Is the NET ranking system working the way it’s supposed to?

But one factor plays an especially important role in this process: the NCAA Evaluation Tool or NET. That’s because the NET is the “primary sorting tool” used to evaluate teams and the only computer-ranking system provided by the NCAA to committee members to aid in their decision-making.

Aware that many coaches, players and fans tend to be obsessed by their team’s NET ranking, the NCAA goes to great lengths to emphasize that the NET is only one of many tools used by the committee to select and bracket the NCAA tournament.

“I think it’s important to note that our committee has 11 different selection criteria,” Rick Nixon, associate director of media relations and statistics, told The Next. “[And while] the NET is certainly important, each committee member is going to look at teams differently in regards to those eleven selection criteria.”

In addition to the NET, the other 10 selection criteria include (in alphabetical order): bad losses, common opponents, competitiveness in losses, early performance versus late performance, head-to-head competition, observation of performance (the “eye test”), overall record, regional rankings, significant wins and strength of schedule.

Many coaches admit they either don’t fully understand the NET or remain mystified about how it works.

“I mean, the NET never ceases to amaze me,” Baylor head coach Nicki Collen told reporters after her Bears beat Colorado by 22 points in Boulder on February 19. “I felt like I had a pretty good grasp of the RPI, and it’s kind of amazing. We’ll win and drop two spots [in the NET].”

---

Order ‘Becoming Caitlin Clark’ and save 30%

---

Seton Hall head coach Tony Bozzella has studied the NET and discussed it with NCAA officials. Does he believe he understands how it works?

“I don’t,” Bozzella told The Next. “And here’s the reason: We had someone come in to try and explain it. But no one really knows what the metric is [for margin of victory]. Does it cut off at 25, does it not cut off at 25? Do you sub [at the end of the game]? Do you not sub? [They said], ‘Subbing a little at the end of the game is not going to affect your metric.’ So I said, ‘When does it affect your metric?’ And, you know, they gave us certain ideas, [about] points per possession, the quality of your shooting defense, [but] at the end of the day, you have a computer basically saying what is a good win and what’s not a good win, as opposed to just winning the game.”

The NET originally was created and introduced by the NCAA for use in selecting the men’s NCAA Tournament bracket in 2019. Starting with the 2020-21 women’s college basketball season, a modified version of the NET was incorporated into the selection process for the D-I women’s tournament, replacing the Ratings Percentage Index or RPI. 

The RPI was a relatively straight-forward algorithm that took into account three factors for each team: winning percentage, opponents’ winning percentage and opponents’ opponents’ winning percentage. Note the common denominator in all three of these components: winning.

According to the NCAA, the NET was created to provide a more “comprehensive and contemporary” sorting tool to the committee. Comprehensiveness was added by, among other things, taking into account the difference between home, road and neutral court contests. The model became more contemporary by incorporating a form of “machine learning” into the algorithm that constantly updates the rankings with data not only from every game, but from every possession of every game.

But the main innovation of the NET was to infuse into the the rankings what the NCAA calls “Adjusted Net Efficiency.” According to the NCAA, Adjusted Net Efficiency reflects the “difference between a team’s offensive efficiency (points per possession) and defensive efficiency (opponents’ points per possession), adjusted for opponent strength and game location.”

In simple terms, Adjusted Net Efficiency rewards a team for scoring every time it has the ball. Likewise, it rewards a team for preventing its opponent from scoring every time they have the ball. According to the NCAA, this makes sense because the intrinsic goal of basketball is to play efficiently, that is, to score any time you can and to prevent your opponent from scoring.

But not every coach agrees with the NCAA that playing efficiently is intrinsically important.

“I think it makes no sense at all,” Seton Hall’s Bozella told The Next. “I think winning and losing [are] the most important criteria.”

Bozzella points out that teams play with different styles and not all of them are geared toward how the NCAA defines offensive and defensive efficiency.

“Listen, some people play totally different. Like we [at Seton Hall], we know we’re going to shoot the ball bad, so we crash the boards. We want to get more shots up, because we don’t shoot the ball particularly well. So we think volume is better.”

Another problem with the NET is the perception that it encourages coaches to run up the score, even when doing so might run counter to a coach’s instincts or long-term interests.

“[You leave your] starters in longer and score faster,” say Bozzella. “You know, in the old days, when you’re up 25, you used to slow the game down, get some subs in there, reverse the ball, not shoot for the first 15 seconds. Now you have coaches trying to score quicker.”

The NCAA resists the idea that the NET encourages coaches to run up the score. According to a NET explainer document published on the NCAA’s website, “A common misconception is that teams can boost their NET rankings by running up the score. However, NET does not factor in margin of victory in a way that encourages this.” (Bolded language in original). 

While the NCAA admits that running up the score may have some effect on a team’s NET ranking, the organization believes the impact is muted by taking into account the entire sweep of a team’s season.

“You’re not going to have 32 games where you’re able to run up the score,” says the NCAA’s Nixon. “I mean, every possession is factored in.”

---

Want even more women’s sports in your inbox?

Subscribe now to our sister publication The IX and receive our independent women’s sports newsletter six days a week. Learn more about your favorite athletes and teams around the world competing in soccer, tennis, basketball, golf, hockey and gymnastics from our incredible team of writers.

Readers of The Next now save 50% on their subscription to The IX.

---

But this explanation does not always comport with coaches’ actual experiences. For example, On Feb. 26, the Colorado Buffaloes celebrated senior night in Boulder by pummeling Arizona State by 35 points, 89-54. The following day, the Buffs’ NET ranking ballooned by eight spots, from 65 to 57, lifting CU into contention for a possible at-large berth.

When I pointed this out to CU head coach JR Payne two days later, she expressed disappointment. “I don’t like that,” Payne told The Next. “You’re rewarded for blowing teams out. I mean, I don’t like it, because I think depending on the mindset of the coach, it eliminates the opportunity for your young players to play in a game where they could come in and contribute, you know?”

Most coaches I’ve spoken with about this issue claim that while they are aware of the incentive to run up the score, they usually resist the temptation to do so for fear of wearing down or risking injury to their best players. 

After watching a 38-point lead on the road erode into a 22-point win at Colorado on February 19, Collen claimed she wasn’t worried about the impact on her team’s NET ranking when she made a decision to rest her starters for nearly all of the fourth quarter. 

“Actually, I wish I had thought about that a little bit,” Collen told reporters after the game. “I’m thinking about Nettie [Vonleh] having to guard [Audi] Crooks [at Iowa State] on Saturday. I’m thinking about, we’ve got three games left, and we have to play Iowa State, Kansas State and TCU, and we’re competing for a conference championship. And so, you know, I feel like that stuff takes care of itself over time.”

But in a weekly media availability last month with reporters, Columbia Lions head coach Megan Griffith threw cold water on the notion that coaches will always resist the temptation to run up the score to boost their NET ranking. “I think any coach would lie to you if they said they didn’t [consider the NET], especially a team that’s on the bubble.”

Even if it were true that it takes multiple blowouts to move the NET dial, many teams are doing just that. Take the example of Harvard University in the Ivy League. Although Harvard finished third in the Ancient Eight standings this season, the Crimson blew out many of the weaker opponents on its schedule, including Boston University by 60 points, Northeastern by 52 points and Yale by 56 points.

Harvard has also consistently maintained the highest NET ranking of any Ivy team throughout the season and currently sits at 37 in the NET, four spots above first place Columbia and eleven spots higher than second place Princeton, which swept Harvard in both games the teams played during the regular season. Both Columbia and Princeton have higher rated strengths of schedule (SOS) than Harvard, whose 117th ranked SOS is the lowest in the Ivy League.

Harvard coach Carrie Moore understands the importance of touting her team’s gaudy NET ranking. In making a case for her team receiving an at-large berth, Moore emphasized one point to reporters at her weekly media availability last Wednesday: “Our team has not been ranked any lower than 40 in the NET all year long.”

Having a system that rewards running up the score can act like a double-edged sword. The teams that run up the score may boost their own NET rankings, but teams on the other side of the beat down pay a heavy price. On January 19, Bozzella’s Seton Hall team traveled to Storrs to play UConn and was demolished by 60 points, 96-36.

“We lost by 60 points at UConn,” said Bozzella. “They played great. We played bad. I subbed a lot because we had kids with the flu. We only had just 10 kids, and I didn’t want to get our stars hurt for the next game. So we lost. We went from 57 to 71 [in the NET]. So, don’t tell me margin didn’t matter because we lost by 60, and again, could I have lost my 40? Yeah, probably, but I didn’t want to risk losing the next game. And that’s my point. You have to look at the whole picture, and [dropping from 57 to 71] has taken me forever to bounce back from.”

When the NCAA is confronted with the concern that the efficiency component of the NET encourages teams to run up the score, the organization retreats to a familiar refrain: “We’ve heard the complaints,” says Nixon. “But, the [NET] ranking is just one of 11 selection criteria.”

To some extent, Nixon is right to point out that the committee is capable of separating the noise from the signal and looking beyond the NET. Last year, for example, Penn State with a NET ranking of 27 and Washington State with a NET ranking of 29 were left out of the tournament field, presumably because their win loss-records of 19-12 and 18-14, respectively, were too mediocre to merit invitations to the Big Dance.

On the other hand, an Arizona squad sporting a 17-15 record received an at-large bid last year at least in part because the Wildcats enjoyed a NET ranking of 35. Indeed, no team with a NET ranking below 41 failed to receive a bid, with the exception of the aforementioned Nittany Lions and Cougars. 

Like Arizona last year, Maryland, Tennessee, North Carolina and Kansas, all received bids even though they had 12 or more losses and fewer than 20 wins entering Selection Sunday. One important factor these teams shared in common was a NET ranking in the top 38.

---

Order ‘Rare Gems’ and save 30%

Howard Megdal, founder and editor of The Next and The IX, released his latest book on May 7, 2024. This deeply reported story follows four connected generations of women’s basketball pioneers, from Elvera “Peps” Neuman to Cheryl Reeve and from Lindsay Whalen to Sylvia Fowles and Paige Bueckers.

If you enjoy his coverage of women’s basketball every Wednesday at The IX, you will love “Rare Gems: How Four Generations of Women Paved the Way for the WNBA.” Click the link below to order and enter MEGDAL30 at checkout.

---

Another common trait between these teams is their conference affiliations: Arizona, Maryland, Tennessee, North Carolina and Kansas are all members of the power four conferences. 

Since the NCAA moved from the RPI to the NET four years ago, teams from the ACC, the Big 12, the Big Ten and the SEC have benefited disproportionately in terms of improved rankings. 

The RPI rankings, though now disregarded by the NCAA, are still available to the public, which makes it possible to evaluate how teams are ranked in the NET compared to the RPI. The biggest winner in this comparison? The Big 12 Conference. All sixteen members of the Big 12 are ranked higher in the NET this season than in the RPI. On average, Big 12 teams pick up a boost of over 21 spots in the NET compared to the RPI.

The effect is similar for the SEC. All but three of the sixteen SEC members are higher ranked in the NET compared to the RPI with the average bump exceeding 27 spots. Arguably, the biggest winner this season in the switch from the RPI to the NET is the ACC, whose teams on average rank nearly 32 spots higher in the NET than in the RPI. 

Overall, the 68 teams comprising the power four conferences receive an average boost of nearly 23 spots in this year’s NET compared to the RPI.

At the opposite end of the spectrum is the mid-major Patriot League. All 10 members of the Patriot League are ranked lower in the NET than the RPI, and the differential is stark. On average, teams in the Patriot League suffer a loss of 41 spots in the NET compared to the RPI.

Is the NCAA aware of or concerned about the impact the NET has had on the various conferences? Yes and no.

The NCAA annually reviews the NET ranking system and seeks feedback from its members on a regular basis. But the RPI is considered defunct from the NCAA’s perspective, so there’s no effort made to compare or analyze the rankings between the NET and the RPI.

“We don’t look at the RPI at all,” the NCAA’s Nixon told The Next. “We’re totally into the NET rankings and what it provides for us. We don’t compare the two. We think the NET is providing a proper snapshot of what a team is all about, and certainly a lot more data points. There’s a lot more depth of knowledge in regards to teams that the NET provides versus what the RPI provides.”

It’s certainly true that the NET considers a more comprehensive set of data then the RPI — that was a stated goal when the NCAA created the NET. But is the additional data from every possession of every game, and the weight given to that data, actually improving the predictive power of the ranking system?

Some coaches seem to be comfortable with or are at least sanguine about the NET. UNLV coach Lindy La Rocque may need to rely on her team’s strong NET ranking of 46 to win an at-large berth in the event the Lady Rebels fail earn the automatic bid at the Mountain West Conference tournament in Las Vegas this week.

“The NET is a very interesting metric,” La Roque told reporters on Tuesday after a win over Boise State in the MWC tourney quarterfinals. “It has, like, what 37 different things calculated in it? You can go through and pick the ones you like and don’t like. I think efficiency is important, but maybe not as important as wins and losses.”

But other coaches, like Bozzella, remained unconvinced of the value of the NET the way it’s currently configured. “I don’t like the NET metric, because it’s based on what they think is a good win or a good loss or good this or a good that.”

On Monday, Bozzella’s Seton Hall squad suffered another blowout loss to Creighton, 73-44, in the semifinals of the BIG EAST Tournament. Seton Hall’s NET ranking dropped by seven spots the following day. With a current ranking of 76, Bozzella’ Pirates hover on the outer perimeter of the bubble, hoping that something other than their NET ranking will catch the committee’s eye.