The Michigan State basketball team (and many of their fans) just wrapped up Christmas. According to what Tom Izzo said following the win last week over Oakland, his players are currently "enjoying" two-a-day practices in preparation for one more non-conference game against Buffalo this Friday and the resumption of the Big Ten conference schedule just after New Year's Day.
It is a good point in the season to take a step back and look at where we are now in the season and what fans can expect to see from Tom Izzo's squad in 2023. Personally, I am not sure what to expect. I see this as one of the more fascinating teams in the last 20 years.
In flashes, we have seen a team with excellent guard play, not one, but two excellent "stretch fours" in Joey Hauser and Malik Hall (when he is healthy and not manning the perimeter) and flashes of plus-play at the center position. There is certainly the potential that the team could emerge from Tom Izzo's boot camp this week much improved. I believe that this team certainly could grow into both a Big Ten and a Final Four contender come March, especially as I look around at the rest of the conference and country.
Then again, everyone saw what happened at Notre Dame and against Northwestern. That version of the Spartans will struggle to make the NCAA Tournament. In other words, the 2022-23 Michigan State squad appears to have both a high ceiling and a very low floor. That is what the "eye test" is telling me.
Fortunately (or unfortunately), we have more than eyes to inform us about what is happening on the court and what may happen going forward. We also have a large arsenal of data and analytics to supplement what we see on the court.
So far in the non-conference season, I have provided a few updates regarding what some of these metrics tell us about the trajectory of the Big Ten season and Michigan State's place in it. I will continue to provide those updates about once a week in the new year.
In addition, today I will start to provide some additional data to help us all understand some of the details of Michigan State's performance and how it compares to past Spartan teams. Let's dig in.
Kenpom Efficiency in Perspective
In my weekly calculation of Michigan State's expected Big Ten win and conference title odds, I draw heavily from the data provided by Ken Pomeroy. "Kenpom" crunches a lot of numbers for all 363 Division One programs.
The most important Kenpom data are his adjusted offensive and defensive efficiency values which are effectively the number of points per 100 possessions a team is expected to score and to allow. The difference between the offensive and defensive efficiency is referred to as an "efficiency margin" which can be used to rank teams overall and to project point spreads for any arbitrary basketball match-up.
Over the years, I have developed a visual to help put all of these numbers into perspective. That visual is shown below in Figure 1 considering the data as of Monday, Dec. 26.
 |
| Figure 1: Kenpom efficiency scatter plot as of Dec. 26, 2022. |
There is A LOT going on in Figure 1, but I believe this visual provides the best way to grasp the entirety of the college basketball landscape and where Michigan State sits within it. Allow me to break down the elements of this visual.
Each data point on the graph represents a specific team's offensive and defensive efficiency, both past and present. The x-axis represents offense and the y-axis represents defense. I have also set the axes such that moving up and to the right both represent improving performance. Therefore, data points closer to the upper right-hand corner are the best, as they represent both good offense and good defense. Also note that the diagonal lines on the graph represent the adjusted efficiency margin (labeled as "EM").
Championship Resumes
There are four different groups of data points in Figure 1. First, there are the unlabeled small blue diamonds. These data points represent the final pre-NCAA tournament efficiency values for the past 20 NCAA champions. All 20 of these data points fall inside of the light blue rectangle in the upper right-hand corner of the Figure.
I generally refer to the blue area as the "championship zone." These blue diamonds act as benchmarks for the kind of metrics past champions have posted. If a team falls outside of this area on Selection Sunday, I would not expect them to be a legitimate national title contender.
The second type of data point in the figure are the brown triangles. These points represent the current teams who meet the criteria (from an efficiency standpoint) to be considered as a national title contender.
As it currently stands, only 14 teams meet those criteria. This includes 12 of the top 14 teams as currently ranked by Kenpom, as well as No. 16 Baylor, and No. 20 North Carolina. Interestingly there are two highly ranked teams that do not make the cut: No. 3 Tennessee (who are not good enough on offense) and No. 8 Arizona (who are not good enough on defense).
Also note that even the top two teams in Kenpom's current rankings (No. 1 Houston, who is off scale in Figure 1 and No. 2 UCONN) only have efficiency margins just over +30. Almost half of the previous national champions entered the NCAA tournament with better efficiencies. This suggests to me that the overall field in 2023 is not as strong as in past years.
Of the 14 current contenders, Michigan State has already played Alabama, Gonzaga, Kentucky, and the Spartans will play Purdue at least twice. The Spartans also scrimmaged against No. 3 Tennessee. So, the Green and White certainly have been tested and were competitive (so far) in all of those contests. That is the good news.
Michigan State's Efficiency, Past, and Present
The other two sets of data points represent current or past Michigan State teams. The large green circles show the pre-tournament efficiency of every Spartan team back to 1997. Remarkably, 17 of Tom Izzo's last 26 teams ended the regular season in the "championship zone."
A glance at the labels on each data point also shows some of the predictive power of Kenpom efficiency data. Of the 17 former Michigan State teams in the championship zone, 13 of those teams advanced at least to the Sweet 16. Only the 2007, 2016, 2018, and 2020 team did not make the second weekend and one of those instances (2020) was due to the cancellation of the entire tournament.
In contrast, the nine Tom Izzo teams who finished the season outside of the championship zone include only one team which advanced past the second round. That also happens to be the best defensive team in this group: the 2003 team which lost to Texas in the regional final.
This brings us to the final set of data points, which are the green squares connected by the green line. These data points represent the current Michigan State team and the changes in adjusted efficiency over the first two months of the season. The larger labeled square represents the current data for the Spartans.
In the preseason, Kenpom's estimated data placed Michigan State just outside of the championship zone with a projected solid defense, but sub-standard offensive efficiency. Once the season started, the Spartans' performance through the PK85 tournament placed them just inside the lower left-hand corner of the championship zone.
But, the performance over the past five games has resulted in the data shifting to the left (indicating offensive regression). Michigan State's current data profile is virtually identical to that of the ill-fated 2011 team who lost in the first round of the NCAA Tournament UCLA in Kalin Lucas and Durell Summers' final year. That is the bad news.
There are still two solid months of basketball in front of the Spartans with plenty of chances to improve. How much improvement can be expected? That is what is indicated by the error bars and green oval surrounding Michigan State's current data point. This oval represents the average change in Kenpom efficiency for all teams over time at this point in the season. In other words, it is the expected uncertainty in the current data relative to the values expected on Selection Sunday.
The oval suggests that if Michigan State can improve, especially on offense, they can return to the lower corner of the championship zone. It would be reasonable to optimistically project that the 2023 Michigan State team could evolve into a team with a similar profile to the 2010 Final Four team. This feels like the best-case scenario, based on the current data.
Michigan State's current statistics are very likely being impacted by the combination of the brutal early schedule and the injuries to Malik Hall and Jaden Akins. If both players are able to return to full strength, it is reasonable to assume that the Spartans can approach the ceiling that was discussed above.
However, the position of the current Michigan State data point should have fans a little concerned. On some level, a team is who the data says that they are. To data in the 2022-23 season, Michigan State has played like a bubble team which is unlikely to survive the first two rounds of the NCAA Tournament.
The roster, the coach, and the on-court performance against teams such as Kentucky and Gonzaga, suggest that better things are on the horizon for this Spartan team. But nothing in life is guaranteed, and there is certainly a lot of work left to be done.
Comments
Post a Comment