UK vs UofL 2017 Preview Part 1 - Season Recap (so far)
This is Part 1 of my 3 part preview of the 2017 UK vs UofL game:
Part 1 (Wednesday): The Season So Far
Part 2 (Thursday): Strengths and Weaknesses
Part 3 (Friday): How Each Team Can Win, and my prediction
Louisville more steady than you'd expect, but not because of experience
The Louisville Cardinals had what you might call a turbulent offseason. After losing their Hall of Fame coach, plus several key contributors to last season's team, nobody knew what to expect from the Cards this season. While the Cards came up short in their two games against NCAA tourney-level competitions (Purdue and Seton Hall), and have had some ugly stretches in their wins against lesser competition, their performance has actually been shockingly steady, given the offseason upheaval.
The Cards had two really standout games, in 40+ point wins over Southern Illinois and Bryant. While neither team is anything to write home about, winning by 40 or more points is a great sign. In both games, UofL was able to limit turnovers to aout 10% of their possessions andrebounded over 75% of their opponents' misses. In most of their other games, UofL has turned the ball over on roughly 20% of their possessions and rebounded less than 70% of opponent misses. Turnovers and allowing offensive rebounds have the effect of giving your opponents extra shots; that's been a huge problem for the Cards this year. The Cardinals have always been good at getting extra shots by forcing turnovers and getting offensive rebounds; this season they haven't had such a huge advantage. This forces the Cards to outshoot opponents; that's not a strength (124th nationally in effective FG%). In order to play at their best, the Cards have to limit the extra possessions they give to teams.
You'd think that the Cardinals would be relying heavily on their more experienced players to carry them, but that's not been the case this season. While their starting lineup of Snider/King/Adel/Spalding/Mahmoud is their most experienced group, that combo has only outscored opponents by +14 points in 186 possessions (+.08/possession). That lineup has played over 5x more time together than any other lineup, but is barely outscoring opponents. By contrast, a lineup of Perry/Sutton/Nwora/Thomas/Williams features 4 freshman and a transfer, and has only played 16 possessions together, but has outscored opponents by +15 points. To reiterate the importance of UofL's bench, UofL has a higher scoring margin when they substitute any starter...no matter which one! Take a look at the numbers:
Quentin Snider; UofL outscores opponents by +0.21 pts/poss when he is in, and +0.27 when he is out
VJ King: +0.20 when in, +0.25 when out
Deng Adel: +0.22 when in, +0.24 when out
Ray Spalding: +0.19 when in, +0.27 when out
Anas Mahmoud: +0.18 when in, +0.27 when out
This means that their bench players are making a positive impact, to a man:
Darius Perry: +0.29 when in, +0.18 when out
Malik Williams: +0.27 when in, +0.21 when out
Ryan McMahon: +0.35 when in, +0.21 when out
Jordan Nwora: +0.23 when in, +0.22 when out
Dwayne Sutton: +0.28 when in, +0.19 when out
It's not just a case of bench players beating up on other bench units, either. UofL's starters have played 119 possessions against the opponent's starters, and have outscored them by +3 points, total. Every other UofL lineup has combined to play 101 possessions against the opponent's starters, and have outscored them by +31 points. So, while UofL's results have been steady when looking at entire games, there has been a lot of fluctuation within games...and UofL is not necessarily playing the lineups that give them the best chance to win. Louisville struggles to tread water with their starters, and only pulls away once the Cardinals start making substitutions.
This was the case in their loss to Seton Hall. UofL's starters played 8 possessions together and were outscored by 7 points. The rest of the game, the Cards outscored Seton Hall by 5 points. Oddly enough, the Purdue loss was the best performance all season for UofL's starters, as they outscored the Boilermakers by 7 points in 23 possessions. UofL had a 4 minute stretch with Mahmoud and Spalding out where they allowed a 12-0 run for Purdue. However, nothing about this stretch resembled anything like the rest of UofL's season. While it's tempting to think the Purdue game showed what UofL's starters can do, it's natural to question why they haven't done that in any other games. Here are the scoring margins in each game when UofL's starters have played:
Vs George Mason: -10 pts
Vs Nebraska Omaha: did not play together
Vs Southern Illinois: +6 pts
Vs St. Francis: +5 pts
Vs Purdue: +7 pts
Vs Seton Hall: -7 pts
Vs Siena: +8 pts
Vs Indiana: -11 pts
Vs Bryant: +17 pts
Vs Memphis: -1 pt
Vs Grand Canyon: +0 pts
UofL's starters have been outscored by double digits twice (Indiana and George Mason), and outscored opponents by double digits once (lowly Bryant). It's reasonable to expect UofL to get out to slow starts more often than not, especially against quality opponents.
Kentucky's season has been a roller coaster
It's been almost impossible to predict whether the Wildcats will play a strong game or a poor one. By my proprietary stats, I have the Wildcats rated as +0.20 points per possessions better than average. However, they have only played one game around that level (+0.19 vs Kansas). Every other game, the Wildcats have been at +0.15 or under, or +0.26 or over. Here's a breakdown of what I would consider poor games and strong games:
Poor games: Utah Valley, Vermont, Troy, Harvard, UCLA
Strong games: East Tennessee St., Fort Wayne, Illinois Chicago, Monmouth, Virginia Tech
If you wanted to reduce UK's season down to one factor, it's been whether or not UK can shoot better than their opponents usually allow. In every one of the games I labelled "strong" performances, UK's effective FG% was dramatically higher than the opponent allows for the season. In every "poor' performance, UK shot right around where the opponent usually allows. Take a look at the data, first for the poor games:
Utah Valley allows 49%; UK shot 43%
Vermont allows 49%, UK shot 49%
Troy allows 51%; UK shot 47%
Harvard allows 47%; UK shot 48%
UCLA allows 48%, UK shot 47%
Now the strong games:
East Tennessee St allows 49%; UK shot 65%
Fort Wayne shot 54%; UK shot 65%
Illinois Chicago allows 49%; UK shot 70%
Monmouth allows 47%; UK shot 57%
Virginia Tech allows 51%; UK shot 57%
UK simply cannot overcome even an average shooting night this season. Most of their other stats have been equally volatile, but haven't followed the same pattern in poor and strong games. For example, UK forced a ton of turnovers against Utah Valley, and allowed a low eFg% to Troy, but had poor games. Conversely, UK turned it over a ton and rebounded poorly against East Tennesee, and fouled way too much against Illinois Chicago, but played well overall. The story of UK's season has been that shooting dictates their performance, to an eerie degree.
UK's typical starting lineup of Green/Diallo/Knox/Washington/Richards has been a bit worse than UK's average, outscoring opponents by +0.11 pts per possession. However, their performance has been drastically different against quality opponents:
UK starters vs:
Kansas: -3 in 12 possessions
Virginia Tech: -1 in 23 possessions
UCLA: -16 in 18 possessions
Rest of season: +34 in 74 possessions
While the starters have dominated against lesser teams, they've struggled to hang with the better opponents on UK's schedule. As with UofL, it's not just an issue of playing against starters. UK's starters are -23 in 32 possessions against the starters of Kansas, Virginia Tech, and UCLA. All other lineups for UK are +3 against these team's starters in 35 possessions. As with UofL, UK's starting lineup doesn't appear to be giving them the best chance to win against the best teams.
For UK, the strongest combination has been a backcourt of Shai Gilgeous-Alexander and Hamidou Diallo. Both players bring length and quickness defensively to help contain guards, and have been able force turnovers at a high rate. Quade Green does not bring the same defensive presence, and struggles at times to contain dribble penetration. Against UCLA, a Diallo/SGA backcourt (without Green) outscored the Bruins by 14 points in 32 possessions. They did this largely by forcing turnovers on 31% of UCLA's possessions; by comparison, UK turned it over on 16% of their own. For UK to put together a complete game, they need Green to hold his own defensively.
In summary, UofL has been much steadier than UK this season. That's not a huge surprise given UofL's edge in experience, but UofL's experienced players have not exactly held up their end of the bargain. Both UK and UofL's starters have generally been less productive than lineups with bench players mixed in. That makes me very interested in seeing who is able to have the advantage in the first 5 minutes of each half during Friday's game. Given that both teams perform better with their bench lineups, it will feel like "found money" if either team is able to establish much of a lead with their starters in the game.
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