KU Not Closing Games

As you know, I create game trend graphs for every KU game as part of my statistical analysis. There were quotes in today's KU Athletics article about a focus for the upcoming games is to close out games better. The following game graphs show they did better against Vermont in closing the game but were still not great.

First of all, the Michigan St. game was incredibly fast-paced with 86 possessions a piece. That is 20% above the norm for KU. It was an up and down game which should generally favor Kansas. The pace was pretty steady throughout the game but actually picked up in the last 10 minutes slightly. Look at the graph and you will see that after feeling each other out in the first 8 minutes, Kansas kicked into gear. They went on a great run from 12:05 to 3:21 in the first half they increased the lead from 2 to 17. That included a 9 point run and a 6 point run. MSU came out in the 2nd half and were clearly ready to play. They took a 14 point halftime lead to 8 in just over 5 minutes by buckling down on defense and running smart offense. To the Jayhawks credit, they responded with an 11-3 run over the next 3 minutes to take the lead back to 17. Sparty made another run to get it back to 11 and it went back and forth from there from the 11:36 mark to about the 3:40 mark. It was that last 4 minutes where KU lost their composure a bit and let Sparty get some confidence. I watched that sequence and yes, SParty hit some shots, but KU's defense broke down more often in that last stretch and Sparty was running like crazy and we were just a step behind. This is good news because all of this is fixable.



The Vermont game was similar except for the last 4 minutes. KUtook a while to get going and Vermont was hitting shots. But in all fairness, our defense was not as tight as it should have been. KU buckled down on defense when the subs came in and that inspired the team. KU was down by 8 with 9:25 left in the 1st half and closed the half on a 22-10 run to lead by 4. We traded baskets for the first 4-1/2 minutes of the second half before KU flashed and went on a 24-10 run to push the lead to 18 points. From there, KU went soft again but never allowed Vermont to make the big run like Sparty did. Vermont got it down to 9 points but KU responded. The Vermont game shows how a game can be won in a 20 or so minute stretch if you are super dialed in. From 9:25 in the first half to 7:26 left in the 2nd half, KU outscored Vermont 50-24. The other 18 minutes of the game we were outscored 44-34. It is that other 18 minutes we need to work on.

Thank you for your time-consuming reports! I may not understand everything completely, but I always come away with much-needed knowledge

Cool info, and very good data. A couple points:

1.) sample size: we're going off 2 games. One against a team that will be in the hunt for the Big 10 title and will almost certainly make the tournament. The other against a "good but not great" team who might get lucky and win their conf tourney to get to the tournament. With a game tonight and 2 next week, I think the data will either confirm or disprove the "not finishing games" theory, although I WAS concerned with how we let MSU back in at the end. Was that bad play on our side, or was that Tom Izzo?

2.) Learning mode: early games like these are helpful in finding weak points and working on them. We've been here before, and Self is very good at identifying and fixing problems. If it turns out that not finishing is an actual problem, we have a LOT of time yet to address it. In the meantime, every game is another data point: see the problem, identify it and focus on it in practice, play a game and get new data....repeat as necessary.

3.) MSU leads me to believe that we didn't finish well. Not sure that I agree on Vermont and we need to be careful interpreting the data at the end of games. With Vermont, we had the game in hand by the 2nd half. Yes, UVM cut it to 10, but that was corrected fairly simply. The data doesn't show effort or coaching. With a 15 point lead and 4 minutes left, I can't recall if the "a" team as in, or if HCBS had bench players in. I do imagine that after a game is well in hand, many coaches will also let off the gas, let their bench guys play, and ease off so as not to run up the score. Again, I'd have to look at the UVM game, but moving forward if we're up by 20 tonight over Louisiana with 6 minutes left and Self decides to let the bench get some experience, that's a far cry from "not finishing".

Anyway, nothing I'm saying here should be earth shattering. Pretty common sense and I figure you'd already thought about it. Just putting it out there for anybody looking at this and to further the discussion.

Thanks for the visual.

Good catch Eric. As I do every year, I spend some time revamping all my systems to make them easier or more powerful. I forgot to update the score formula. That is fixed now. The scores are populated automatically as long as there is a formula.

I really like the point you made about the pace of the MSU game and it actually speeding up late. My opinion is when you have a lead late you want to play more deliberate. Allowing MSU to dictate a faster pace is part of not finishing.