The Limitations of Season-Long Analytics: A Look at North Carolina's Championship Game Run

by Hollywood Sports

Before the college basketball season started, I vowed to try to pay attention to teams that made significant advances during the season and into March. I had three reasons why the 2021-22 season would be unique in this regard.

(1) The first full season of normality after the COVID-impacted 2020-21 campaign. The previous season was chaotic for all college basketball teams. Practice sessions were different with coaches and players navigating the rules of the road regarding what was safe and appropriate during a pandemic. Players were frequently unavailable to practice and/or play in games if they tested positive for COVID. These necessary disruptions impacted player and team development. While things were not back to pre-COVID levels of normalcy, the easing of COVID regulations provided coaches the opportunity to better develop individual skills and team cohesion. 

(2) The NCAA’s decision to grant everyone an extra year of eligibility would create some intriguing teams with veteran experience. The NCAA did not deem any collegiate player who competed during the COVID-impacted 2020-21 season to lose that year of college eligibility. This allowed many teams to bring back fifth-year seniors and even six-year “super seniors” who had already been granted an extra year of eligibility earlier in their collegiate careers. These teams who may lack the elite talent of the blue-chip programs could make up the divide with older groups with depth — and their experience may help them peak once March hit with single-elimination tournaments. Miami, Florida was a great example of this phenomenon as they were led by Kameron McCusty and Charlie Moore as super sixth-year seniors who led a starting five consisting of four seniors. Despite being a #10 seed, the Hurricanes reached the Elite Eight before losing to the eventual National Champion Kansas Jayhawks. 

(3) The opening up of the transfer portal is changing college basketball, irrespective of the after-effects of the pandemic. Most coaches say that the new transfer window rules have had the biggest impact on how college basketball operates. Good for the players (and boo-hoo for the well-paid head coaches whose jobs became more challenging). How transfers will fit into new rosters and systems is not something that is known in the early games in November. Coaches are building a team identity for March and the single-elimination tournaments at the end of the season. 

College basketball teams have always improved or regressed as the season goes on. Evaluating season-long team data and analytics should always be taken with a grain of salt. But, now more than ever, even the fantastic data provided by folks like Ken Pomeroy needed to be treated with additional scrutiny simply because it might include statistics that changed as qualitative adjustments were made to the individual teams in question. 

Frankly, I missed it on Miami (FL) this season. I did not appreciate how good they would be in the NCAA Tournament (and too quickly dismissed their results as a result of a favorable road to the Elite Eight). Fortunately, I finally figured out North Carolina — but it took their victory against a very good UCLA team for me to finally appreciate what rookie head coach Hubert Davis has done with this Tar Heels team. 

I thought that Bruins' team that went to the Final Four last year was really good — but North Carolina rallied from a 56-51 deficit with under nine minutes to go to cleanly win by a 73-66 score. A 22-10 run to close out that game was impressive. A mistake the analytics community in all-sports can make is not knowing when to pivot away from season-long stats to privilege recent data — ask those bettors who did not figure out the Cincinnati Bengals. One of the power rankings systems I use considered the Tar Heels the best team in the nation going into the Final Four weekend in terms of Net Adjusted Efficiency over their last ten games. Those numbers also ranked North Carolina with the fifth-best defense in terms of Adjusted Efficiency since February 17th (before the Final Four). 

What made an impression on me was the team balance Davis had fostered. Davis’ decision to have R.J. Davis take over more of the ball-handling duties to let Caleb Love play more off the ball as a wing made a significant difference. Four different players led them in scoring in each of their first four NCAA Tournament games. Brady Manek scored 28 points in their opening win against Marquette. R.J. Davis poured in 30 points in their dethroning of the reigning champion Baylor. Caleb Love scored 30 points in the win against UCLA. And then Armando Bacot scored 22 points and added 20 rebounds in their 20-point win against St. Peter’s. It was then Love on in the Final Four showdown with Duke who took charge for this team by scoring 28 points against the Blue Devils. What these Tar Heels could do that too many of the teams under Roy Williams could not is make their 3s. North Carolina saw at least one player nail at least three 3s in each of their games in the Big Dance. And while them taking 38.0% of their shots from behind the mark is just above the 37.7% national average, there were many years under Williams where they were not taking more than 30% of their shots from 3-point range. 

The Tar Heels were rolling — and I expected the momentum to continue after the continued confidence they have after beating a Duke team loaded with future NBA talent. North Carolina was our 25* College Basketball Game of the Year in the National Championship Game against Kansas. After taking a 38-22 lead late in the first half, the Tar Heels blew the lead in the second half but, fortunately, held on to cover the point spread getting +3.5 to +4 points in most spots. The oddsmakers and the market evaluated North Carolina for their season-long efforts — and the winning difference (by just a point!) was appreciating the improvements the Tar Heels made along the way that made a different and better team. 

A good lesson moving forward. 

Best of luck — Frank.

All photographic images used for editorial content have been licensed from the Associated Press.

© 2022 Al McMordie's BigAl.com. All Rights Reserved.