Ever watch a game where a team wins by twenty, yet the guys on Twitter are acting like they just saw a car crash? You look at the box score. 80-60. Seems fine. But then you see the "nerd" metrics—the college basketball advanced stats—and suddenly you’re told that the winning team actually played like garbage.
It feels pretentious. Maybe even a little annoying. But honestly, if you’re still trying to understand this sport using just points per game or field goal percentage, you’re basically reading a book with half the pages ripped out. The game has changed.
The 2025-26 season has been a total fever dream. We’ve seen scoring averages jump by nearly five points across the board compared to last year. Is everyone suddenly Steph Curry? No. They’re just playing faster. If you don't adjust for tempo, you’re going to think every team in the Big 12 is an offensive juggernaut, when really, some of them are just sprint-and-pray squads.
Why Your Eyes Are Lying to You
Humans are suckers for "clutch" shots. We remember the buzzer-beater, but we forget the four possessions in the first half where a team settled for a contested, fadeaway long two-pointer. That’s where ShotQuality comes in.
I’ve been tracking their data this season, and it’s wild how often the "actual" score is a total fraud. They use computer vision to calculate what a shot should have been worth based on the shooter's ability, the defender’s proximity, and the play type.
💡 You might also like: Oilers Rumors and News: What Most People Get Wrong About Edmonton’s Deadline Strategy
Remember that Duke-Arkansas game on Thanksgiving? The one with nearly 7 million viewers? If you just looked at the final score, you might miss the fact that one team was essentially living on "prayer" shots. Advanced metrics tell us who is actually good versus who is just getting lucky with their shooting variance.
The KenPom Gospel
You can’t talk about this stuff without mentioning Ken Pomeroy. His site, KenPom, is the industry standard for a reason. He basically pioneered the idea of "efficiency margins."
Instead of looking at how many points a team scores, he looks at Adjusted Offensive Efficiency (AdjO)—how many points they’d score against an average defense over 100 possessions.
- Pace matters: Some teams, like Michigan under Dusty May this year, want to fly. Others want to grind the game into a halt.
- The "Pythagorean" Wins: This sounds like high school math because it basically is. It’s a formula that predicts a team’s winning percentage based on their scoring margin, not just their record.
This year, Michigan has surged to the #1 spot on KenPom because they aren't just winning; they are systematically dismantling teams. They are currently a consensus top-four team across almost every model, including BartTorvik and EvanMiya.
The Rivalry of the Models: KenPom vs. Torvik vs. EvanMiya
It's kinda funny how fans pick their favorite "stat guru" like they’re picking a political party.
BartTorvik is the choice for the hardcore "what have you done for me lately" crowd. Why? Because Torvik allows you to filter by date. If a team lost their star point guard in November but has been on a tear since January 1st, Torvik’s "Recent" filter will show you they’re actually a top-10 team, while KenPom might still have them at 25th because he weights the whole season more heavily.
Then there’s EvanMiya. He’s doing some cool stuff with Kill Shots—that’s when a team goes on a 10-0 run. It’s a metric for momentum and "knockout" ability. He also breaks down individual player impact much better than the older sites. If you want to know which transfer from the portal is actually carrying a team's defense, Miya is your guy.
Take NC State this season. Depending on which site you check, they fluctuate between 24th and 35th.
- EvanMiya: Ranks them 24th.
- KenPom: Ranks them 25th.
- BartTorvik: Ranks them 35th.
Why the gap? Torvik thinks their defense is a bit of a sieve (64th), while Miya’s model thinks their defensive floor impact is much higher. Who's right? Honestly, usually the truth is somewhere in the middle, which is why most "bracketologists" use a composite of all three.
The NET Ranking: The Only One the Committee Cares About
You've probably heard the term "Quadrant 1 win" about a thousand times during a broadcast. That's the NCAA Evaluation Tool (NET).
The NET is a bit of a black box. The NCAA doesn't release the exact formula, but we know it’s a mix of Team Value Index (results against strength of schedule) and Adjusted Efficiency.
It’s not meant to be a "who would win today" ranking. It’s a "what is your resume" ranking.
This is where things get messy. A team can be #15 on KenPom (meaning they are very strong) but #40 in the NET because they haven't played anyone good yet. This is exactly what’s happening with some Big Ten teams right now. They’re blowing out "cupcakes" in the non-conference, which the computers love, but the NET committee won't reward them until they beat a real opponent on the road.
Breaking Down the Quadrants
- Quad 1: Home games vs. 1-30, Neutral vs. 1-50, Away vs. 1-75.
- Quad 2: Home vs. 31-75, Neutral vs. 51-100, Away vs. 76-135.
- Quad 3: Home vs. 76-160, Neutral vs. 101-200, Away vs. 136-240.
- Quad 4: Everything else.
If you lose a Quad 4 game, your season is basically on life support. Just ask some of the SEC teams that have struggled early this year. The SEC was hyped as the best conference ever, but the college basketball advanced stats suggest they might be a bit overrated this time around.
The Transfer Portal Has Broken the "Continuity" Stat
It used to be that you could look at "minutes returning" and guess who would be good. Not anymore.
With the portal, a coach can flip a roster in 48 hours. This has made preseason advanced stats almost worthless. However, once we get about 10 games into the season—usually right around Christmas—the noise starts to clear.
One trend we're seeing in 2026 is the "Super-Senior" effect finally fading, but being replaced by "Professionalized" rosters. Teams like Arkansas and Kentucky are essentially built like NBA G-League squads. Their Effective Field Goal Percentage (eFG%)—which gives extra weight to three-pointers—is through the roof because they aren't recruiting projects anymore. They’re recruiting finished products.
Actionable Insights for Your Bracket and Betting
Look, if you want to actually win your pool or just sound like you know what you’re talking about at the bar, stop looking at the AP Poll. It’s a beauty contest.
1. Watch the "Kill Shots": Check EvanMiya’s data for teams that go on 10-0 runs. Those are the teams that can erase a deficit in three minutes in March.
2. Fade the "Lucky" Teams: If a team is winning but their ShotQuality "Expected Score" says they should have lost, they are a prime candidate for an upset.
3. Efficiency Margin is King: Every national champion since 2002 (except for 2014 UConn) has been in the Top 20 for both Offensive and Defensive Efficiency on KenPom by Selection Sunday. If a team is 5th in offense but 90th in defense? They are a "fake" contender.
The 2026 season is proving that pace is the new frontier. Teams are averaging 67.6 possessions per game, up from 65.1 just a year ago. That’s a huge jump. It means more fatigue, more bench depth requirements, and more variance.
Go to BartTorvik.com. Set the date to the last 30 days. See who is actually playing well right now. That’s how you find the next Cinderella before everyone else starts talking about them.
Stop counting points. Start counting efficiency. The numbers don't lie, even when the scoreboard does.