It makes no sense to anyone. Basically it's taken as a step-stone in calculating quads, how many quad 1/2 wins and how many 3/4 loss are supposed to be what count at the end of the day.

Early NET rankings are wonky. Some schools are gaming the system by beating quad 3 and 4 schools by 40+.

It will (mostly) work its self out.

The thing efficiency metrics don't account for, and this is why I still like there being human voters on the committee I just wish there were a lot more of them and they didn't gather together in a room, is the value of actually winning close games against good teams.

This actually applies to college football too but I think at times people are too hung up on the data and don't put enough emphasis on actually winning games.

Data is a good backdrop to put results in context but the goal of the game isn't to play the most efficient basketball, it's to win games.

Net rankings are efficiency rankings. It is a measure of how many points per possession you score vs how many points per possession you allow. This then only has an adjustment factor for big wins, but you could play only the worst 300+ teams in the country and if you beat them by 70 in every game you will have a high rank.

I believe a major component of the NET is a team's points per possession perhaps minus allowed points per possession. I can't find those stats for any team, much less for A&M. Does anyone know a site where that data is?

I guess you could get most of it from the box scores. Assuming that missed shots when the shooter is fouled counts as a shot, you could add the team's total shots from the floor, subtract the number of offensive rebounds and add their turnovers to get a close estimate of the possessions. What is missing are the non-shooting fouls that result in foul shots.

I didn't try to calculate the number for all the Aggie games, but I did look at several of them to see if the number of possessions was anywhere in the box score. What I discovered indicates why teams that play a lot of Quad 3 and 4 games are ranked so highly.

When we beat Lamar 97-71, we scored 97 points on 58 shots with 14 turnovers and 21 offensive rebounds. Ignoring any possible non-shooting fouls, our points per possession = 97 / (58-21+14) = 1.902. Lamar's points per possession = 71 / (61-14+10) = 1.183. The difference is 0.719 which I understand is very good. To me it means that we are outscoring them almost three-quarters of a point on mutual each possession. That constitutes a blowout.

However, when you play good teams, like Purdue, everything is much closer. A&M was 70/(50-14+18) = 1.296, while Purdue was 66/(53-8+16) = 1.167. The difference was .130, which is obviously a much closer game. But this is what happens when you play better teams.

One final one is our loss to UCF 61 - 64. Our PPP = 61 / (63 - 16 + 12) = 1.034, while UCF = 64 / (50 - 8 + 11) = 1.167. The difference was = -.133. The difference indicates our close loss, 3 points, while the positive .130 in the Purdue game was a close 4 point win.

Unless you weight the level of your opponent, teams which constantly blow out inferior opponents, like UGA, are going to look a lot better in the PPP calculation than teams which play a better opponents. I don't know everything that goes into the NET calculation, but I at least understand why some teams playing weak schedules are ranked more highly than I believe they deserve.

I looked it up and last night:

Texas Tech 101 - Lamar 57
Arizona 94 - Central Michigan 41

As mentioned in above posts......blow outs matter.

Can't just go the only stat that matters. But if it were 109% transparent it would be a better metric.

And they should matter.

There's enough games in college hoops that your offensive data isn't going to be skewed greatly by an anomaly game against a powder puff.

Tech's blown a lot of teams out and their 2 losses were close.

JJxvi said:

---

Net rankings are efficiency rankings. It is a measure of how many points per possession you score vs how many points per possession you allow. This then only has an adjustment factor for big wins, but you could play only the worst 300+ teams in the country and if you beat them by 70 in every game you will have a high rank.

---

Yeah, it's like the stupid game control metric the committee uses in football

When you play bad teams and dominate your game control goes up

We won by 32, so it's not like we didn't blow out the team we played

Unless the quality of the opponent is considered, then blowouts do matter in the net. At this time Arizona is ranked #26 and we are #27 in the net. Arizona is 6-5 being 0-5 against quad1 teams. They haven't played any quad 2 teams currently, but they are 6-0 against quad 3 and 4 teams. Via the eyeball test, their six wins were at least by 30 points against teams without a pulse. They were competitive against most of the Q1 opponents, but three of them were at home. Based on their actual record, it's hard to say that they are a top 50 team. Yet the net says they are better than us.

We are 5-2 against Q1 and Q2 teams, and only two of our big wins were at home. Less than half our schedule is against Q3 and Q4 teams We haven't beaten anyone by 40 points, but all the Q3 and Q4 woms except perhaps the Southern game were very convincing.

The only way UA can be ranked ahead of us is my overweighting their Q4 margins of victory. The same is true of #17 Texas Tech. They are 0-2 versus Q1 and Q2 teams and 9-0 against teams barely alive. But their margins of victory are often by 30 even 50 points.

But one of their losses was to us, and they are still ranked ahead of us.

There needs to be more transparency on the calculations and likely a major adjustment.

It's roughly the same in KenPom and Torvik too though and their formulas are more transparent so it's not some unique flaw in the NET calculations. Every game matters. Bigger games matter more. And it's an efficiency metric, not a prize for winning games. That's why none of these are end all be all rankings, you have to use some common sense too.

That's why I always preach that people are too hung up on the specifics. Especially in the middle of the season.