Quadrant records and the NCAA selection committee's NET
Quadrant records sort every college game by the NET rank of the opponent and the location. This article explains the four tiers, why a road win counts for more, how the cutoffs create cliffs, and how the committee uses them.
Every March, a few phrases show up in broadcasts and bracket columns that sound like jargon and carry real weight: "Quad 1 wins," "bad losses," "the team sheet." They refer to a system the NCAA introduced alongside its NET ranking in the 2018-19 season to help the men's selection committee judge a team's résumé. The idea is easy to state. Not every win is worth the same, and not every loss costs the same, so sort each game into one of four tiers according to how good the opponent was and where the game was played.
The result is a short table on every bubble team's team sheet: a record in Quadrant 1, a record in Quadrant 2, and so on. A team that is 5-9 in Quadrant 1 and 12-0 in Quadrant 4 tells a different story from one that is 2-3 and 14-1. This article explains how the tiers are drawn, why the location of a game changes its tier, what the system gets right, and where it creates strange incentives. For the ratings behind the NET itself, see College efficiency metrics: KenPom, NET and the ratings used to select the field.
The idea behind the tiers
Think about two wins. One is a home game against the 20th-best team in the country. The other is a road game against the 70th-best team. Which is more impressive? Most coaches would say the second is at least comparable, because road games are harder, and the numbers back that up. Home teams win more often, by a few points on average, as discussed in Home-court advantage in the numbers. A 70th-ranked team at home is a much tougher opponent than a 70th-ranked team in your own building.
The quadrant system builds that logic into a table. Each game is placed in a tier based on the opponent's NET rank on the day the committee looks, and the rank that qualifies for a given tier depends on whether the game was home, neutral or away. The better the opponent and the harder the venue, the higher the tier. Quadrant 1 holds the games that carry the most credit when won and the least blame when lost, and Quadrant 4 holds the games that a tournament-caliber team is expected to win.
The cutoffs
The NCAA defines the tiers by opponent NET rank. The table shows the published ranges, with the lower bound of the bottom tier running to the last team in Division I, a number that changes as the number of Division I teams changes.
| Tier | Home | Neutral | Away |
|---|---|---|---|
| Quadrant 1 | 1 to 30 | 1 to 50 | 1 to 75 |
| Quadrant 2 | 31 to 75 | 51 to 100 | 76 to 135 |
| Quadrant 3 | 76 to 160 | 101 to 200 | 136 to 240 |
| Quadrant 4 | 161 and below | 201 and below | 241 and below |
Read across the first row to see the principle. A home win over the 30th-ranked team earns Quadrant 1 credit, but so does a neutral-court win over the 50th-ranked team and a road win over the 75th. The cutoffs are not arbitrary. They are chosen so that the tiers contain roughly comparable games, taking the extra difficulty of playing away from home into account.
Remember what counts as a "Quadrant 1 opportunity." A team in a strong conference may have a dozen or more chances for Quadrant 1 games simply by playing its league schedule. A team from a smaller conference may have only a few, and may need to arrange them in the non-conference portion of the schedule, often by traveling to a major program's building. That is a theme in College basketball scheduling: non-conference games and neutral sites and in Mid-major basketball: how smaller programs make noise.
How a team sheet reads
The team sheets the committee receives show, for each team, its NET rank, its record overall and in conference, its record in each quadrant, its non-conference strength of schedule and several other ratings. The quadrant records are meant to be read as a pattern, not a total.
A good profile for a bubble team might look like this: a respectable number of Quadrant 1 wins, a Quadrant 2 record well above .500, no losses in Quadrants 3 and 4, and a NET rank comfortably inside the field's range. A concerning profile might have a very good overall record built almost entirely from Quadrant 3 and 4 wins, with no top-tier wins at all. A mixed profile might have one or two Quadrant 1 victories and a Quadrant 3 loss that stands out.
The weighting reflects an asymmetry in how people think about résumés. A win in the top tier is a positive that cannot be erased, because it proves the team can beat the best. A loss in the bottom two tiers is a negative that stands out because it shows the team can lose to teams that should not trouble it. The common phrase "bad loss" is shorthand for a defeat in Quadrant 3 or 4, and "good win" for Quadrant 1 or the upper end of Quadrant 2. The committee has said it looks at the whole picture rather than counting wins and losses mechanically, but the quadrant table is the quickest summary of that picture.
A worked example
Imagine two hypothetical bubble teams, both with 21 wins. Team A played in a strong league. It won three games that fall in Quadrant 1, went 6-4 in Quadrant 2, lost once in Quadrant 3 and won everything in Quadrant 4. Team B played in a weaker league. It has no Quadrant 1 wins, a 3-2 record in Quadrant 2, no losses below that and a long list of Quadrant 3 and 4 wins.
On raw record they are twins. On the table they are different teams. Team A has proved it can beat the best, with a blemish; Team B has proved it can beat those it should and has little evidence about anything harder. A committee member scanning the sheet would pause on Team A's Quadrant 3 loss and on Team B's empty top row, and the discussion would turn on which gap matters more. Notice, too, that nothing in the table tells you whether Team B was timid or merely unlucky in what was available. Perhaps its best non-conference opponent canceled, or a road game it scheduled against a major program was rescheduled. The numbers cannot say, and that is the point at which the committee's judgment and the other ratings come in.
Now change one detail. Suppose Team A's three Quadrant 1 wins came against opponents ranked 28th at home, 49th on a neutral court and 74th on the road. All three sit just inside the cutoffs. If those opponents slip a few places by March, all three wins drop a tier, and Team A's best row shrinks from three wins to none without the team playing another minute. That sensitivity, which the next section describes, is the most important thing to understand about the system.
The cliff problem
Any system that sorts continuous quality into four boxes creates cliffs, and the quadrant system has them. Consider two opponents ranked 30th and 31st in the NET at home. By any reasonable measure they are almost indistinguishable teams. Yet a win over the first is a Quadrant 1 win and a win over the second is a Quadrant 2 win. The line is sharp even though the difference in quality is tiny.
The cliffs matter most for teams near the cutoffs. A bubble team that beat an opponent ranked 74th on the road holds a Quadrant 1 win; had that opponent been ranked 76th, it would be a Quadrant 2 win. Because the opponent's rank is determined at the time of review, not at the time of the game, the same result can move between tiers during the season as rankings shift. A road win in November over a team ranked 60th may look like a prize in the winter and, if that team collapses in February and falls to 90th, sink into a lower tier by Selection Sunday.
Fans often complain that this is unfair, since a team cannot control how its past opponents perform afterward. That complaint is true, but it also points to why the committee is told not to use the quadrants mechanically. A coach whose team beat a strong team that later struggled with injuries has a good case that the win was worth more than the label shows, and the committee can consider that. The tiers are a sorting convenience, not a scoring formula.
What the system gets right
The quadrant approach has real virtues, and they are easier to see by contrast with what came before.
It rewards road wins. Earlier team sheets sorted wins into broad groups, such as victories over the top 50 or top 100 teams. The quadrant system refines that by building the venue directly into the cutoffs, so the difference between playing at home and away is explicit. A team that wins on the road earns more credit, which matches how coaches and players think.
It separates quality of wins from quantity of wins. A record of 25-6 can mean many things. The quadrant table shows how many of those wins were over strong teams and how many were over opponents the team was expected to beat. A team that has nine Quadrant 1 wins is a very different proposition from a team that has none.
It makes losses legible. A loss to a top team on the road is a Quadrant 1 loss and barely counts against a team, while a loss to a weak opponent at home is a Quadrant 4 loss that stands out. That is closer to how a basketball fan thinks about a bad night.
It gives the committee a shared language. A small group of people from different backgrounds, who may have watched very different games, need a common framework. The quadrants provide one, so that discussions about a team begin from the same facts.
What the system gets wrong, or at least oddly
The system has drawbacks, and they are worth naming.
It encourages gaming of schedules. Coaches now think about how many Quadrant 1 opportunities their schedule offers and how many Quadrant 3 and 4 games they must play. A team might avoid a game against a mid-ranked opponent because a loss would be a bad one, or seek a road game against a program likely to finish in the top 75. The incentive to schedule for quadrants, instead of for the best basketball, is the same kind of problem the RPI created, in a milder form.
It disadvantages teams with few opportunities. A team from a smaller conference can win nearly every game and still have almost no Quadrant 1 wins, because there were almost none available. The sheet makes that a deficit, even though the team did everything it could. Larger conferences have the opposite advantage: a team can finish with a losing Quadrant 1 record and still be seen as having tested itself.
It depends on the ranking underneath. If the NET misjudges a team, every quadrant that team touches shifts. The tiers inherit the strengths and weaknesses of the ranking, and the NET's own formula is not fully public. This is the same black-box concern raised in the discussion of the NET itself.
It is a blunt record, not a measure of margin. A team that lost three Quadrant 1 games by a possession and one that lost them by 25 have the same record in the table. The sheet shows other ratings beside it, so the committee can see the difference, but the quadrant record alone hides it.
How the committee uses it
The selection process, described in Selection Sunday: how the committee works and How the NCAA tournament field is chosen, is a human process with a set of rules. Members rank and compare teams in rounds of voting, and the information they use includes the NET, the quadrant records, the strength-of-schedule figures, other power ratings, results against common opponents and the committee's own judgment from watching games.
The NCAA has described the NET as a sorting tool, which is a careful phrase: it organizes teams so that the committee can group them and compare them, but it does not by itself decide who is in. The quadrants play the same role. They narrow a long list to a short comparison, and then people debate the cases that remain. Typical debates around the bubble go along the lines of "Team A has more Quadrant 1 wins but also a Quadrant 3 loss, and Team B has fewer top wins but a cleaner record." The table frames the question without answering it. The practical effects are covered in The bubble: who is in and who is out.
Two other facts are useful. First, the automatic bids, 32 of the 68 places in the field, are decided by conference championships, so the quadrant system only matters for the 36 at-large places and for seeding. Second, the system applies at the Division I level and its exact structure is the NCAA's to change; the cutoffs and the details can be revised, and a reader should check the current rules, not rely on memory.
Using the quadrants as a fan
You do not need the committee's access to use this information well. A few habits will keep you from reading too much into the table.
- Look at the pattern, not just the count. Ask how many opportunities a team had. Five Quadrant 1 wins out of eight tries is a better sign than five out of twenty.
- Check the losses. A single bad loss is common; a pair is a warning. The resume sits in the unusual results.
- Watch the rankings move. An opponent's rank in November is not its rank in March. If a team's best wins come against opponents that have since dropped, its profile will weaken without the team playing a game.
- Pair the table with a predictive rating. If a team's quadrant record is mediocre but its efficiency ratings are strong, it may have been unlucky in close games; if the reverse, it may have been fortunate. The contrast between result-based and predictive systems is explained in the companion article on college efficiency metrics.
- Keep it in proportion. The quadrant record is one summary of a résumé, and the committee has said so repeatedly.
What to take away
The quadrant system takes a simple idea, that a road win over a strong team means more than a home win over a weak one, and turns it into four tiers defined by the opponent's NET rank and the game's location. It gives the selection committee a common language and rewards teams for testing themselves on the road. It also creates cliffs at the cutoffs, rewards scheduling for the table and leaves smaller-conference teams short of top-tier opportunities. Read it as a map of a team's résumé, not a verdict. You can follow the rankings on the rankings page, see how the field takes shape on the March Madness hub and look at the men's and women's brackets under brackets.