most on the team), and made only 7. Jaylen Johnson (8 for 19) and Quentin Snider (a staggering 3 for 17) placed 3rd and 4th in shot volume at the rim in these situations, and were off target more often than not.
Having Lewis and Snider taking so many shots at the rim was a bad situation for Louisville’s offense. They were 2 of UofL’s 3 worst shooters at the rim for the season; in halfcourt situations, they were actually the two worst. Despite this, when Mitchell sat Louisville’s offense featured a steady diet of pick-and-roll with one of these two as the ballhandler. This led to a large number of situations where Lewis and Snider were trying to create their offense at the rim in the halfcourt; this was not their strong suit…especially against top-100 caliber teams.
Lewis hit 53% of his shots at the rim in halfcourt situations against teams outside the top-100, and Snider converted a sparkling 63%. But against top-100 teams, Lewis made only 37% and Snider converted 35%! By comparison, Mitchell made 67% of his halfcourt shots at the rim against top-100 teams and Damion Lee made 61%.
Not enough answers when Mitchell sat, and Mitchell sat a lot
Louisville had a tremendous defensive team in 2015-16, but had some vulnerabilities on offense. Most notably, they had only two truly high-level offensive creators at guard: Damion Lee and Donovan Mitchell. Against top-100 level teams Damion Lee played 93% of the non-garbage time minutes, so Louisville didn’t have to worry much about compensating for him sitting out. Mitchell only played 45% of the non-garbage minutes against higher level teams, however, leaving Louisville without one of their best offensive weapons in halfcourt situations.
In these non-Mitchell minutes, Louisville turned to Trey Lewis and Quentin Snider, but those two were not able to create quality offense. In the pick-and-roll sets that the Cards favored, neither Lewis nor Snider was a threat at the basket against quality teams.
I don’t know if Louisville’s coaches purposefully changed their rotations once the postseason ban was announced, but the Cards did start playing fewer minutes with Lee and Mitchell together after that development. In the 10 games prior to the ban, Lee and Mitchell averaged 19 minutes per game together; in the next game they played 22, but then they never played as many as 19 together the rest of the season. In 4 of the last 5 games of the season, Lee and Mitchell played 11 or fewer minutes together.
The decision to play fewer minutes with both of their perimeter offensive threats in the game is one that hurt Louisville in the 2015-16 season. The Cards’ difficulties on the road can be traced to the minutes where one or both was on the bench, as the team struggled to score at the rim. We’ll never know if the coaching staff would have patched this leak by tweaking the rotation in the postseason, but if they had this team could have done something to be remembered more fondly than they are today.