Mizzou and Beyond

I know there have been a lot of opinions and opining about the drama surrounding Darryn, his role in the team, chemistry, the mystery injury, Bill Self's cryptic remarks with a Shakespearean "Thou do'st protest too much" vibe. I am going to ty to make the case, and I think it is a very strong case, that there is absolutely no doubt, this team is better with Darryn in the lineup and pretty much every other player benefits from Darryn being in the lineup. Of course, I am going to base my assessment on the quantitative analysis as I am prone to do. But I will not stop there. Quantitative analyses are indicative in nature, meaning they do not define cause, they only indicate or reveal insight to aide in figuring out the causes. The reason guys like Bill Self make trillions of dollars is not because they are brilliant quants. Any old hack like me can do that. They make the big bucks because they truly understand the cause effect sources. I will do my best to relate "what" of the numbers to the "why" of the numbers. First up we need to look at the "what" of the numbers.

This first chart is that big kahuna chart that I am just now unveiling. I frequently publish my player ratings. But I have heretofore not shown the entire team ratings. This is a critical number. This is the direct measure of the teams performance against some standard. The standard I use is the equivalent of the top 5 teams in the country. This chart is really pretty simple. The bars represent how KU performed in each game compared to what the average of the top 5 teams would have done playing the same opponent in the same venue. You will see that KU busted out of the gate against Green Bay looking like they were wildly underrated. They had what is still their best game of the season. Sure, every game has a bit of randomness that you have to factor in. As Fran Fraschilla says, every team will play 5 games way over their head and 5 games way below their ability. Perhaps the Green Bay game was that crazy game where the stars aligned. But from a quantitative standpoint, the data point is very real. For whatever reason, we played extremely well that game. Then something seemed to happen. We laid a huge egg against UNC. Then we had another up game against TAMU-CC but then right back to the dumps against Princeton. But then something that looks significant happened. Every game after that we seemed to get a little better. That is until UConn came to town and the team fell apart again.

So here is the challenge and here is where guys like Self have the distinct advantage. The challenge is, why are each of these game results and ratings as they are? Why did we lay an egg against UNC? Why did we lay an egg against UConn. Was there really a steady improvement after Princeton, basically our games in Las Vegas? And then the big question. What the heck happened against Mizzou when we had our best second best game of the season and played at a championship (top 5) level or better? The reality is, the reasons are myriad. The number of variables that go into how a team performs are legion. Heck it can come down to a player getting in an argument with his girlfriend. There are just way too many causes for performance variance including the noise in the data or the randomness, if you will. So the challenge is to try to uncover and figure out the causes. So question one, what do the Green Bay and Mizzou games have in common. Well obviously Darryn played in both games. But then again, Darryn played against UNC and that was a nightmare. But UNC was a very unique game. We clearly outplayed them in the 1st half and then they went nuclear and couldn't miss a shot in the 2nd half. Then there is the UConn game. We seemed to be progressing and then we had a terrible game, and at home no less. But UConn also was like the UNC game. We really seemed to outplay them for the 1st half and then things fell apart. Other than Darryn playing, there is another narrative that seems to emerge. That is Flory playing against other mobile and strong big guys. Tennessee, UNC, UConn and especially Mizzou. I am still a little perplexed about the UConn game. The The Flory struggling against good bigs does not hold up because Flory had a very good game. It was literally everyone else that struggled. Then I heard Self reference that most of the team had spent some time on the sick bed in the past week. He did not specifically mention anyone suffering during the UConn game. But it does beg the question, was there an illness cause behind the poor performance at UConn? If you remember the game chart (below) it was literally an across the board malaise except Flory. And there is the "dynamic" factor. This is sort of like "multi variable neural network" kind of stuff. But in basketball it goes like this. Certain players depend on other players to play well for them to play well. The most obvious case of this is a team that has a dominant playmaking PG. When that PG is off, he is not distributing the ball well and that diminishes other players. Another obvious example is when you put a guard in who is no threat to shoot, the opponent can adjust their defense to put more pressure in the paint or more pressure on other perimeter players. So the problem is not that a particular player is playing poorly of his own accord but is facing a much tougher defense than he should because of a weakness in his teammate. So when Tre plays poorly for instance, could that affect other players. Or more likely, when Melvin, the guy with the ball in his hands the most, plays poorly are other players like Tre affected? The answer is clearly yes. So it is sort of a force multiplier in reverse. Bad play cascades through the team at times. Having said all that, the best cause I can come up with for UConn is that it was a combo of one or two players off for possibly illness reasons and the rest of the team being affected. Or perhaps many were ill. That can also be a reason why 2nd halves are worse. Players run out of juice when they are under the weather.

But back to Darryn, Darryn absolutely and hugely changes the dynamic of this team. His presence is wildly disruptive to defenses. This allows KU's other players to take advantage. SOme might argue that Melvin should never shoot the ball. I am sort of one of those. I think KU needs to be in desperation with just a few seconds on the clock for Melvin to shoot a three. But Melvin attacking the rim is a decent play for KU. He still is not great but I think it is a decent play. With Darryn in the game, Melvin took far fewer shots. And he shot zero 3's (thank God). That is a huge change in this team. And Darryn's presence absolutely freed up Tre White. The defense could not afford to cover Tre with their best man. And what we saw against Mizzou was not the best Darryn nor the best team play with Darryn. But his very presence is so disruptive to the defense that it lifts the whole team incrementally. So overall, the mere presence of Darryn is a huge lift to this team. And what we saw against Mizzou is not the best version of Darryn nor the team. That is encouraging. And the most encouraging thing (although statistically not yet meaningful) is we played below our ceiling against Mizzou but still at a championship level. The real test will be 5 games from now if this becomes a trend rather than statistical noise.

So who brought home the bacon in the Mizzou game? I have modified the player rating only so that one player playing well above elite level can make up for a player playing coming up short. So we really do not need every player above the line. But we need the entire team to be above the line. I am going to load you up with charts. I added some charts to my program. The first is the standard total player rating. Recall, the little Jayhawk is the rating the player needed to play up to the current rating of the team (currently 11th in the DPPI). The basketball is what the player rating should have been for the player to play at a level equivalent to the top 5 teams. The 2nd table is the same data just expressed as the percentage of the player rating to the top 5 expected rating. You can see that we had 5 guys play right at (Melvin) or well above the level needed for this team to perform at a top 5 level. And 5 of the top 6 players were above. That is a recipe for great success. But these charts also raise several questions. If you watched the game you might be inclined to say that Jamari played well and deserves a better player rating. I agreed, he played pretty well. So why the dreadful player rating? In his case, it really boils down to the +/-. This is an aspect of the game that is sometimes tough to see since we tend to focus on offense and other ore obvious things. Take note, we outscored Mizzou by 20 points. But with Jamari in the game KU was actually outscored by Mizzou by 2 points. That is huge. That is very significant. That says that Jamari's overall impact did not lead the team to success in terms of outscoring Mizzou even though his stat line looked decent. And in this case it was almost all on defense. The +/- defensive numbers do not lie. Mizzou scored 1.91 points per minute with Jamari in the game. When Jamari was on the bench, they scored less than half of that, 0.91 points per minute. The eye test might tell you that Jamari played well, and offensively that is more or less true, with KU scoring 1.82 points per minute with Jamari on the floor. But even offensively, KU scored 2.25 PPM with Jamari on the bench. These are important numbers. And this is the source of the dismal rating for both Jamari and Elmarko. Elmarko's +/- numbers were only slightly better than Jamari's. But once again, Tre and Flory were outstanding and Bryson had one of his best, or maybe his best overall game. And Darryn, while maybe muted compared to his eventual ceiling, was great for the minutes he played. And here is a take that might surprise you because defense is tough to measure with the eye test sometimes. It is easy to see Flory's defensive impact. Oe might conjecture that surely Flory's defensive +/- numbers have to lead the team. But there is one player who had similar, but slightly better +/- numbers to Flory. That is maybe not obvious. But Darryn's defense was tremendous against Mizzou. This is why love doing the in-depth analyses because it often highlights things that aren't as obvious to the eye test.

I want to show one more chart. This is the same player rating data as above just normalized to minutes. This really highlights players who play well even if for fewer minutes. This shows, for instance, that Darryn, while a lower overall total player rating because of fewer minutes played, was actually more effective overall than Flory (barely). But Tre was the heavy lifter in this game. He had an amazing game 0.65 per minute is just great by any measure Recall, the red line is the top 5 equivalent rating. I will say it boldly. If KU continues to play at the level we played against Mizzou, we will be a strong Final Four contender. Time will tell.

Lets look at the updated DPPI. Very interesting things are happening. I have said before, I am having some differences of opinion with the AP voters. I predicted Iowa State defeating Purdue. In my opinion, Purdue was not nearly the top team in the country. Not even among the top 5. Maybe not even top 10, but probably right there in the bottom half of the top 10. Right now I have Purdue ranked 12th, 1 spot below KU at 11th. KU is on the verge of breaking into the top 10. We need another nice game this weekend against NC State to propel us into the top 10. Michigan continues as the dominant team. In fact without doubt the most impressive game of the year was Michigan's 40 point drubbing of No. 2 Gonzaga. Gonzaga looks extremely impressive this season other than that one game and Gonzaga is a solid No. 2 in the DPPI.

I like to watch the competition between conferences. For most of the year, the Big 12 and the SEC have been neck and neck with the ACC well back and the Big 10 a notch down from that. December has been a little tough on the SEC. Mostly the Big 12 has been tough on the SEC. There have been 17 games between the SEC and the Big 12. The Big 12 has won 15 of those. Only the Tennessee comeback over Houston in Las Vegas and Vandy ( top 10 team) defeating UCF fall for the SEC'ers. The Big 12 has carved a nice lead in the battle of conferences. You can see below. Over the weekend, the Big 12 dominated the sec with 4 huge blowout wins. Arizona blew out Auburn, Texas Tech blew out LSU, KU blew out Mizzou and Arizona State, surprisingly, blew out Oklahoma. That really separated the ratings of the two conferences with the Big 12 now solidly in the lead. As the Thanksgiving holiday games gave way to December, the other power 5 conferences picked up their pace of tough SOS but so did the Big 12 which still retains by far the toughest SOS.