UofL Notes: VA Tech is weird, and free throws are a problem
One of the weirder games you'll see UofL play
In a previous newsletter, I predicted Virginia Tech would be a great matchup for UofL because they fit a pattern: teams relying on great shooting, but who were suspect at rebounding. UofL tends to limit teams shooting, and the Cards step up and rebound well when playing teams who are generally poor rebounders. Virginia Tech also rarely forces turnovers, so the Cards should have been able to execute on defense. That should add up to a strong offensive game for UofL, and their usually strong defensive game should lead to a blowout.
I turned out to be half right. UofL set a season high in points per possessions (1.54). However, they also had by far their season high in points allowed per possession (1.48) and escaped with a 94-90 win. It's incredibly rare to have your best offensive game and worst defensive game in the exact same game!
Let's look at what I was right about first. Tech rarely forces turnovers, and this game was no exception. UofL posted a season best turnover rate of 4.9%, turning the ball over 3 times all game.
However, Virginia Tech didn't want to conform to my theory that UofL shuts down (or slows down) opponent's shooting. They went the other way, and became the 6th team all year to exceed their season eFG% against the Cardinals. Not only did they exceed their average, but did so by over 20%. They set a season high (by far) for eFG% by a UofL opponent at 77%. To put that in perspective, the next highest eFG% by a UofL opponent this year is 56% by Virginia on Feb 6th. Even if Virginina Tech had matched their season average (currently 57%) they would have set a season high for a UofL opponent, so they are a great shooting team. On this night, they were legendary.
To make matters worse, literally nothing about the type of shots they were taking pointed towards this performance. They weren't getting layups and dunks; Virginia Tech took 20% of their shots at the rim, while UofL opponents typically take 34% of their shots there. They weren't getting easy shots by offensive rebounding; Tech got 26% of their misses, down from the 29% UofL opponents typically get. UofL allowed fewer easy shots than they usually do, but Virginia Tech still managed to make offense look easy.
Lastly, Virginia Tech has had some great shooting games this year. They had posted an eFG% above 60% 6 times before this game, although never as high as 77%. They were 6-0 in those games, including an earlier win over Duke. But somehow in this game, despite shooting their best all season, the Hokies lost.
I would normally call a 77% eFG% unsustainable or once-a-season, but the Hokies did something ever weirder. They went out and beat that number on Feb 25th, posting an 81% eFG% against Boston College. Oddly enough, the Hokies again did so despite under-indexing in shots at the rim and offensive rebounds; it was all jumpers. Virginia Tech is probably the weirdest offensive team in college basketball, and for one night they dragged Louisville into their alternate universe.
Louisville has a bad relationship with free throws
Before the UofL-UNC game on Feb 22, I tweeted out something to the effect of "UNC doesn't usually get many FT's, and UofL gives up a lot." and "UNC is good without FT's. UofL will be under pressure if they give them up."
That's pretty much how it played out. UofL limited UNC to a 43% eFG (below their avg of 53%) and forced turnovers at UNC's usual rate. However, the Cards lost by 11 and were basically out of the game with 10 minutes left. The issue is that UNC drew 29 free throws to UofL's 13. While some fans might want to point to biased refereeing on the road, the fact is that UofL has repeatedly dug themselves a hole by allowing free throws more than they earn them.
To measure this, I'll look at FTA/FGA, which divides the number of free throws attempts by field goal attempts. This comes out to a percentage, but I'll express it as a number to avoid typing % over and over. So, 50 means a team shot 50% as many free throws as field goals (say, 20 FTA and 40 FGA). A ratio above 40 or so is really high. For the season, UofL on offense is at 35, and at 39 on defense. So, UofL draws free throws slightly less often than they allow them. Note that this metric ignores possessions which end in a turnover, since no FTA or FGA is recorded, and doesn't care if there are multiple FGA on 1 possession...every FGA and FTA counts as 1.
Free throws are a great source of offense, because they're pretty easy to hit. If a team shoots 70% on free throws, every time they earn 2 FTA they will average 1.4 points per possessions (70% chance on each FTA). No team in the country shoots 70% on field goals, so 1.4 points per possession doesn't really happen unless you're getting free throws.
It's therefore a really good idea to draw more free throws than your opponent. Otherwise, you have to do one of the following things to win:
Force a huge number of turnovers, so your opponent has a lot of zero point possessions
Shoot much better from the field than your opponent
Hope your opponent is terrible at shooting free throws
3 is pretty much out, because no team is so bad at free throws that they'd rather be taking field goals. So, teams who give up lots of free throws can still be very good by doing one of the first 2 things. Last season, West Virginina fouled more than anyone, but forced turnovers more than almost anyone. Kentucky fouled a lot last season, but had an eFG% 10 pts better than their opponents. So, it's possible to be a good team while allowing free throws, but it's tough.
This season, UofL has tried a bit of both. The Cardinals force turnovers more often than they commit them, and have an eFG% 6.5 pts better than their opponents. However, I'm a bit concerned about how this holds up against quality teams because everything gets much worse when you just look at performance vs NCAA tournament-caliber teams (defined here as teams in KenPom's top 50). Take a look:
UofL's eFG% advantage over opponents shrinks to 1pt (49% vs 48%)
UofL's turnover advantage decreases slightly (16%-19%, from 16%-21%)
The FTA/FGA gap widens to 32 for UofL and 42 for opponents
These numbers have been consistent whether UofL is at home, on the road, or at a neutral site, so I would expect them to continue in postseason play. Does UofL have any opportunities to change up their rotation against top teams to be more successful?
UofL may have some better options against top teams
We can look at 3 options for UofL to really improve their performance against NCAA teams:
Reverse the free throw disparity
Improve the turnover difference
Improve the eFG% difference
Looking at various combinations of players, there doesn't seem to be much hope on the first two accounts. Regardless of who is in the game, UofL puts up pretty similar numbers around fouling and turnovers. These are likely related, since aggressive defense leads to turnovers and free throws, and a core part of Louisville's team DNA. But, there does appear to be a glimmer of hope in improving the shooting difference against good teams
UofL has stuck to a core rotation of 4 players at PF and C this season: Jaylen Johnson, Mangok Mathiang, Anas Mahmoud, and Ray Spalding. Here's how various combos of these players have impacted UofL's eFG% differential:
UofL has performed much better against top-50 teams with Anas Mahmoud in the game.
UofL has shot better and limited opponent shooting much better with Anas Mahmoud in the game against top-50 teams. The Cardinals are prone to giving up a lot of free throws due to their aggressiveness on defense, and need to make up that difference by shooting better than their opponents. Playing Anas Mahmoud has been the best way to accomplish that against top-50 teams; the type of teams UofL will see in the tournament. I would encourage Louisville to play Mahmoud 30+ minutes in ACC and NCAA tournament play to give the Cards the best chance of winning.
Thanks for reading my newsletter. If you have any questions, want to argue a point, or have some feedback, feel free to reach out via email at sean@hoopsinsight.com, or on Twitter @hoopsinsights. If you liked this, let me know as well, and tell your friends to subscribe at www.hoopsinsight.com.
You can forward this to others, but please ask them to subscribe as well so I can keep track of who's enjoying my insight and analysis