Was I Right? Looking Back at Some of My UK Insights This Season
Sometimes, trying to get meaningful insights from a few performances by college kids is a fruitless exercise. But sometimes, you can spot trends early if you're willing to look hard enough. In this Hoops Insight newsletter, I want to look back at a few things I wrote earlier this year to see if I was on to something, or if I was full of it. Before I get too self-reflective, here are one-page stats profiles on UK.
Vs Mississippi State and Kansas, Jan 22-26
UK was smothering on defense in these 2 games, allowing a 39% eFg%; the Cats also outscored their opponents by 29 points when playing against 4 or 5 starters, with UK's starters posting a +20 together playing less than half of the 2 games
UK Stats Profile, Nov 6-Nov 28
A bumpy start to the season, as UK had 1 big loss and several wins over weak teams; Immanuel Quickley and PJ Washington were playing well, but UK was outscored when they sat
UK Stats Profile, Dec 1-Jan 26
The team started to click as competition improved; PJ Washington and Ashton Hagans played well, and the 4 most played lineups all were about equally good at outscoring opponents
I shared a couple of stats profiles for longer segments of the season to provide some context as I look back at some of my articles from earlier this season. UK has been a much stronger team since Dec 1, compared to the first 7 games of the season. When I wrote many of these articles, I was comparing some of the early returns from this hot streak to the more scuffling start to the season. As such, some of the insights I shared have held up well, but some have not. With that said, let's look at a few of my earlier articles and see how they hold up.
From Dec 4, 2018: "Is UK Becoming Elite?"
In this article, I noted that UK may have turned a corner in the 2nd half against UNC Greensboro. It seems like a lifetime ago now, but UK was down 40-37 at the half to UNCG. UK is a much better team now than they were then, and my article at the time hinted as much, noting that "recent evidence indicates they can be an elite team". Are they elite for the reasons I noted, or is it just coincidence?
First, I noted that UK had a 7 man core who, at that point, had been dominating over a 4 game stretch: Quickley, Hagans, Herro, Johnson, Washington, Travis, and Montgomery. The issue had come when giving extended time to Quade Green and Nick Richards. UK was posting an adjusted margin of +0.45 points per possession over their prior 4 games with this 7 man core. Since the UNCG game, this 7 man rotation has posted an adjusted margin of +0.26 pts/poss; still very good, although not as otherworldly as before. However, Nick Richards has stepped up from being a semi-liability to being a key member of the rotation. In UK's first 7 games, they had an adjusted margin of +0.05 with him in and +0.22 with him out. Since Dec 1, they are +0.29 with him in and +0.28 with him out. UK has also integrated Jemarl Baker into the rotation without much of a hiccup; in fact, UK is slightly better with him in (+0.31/poss) than out (+0.28/poss). Even with the loss of Quade Green, UK has found a rotation without a significant weakness. I wasn't quite right that UK only had a 7 man core, but I was right that UK needed a consistent rotation without any weaknesses. Nick Richards stepped up enough to be included in that rotation, to his credit,
In that same article, I noted that the trio of Quickley/Herro/Johnson was on an absolute tear over their prior 4 games. They were a stunning +55 in 96 possessions in that stretch, accounting for nearly the entire margin of victory against Winthrop, Tennessee State, and Greensboro. I also noted that the results had been mixed when Hagans took Quickley's spot, although his best 2 games had been the 2 most recent and noting "If the returns from the Greensboro game are reflective of his contributions going forward, UK...will have few stretches where they get outscored". That trio has not recaptured their dominant form for UK, as Quickley has moved into a bench role; their adjusted margin is +0.24 pts/poss, while UK is +0.29 pts/poss with Hagans in for Quickley. I didn't anticipate Hagans' rise to becoming maybe UK's most important player, although I did notice that he had started to show signs of improved play.
I wasn't exactly Nostradamus in predicting UK's rotation going forward, but I will give myself credit for spotting signs of Kentucky's rise much earlier than pretty much any other media outlet. I also noted how UK had signs of being elite offensively and defensively; given that UK has risen to 22nd in offense and 11th in defense per KenPom, I think I was pretty on the money with that prediction.
From Dec 15, 2018: "What You Should Know about UK vs Seton Hall"
Just a few days after I sang UK's praises, they go and lose to a Seton Hall team that will likely be a bubble NCAA tournament team. Partly to ease my own anxiety, I wrote this article about why UK's loss against Seton Hall was not indicative of their true talent level. In particular, I noted that UK struggled tremendously on offense when Immanuel Quickley played with the starters at point guard. UK was 0-7 from the field with this lineup, missing several quality shots, and I noted that UK would have likely won by 5 or 6 points if this group had shot anywhere near their typical form. While this group has seen less time since Hagans' emergence, they have rebounded to post a +9 in 33 possessions and a +0.42 adjusted margin. This lineup hasn't had any games anywhere near as bad as Seton Hall, and UK has played quite well. Even the Alabama loss can't be pinned on their performance, because this group did not play together during that one.
To toot my own horn, my takeaway from the Seton Hall loss was that "I'd expect UK to bounce back in their upcoming games, and to particularly play very well against UNC next weekend in a neutral site game." I'll mark that up as a win.
From Dec 27, 2018: Lineups and Plus/Minus
In this article I noted that UK had moved away from playing the Quickley/Herro/Johnson trio, and was playing the new starting lineup heavy minutes. In fact, the starters were playing more than any UK lineups had played in any game. I took this as a sign that the coaching staff had settled on their rotation and player roles, and that's proven correct. This lineup accounts for all 10 instances where a lineup played 20 or more possessions in a game, including a season high 41 against Alabama. The only game since Dec 15 where they didn't play 20 possessions was against Vanderbilt, where they played 18 in a 56 possession game (still a team high for the game).
I also noted that the starting lineup had been posting results that were a bit worse than UK's average, according to adjusted margin. In the 3 prior games, they had an adjusted margin of +0.29, below UK's +0.35 average as a team and the worst of UK's 4 most played lineups. However, I noted that "it seems like the starters are due for more success than they've been having. Opponents seem to be shooting unsustainably well against them, and Hagans' defensive skills in particular should begin showing up in these lineup stats". I also predicted "they may be in line for a really dominating game, plus-minus wise". In the very next game against Louisville, UK's starters put up a +14 in 26 possessions. They held UofL to a 50% eFG% and forced turnovers on 23% of Louisville's possessions. They absolutely bounced back as I predicted, so I'll call that a win as well.
Sorting through small samples of data and trying to glean insights on a college basketball team can feel foolish. However, it's very rewarding when I spot a small trend and it plays out as I expected. It is also very humbling when I expect a trend to continue and it completely falls apart. Like anyone, I hope to be right more times than I am wrong, learn from my mistakes, and enjoy every step along the way. I hope you're enjoying the insights I share, and as always feel free to reach out with any questions or comments.
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.