The All-Encompassing Master Bracketnomics Paradigm–2023
Note: This Bracketnomics Tutorial makes past ones on our site obsolete.
One year ago, we issued our updated bracketnomics paradigm and predicted Kansas to win the National Championship. We also selected Villanova to make the Final Four. Yet, we decided to make considerable changes to this year’s paradigm. Why is this, you may ask?
For the last several weeks, we have spent hours researching the changes in college basketball over the last few years. The age of analytics has changed the game just as much as it did with baseball. Five years ago, a small handful of coaches even recognized analytical data. Today, almost every college team not only realizes its importance; they have staff on hand to analyze the data and use it to best implement their game plans.
In baseball, on-base percentage is the most important offensive factor, with slugging percentage in second place. Pitchers’ fielding independent pitching stats are the true signs of their effectiveness. In basketball, there is a tier of four different stats and how teams best implement plans to maximize these tiers.
First, let’s break the game down into its proper components. Basketball is a continual game, where the two teams have the ball basically the same number of possessions per game and trade off on those possessions in a typical time frame of every 10 to 30 seconds (offensive rebounds can extend these possessions).
When a team has the ball, their obvious mission is to score points, and when the other team has the ball, the obvious mission is to prevent the other team from scoring points and gaining possession of the ball.
Offensively, the highest expected points per possession comes from connecting on a 3-point shot or getting a 2-point shot opportunity with the highest possible percentage chance. Thus, 3-point shots and very close 2-point shots have taken over the game. And, preventing 3-point shots and very close 2-point shots has become more important defensively.
But, if a team can get more opportunities to shoot these shots and conversely keep the opponent from having these opportunities to shoot these shots, the overall effect is almost as important as the ability to shoot these shots.
What all this means is that we have spent hours scrutinizing what works today in college basketball and what may be not as important as one might think. Thus, we have thinned out our criteria this year, getting rid of old criteria that no longer portends success, while giving more weight to what does lead to success. Here, then, is your primer for your 2023 Bracketnomics PhD. If you already own a Bracketnomics PhD from taking this course in the past, then think of it as continuing education to keep up to date on the latest methods.
There are four major criteria to look at in each team’s resume. The national champion will come from the teams that best meet these four criteria.
Criteria #1: Adjusted Offensive Efficiency
How many points does a team average per possession when it is adjusted to reflect the average defense? If two teams both average 1.1 points per possession, but one team does it against teams that typically give up 1.15 points per possession, while the other does it against teams that typically give up 1.05 points per possession, these identical 1.1 points per possessions are not the same. Efficiencies must be adjusted to an equal scale for all teams.
What Matters: Teams with adjusted offensive efficiencies in the top 20 and especially in the top 10 tend to win the national championship most years. Kansas was #6 last year. Baylor was #2 in 2021. Virginia was #2 in 2019. Villanova was #1 in 2018. North Carolina was #9 in 2017. The last time a team won the national title and was not in the top 10 in adjusted offensive efficiency was Connecticut in 2014, when they were #39.
Criteria #2: Adjusted Defensive Efficiency
Many people believe in the old axiom that defense wins championships. Defense is more important than offense in baseball, but not by a lot. In basketball, offense is more important than defense, but also not by a lot. An A+ offense and C defense will beat a C offense and A+ defense about 58 to 63% of the time, not a slam dunk, but definitely measurably superior. Still, even though offense is more important, defense must be considerably better than average. We have found that being in the top 15% of all adjusted defensive efficiency suffices for picking teams to get to the Final 4, but being in the top 20 is necessary to cut the nets down as “One Shining Moment” plays. In the 21st Century, the weakest adjusted defensive efficiency was #22. All the others were in the top 15.
Criteria #3 Schedule Strength (eye on specific leagues)
When was the last time that the national champion came from a conference outside the major powers? How about UNLV in 1990? Yes, the last 31 national champions (no tournament in 2020) have come from teams that are in the ACC, Big 12, Big East, Big Ten, Pac-12, or SEC. To parse this a little more, no Pac-12 team has won the title since Arizona in 1997, and no Big Ten team has won the title since Michigan St. in 2000.
The ACC, Big East, Big 12, and SEC have dominated the last 21 national titles. Additionally, all the title winners had schedule strengths 7 points stronger than average and greater.
We believe that the Big Ten has had troubles, because the league stresses half-court possession control and does not emphasize fast breaking or pressure defense that forces turnovers and creates steals. As you will see later, teams that have the ability to go on big scoring spurts have the advantage in the NCAA Tournament. Possession basketball might work in the regular season in the Big Ten, because all the teams play it. Gone are the days when Big Ten teams led the nation in total possessions, scoring, forced turnovers, and steals. We have talked at length in the past about how successful former Purdue coach Gene Keady was, until March 15, when his Boilermakers made it to the Sweet 16 in their better years but couldn’t make the Final Four and only once made the Elite 8 one time, when they had the best player in the nation and a number one seed. Keady’s top four assistant coaches that became head coaches suffered the same fate, as Steve Lavin couldn’t do it at UCLA; Cuonzo Martin couldn’t do it at multiple schools; Kevin Stallings couldn’t do it at Vanderbilt; and so far Matt Painter hasn’t done it at Purdue.
If a team is in the ACC, Big 12, Big East, or SEC and has a schedule strength at least 7 points stronger than average, they meet this criteria. Don’t completely throw out the Pac-12 and Big Ten teams (or Gonzaga), because eventually, one of these teams will win the national championship. Purdue, Gonzaga, UCLA, and Arizona aren’t going to be easy outs this year. They just must buck a mighty big trend to get there. You could have made good money betting against the Boston Red Sox in the playoffs for many years, until sabermetric patriarch Bill James and boy genius Theo Epstein realized what the Red Sox needed to win. The Big Ten, Pac-12, and Gonzaga might eventually have their 2004. The odds might be low this year, but there’s a chance, and getting to the Final Four is not a long shot.
Criteria #4 PiRate Ratings R+T Factor
We are big fans of CBS basketball commentator Clark Kellogg. In the early 1990’s Kellogg coined the term, “spurtability,” where he described some teams’ ability to consistently go on scoring spurts to put games away. One team might trail another by three to five points for a good piece of the game, and then one scoring spurt by the trailing team gives them the lead that they never relinquish. It happens frequently, especially in tight games. It happens in the NCAA Tournament with great regularity.
The first great example of such spurtability came in legendary coach John Wooden’s first national championship at UCLA in 1964. The Bruins, with no starter over 6 foot 5, used an incredible zone press defense to force turnovers, gain steals that led to fast break points, limited opponents to one shot, and then got extra scoring opportunities by crashing the offensive boards. UCLA went 30-0 that year, and they had big scoring spurts in all 30 games. Most notably, the 1964 Championship Game presented the perfect example for spurtability. Duke was favored to win the title. They had two starters at 6 foot 10 plus an All-American tall guard that could shoot from outside and drive the ball to the hoop.
The game started looking just like the experts predicted, as Duke’s extra size and muscle allowed the Blue Devils to work the ball inside for high percentage shots, while forcing UCLA to play from the perimeter. The Bruin press was handled easily, and Duke led 30-27 with about 6 minutes to go in the first half. UCLA then went on a 16-0 run in less than 2 minutes! The press totally disrupted Duke’s offense, and the Bruin fast break looked like an NBA All-star game. The game was over before halftime.
We have analyzed what leads teams to enjoy winning scoring spurts. It almost never happens in possession basketball, where the spurting team scores on six consecutive half-court possessions, while the other team fails to score on six consecutive half-court possessions. Rebounding and Turnovers bring on scoring spurts. Seldom does a team take one shot per possession for multiple possessions and goes on a scoring spurt. They need multiple shots per possession and/or they need to prevent the opponent from getting any shots and if so, just one.
The PiRate R+T factor uses a unique set of formulas to see the spurtabilities of teams. For years, we had one R+T that used counting stats (rebound margin, average number of steals per game, and turnover margin), and we found that the teams with the highest R+T were the one’s advancing the deepest into the tournament. But, as we realized that rates were more important than counting, we also devised a rate formula using offensive rebounding rates, steal rates, and turnover rates. This places a team that averages 75 possessions per game on the same playing field with a team that averages 65 possessions per game.
We have two formulas, one for old R+T and one for the new rate-generated R+T. Both have their merits. The formula for old R+T is quite simple: (R*2)+(0.5*S)+(6-Opponents S)+T, where R is rebounding margin, S is steals, and T is turnover margin. When a team has a rating of 10.0 or better, they have a good chance to have a winning spurt. When the number is 15.0 or better, opponents better watch out, because a spurt is almost assured when playing this team. If the number is 20.0 or better, this is a team that typically goes on one winning spurt every game and only loses to teams that can do the same.
Of course, this R+T rating must be used in step with schedule strength, because a Big Ten team can much easier go on a spurt against a Summit League team than it can against a Big 12 team.
The new Rate R+T formula is quite lengthy, so we won’t attempt to explain it. Anything over 4.0 is considered strong. Anything over 8.0 is considered very strong, and anything over 12.0 is considered lethal when taken in context with schedule strength.
The other important part of this is how to treat a team with a negative old R+T rating. These teams are ripe to be upset early in the tournament. If the team has a new R+T rating below -3, they are also in jeopardy of losing early in the tournament. It is low R+T that has doomed so many Big Ten teams in recent years, especially Purdue and the teams coached by Keady proteges.
The remaining criteria are secondary to the four primary criteria, but they are important when looking at games between teams that are evenly matched in the primary criteria.
Criteria #5: 3-point Shooting Percentage
For years, we believed that the teams that relied on the 3-point shot didn’t win in the Sweet 16 to the Championship Game, because the larger arenas with weird sitelines and the tighter rims were not good for these teams. However, as analytics began to change the game, and players became more 3-point shooting conscious (copying the NBA), this proved to not be the case.
Note that this criteria makes no notice of the number of 3-point attempts taken. Only the percentage matters. If a team can hit better than 35% from behind the arc, defenses must widen, and thus they are weaker against the high percentage inside shot. If a team shoots less than 33.3% from behind the arc, defenses can tighten and keep the ball away from the 65% scoring area near the basket.
Criteria #6: Offensive Rebounding Percentage
This is covered in other criteria, but it is important enough to isolate it into its own category, because there is history showing the point where offensive rebounding can lead a team to extra victories. At 35%, a team that shoots 43% can get 10 extra points on second chance shots. The cut-off number here is 30%. If a team cannot average 30% offensive rebounds, they are not going to make it far in the Big Dance. If a team has an O-Rebound rate of 37.5% or higher, they are dangerous.
Criteria #7: 2-point Percentage Defense
While we are interested in the offensive 3-point percentage, on the defensive side, it’s the ability to prevent made 2-point shots that is more important. If a team gives up more than 52.5% inside the arc, they are going to have to drop defenders in the paint, leaving the 3-point line open, where good shooters will hurt them.
When a defense limits opponents to less than 46% from inside the arc, their defense is tough, tough enough to keep winning in the Big Dance if their offense is better than average. The important number is 48.5%. Most national champions have met this requirement.
Criteria #8: Free Throw Rate
We were late coming to this party. For many years, it was our belief that teams that relied on hitting free throws to win games in the regular season did not succeed in the NCAA Tournament, because the games were officiated differently and fewer fouls were called. The NCAA eventually made some changes in game officiating, where most games are now called the same way, and officials are graded just like players and coaches. Now, the frequency of fouls called in the Big Dance is about the same as what is called in the regular season.
The next part of this criteria that we had to come around to supporting was how it was calculated. The original FT Rate was simply FTA/FGA. How many of these free throws were made did not matter, as the formula was merely a rate of how many free throw attempts a team made compared to field goal attempts.
There were two other schools of thought on this. Some metrics experts changed it to FTM/FGA. Others, including us, went with free throws made per 100 possessions.
But, it soon came to us in a brainstorm that the original formula or FTA/FGA is the best of the three. There’s more to a foul than just making foul shots. Fouls limit playing time. If a star player commits two quick fouls in the first four minutes of a crucial game, he’s likely to miss many minutes in the first half. Losing a star player in this situation is worth many more points than the possible foul shots. Additionally, players in the game in foul trouble must watch how they play defense and how they drive toward the basket on offense.
A free throw rate (FTA/FGA) of 28.5% or better is important in the NCAA Tournament. Above 32% is really good. Above 37.5% is remarkable.
Criteria #9: Scoring Margin
Very few teams have ever won a national championship with a scoring margin below 8 points per game, and a majority of national champions have had scoring margins above 10. When you combine a double-digit scoring margin with a schedule strength at least 7 points above average, this is a dangerous team. Teams with scoring margins under 5 points per game rarely win four NCAA Tournament games.
Criteria #10: A Lengthy Winning Streak
To get to the Final Four, teams have to win four consecutive games against very strong competition. To win the national title, they must win six consecutive games, four of which must be against great teams.
Do you expect a team that hasn’t won six games in a row during the regular season, where many of those games were against subpar teams, to become capable of winning six games after March 15?
A 10-game winning streak is typical of most national champions of the past, and two streaks of six are also markers of teams capable of going deep in the tournament.
Criteria #11: Preseason Top 25
You might think that being ranked in the top 25 in the preseason means absolutely nothing, but the sports media has experts. They know which teams have all the talent, and even though North Carolina just failed to make the field after beginning the season ranked #1, there is no reason to throw this stat aside. It’s been many years since a national champion wasn’t ranked in the preseason.
Criteria #12: Regular Season or Conference Tournament Champion
It’s been many years since the national champion didn’t win a major college conference regular season or tournament title. Even the biggest outlier champion, 2011 Connecticut won the Big East Conference Tournament as a #9 seed, winning five games in five days.
Criteria #13: Coach With NCAA Tournament Experience
Occasionally, a first year coach will direct his team to the NCAA Tournament and have some success, but it doesn’t happen often. Bill Hodges took a team to the Championship Game in his first year as a head coach, but Indiana State with Larry Bird might have made it to the Final Four with the head coach of Terre Haute High School coaching the team. Steve Fisher won a national championship at Michigan in 1989 when he began the tournament with a career record of 0-0.
On the other hand, a coach like Bill Self, Tom Izzo, and Mick Cronin has been there so many times, there are no unexpected variables that arise and throw their teams off sync. They know the process and can better prepare their teams through the extra long timeouts, the pressure of being on the national stage, and even how to better negotiate the logistics of the tournament.
Criteria #14: 3 Upperclassmen Getting 24+ Minutes Per Game
This criteria is almost null and void these days, because thanks to the NCAA giving an extra year to all players that played during the Covid season, almost every team is loaded with upperclassmen. There are a couple of teams with youth and inexperience, but you won’t find many. Once this waiver is gone, this will be important again.
Criteria #15: An Inside Power Game With 2 Starters Combining For 20 & 12
If a team has two starters that combine to average 20 or more points and 12 or more rebounds per game, they can bang it inside at crucial junctures of the games, when an inside power play is important. A team can win with a great perimeter game, but eventually, they are going to need some inside points and cheap baskets.
Criteria #16: Three or More Go-to Guys
If a team relies on one or even two players to do most of their scoring, and the star has an off game, or the opponent can shut him down, they aren’t likely to win at this level. If a team has three or more players that can carry the team on their backs, it is almost impossible to stop all three in any game. A star like LeBron James might be able to take his team to the NBA Finals, but remember that everybody on an NBA roster is capable of scoring a bunch of points on a given night. At the college level, stopping a one-star team is possible and happens every year in the NCAA Tournament. Remember that the great Michael Jordan had James Worthy and Sam Perkins on his North Carolina team. Kareem Abdul Jabbar was surrounded with players like Lucius Allen, Sidney Wicks, and Curtis Rowe at UCLA, and Jabbar’s stats were lower than they could have been had UCLA not been balanced elsewhere.
The Pi-Rate Ratings 