Gaming the NET

[1SE]

Apologies if this has been discussed somewhere in a different thread I haven't seen, but here's Van Pelt talking about Brad Brownell's comments that Houston is gaming the NET.

https://www.espn.com/video/clip?id=39615644

He doesn't talk too much though about if this is a conference-wide conspiracy. I haven't looks through all their OOC schedules, but ISU really is a great case in point - not a single Q1 OOC game, and only 2-2 in their Q2 games.

And yet, in bracketmatrix, they're only two spots behind us on the S-curve.

Much ado about nothing or gaming the system?

[FairWeatherEagle]

The only gaming I see that's mattered schedule wise is the past efforts of teams to bolster their "record" with a set of nonconf cupcakes (like Marquette in some past years). So a 21-10 record looks sexy on the outside but diabetes inside.

The NET purpose is to expose that with a relatively objective Q rating. And in my opinion it succeeds. I'm not saying it can't be gamed a bit ...but eventually you have to win games away, neutral and at home that are tougher.

So if Houston wants to figure a nuanced way to game that, go for it. But they're easy and tough wins and losses are there to see

[FairWeatherEagle]

What are we doing in the middle of the night posting on scoop 😜

[1SE]

What are we doing in the middle of the night posting on scoop 😜

Well - for those of us in Europe it's morning - not sure about anyone stateside!

[1SE]

The only gaming I see that's mattered schedule wise is the past efforts of teams to bolster their "record" with a set of nonconf cupcakes (like Marquette in some past years). So a 21-10 record looks sexy on the outside but diabetes inside.

The NET purpose is to expose that with a relatively objective Q rating. And in my opinion it succeeds. I'm not saying it can't be gamed a bit ...but eventually you have to win games away, neutral and at home that are tougher.

So if Houston wants to figure a nuanced way to game that, go for it. But they're easy and tough wins and losses are there to see

I guess Van Pelt's point was that the NET may be rewarding cupcake blowouts (in terms of O and D efficiency) more than it penalizes playing cupcakes. I don't know if any of those who have looked under the hood of NET can comment, but I think it kind of makes sense - there's certainly heteroskedasticity in the predicted win margins - does the NET account for that? Or - to simplify - does winning a game you were predicted to win by 20 but actually you win by 30 do more for your NET (on balance) than a game you were predicted to win by 2 but actually won by 3?

[MUDPT]

It's a made up story and not true.

https://twitter.com/SethBurn/status/1763278624654872878

"There's been a fair amount of discussion about the Big 12 potentially gaming the NET. If they're trying, they're not succeeding. The MVC and MAAC are the conferences whose NET ratings most exceed their predictive ratings. Conference USA is getting punished. I have no clue why..."

[wisblue]

Seems to me that that gaming might work to give a team an inflated NET but the team would be downgraded by the Selection Committee for its weak nonconference SOS.

It's not as if teams are seeded by the NCAA based only on the NET.

[lawdog77]

Seems to me that that gaming might work to give a team an inflated NET but the team would be downgraded by the Selection Committee for its weak nonconference SOS.

It's not as if teams are seeded by the NCAA based only on the NET.

I guess we'll see if the selection committee is consistent about Non con SOS. Especially these teams:
ISU-308
BYU-295
Duke-177
UConn - 220

[bilsu]

I said before that I do not think wins or losses in November should matter as much as they do.
No team is the same as it was in November.
Not that I like Seton Hall, but if they get left out of NCAA tournament with a winning record in Big East, because they played an easier non-conference schedule is just not right.

The Big East could very well end up with only three bids. Of course, part of the problem is basically the top 9 Big East teams are guaranteed 4 conference wins, because DePaul and Georgetown are so bad.

[Juan Anderson's Mixtape]

I said before that I do not think wins or losses in November should matter as much as they do.
No team is the same as it was in November.
Not that I like Seton Hall, but if they get left out of NCAA tournament with a winning record in Big East, because they played an easier non-conference schedule is just not right.

The Big East could very well end up with only three bids. Of course, part of the problem is basically the top 9 Big East teams are guaranteed 4 conference wins, because DePaul and Georgetown are so bad.

I do not share your opinion about devaluing November and December games.  Conference games already make up 2/3 of a team's resumé.  They don't need to be given extra value on top of that.

Seton Hall had chances in non-conference and lost to Baylor, Iowa, Rutgers, and USC.  Best non-conference win was vs a very bad Missouri. The Pirates should still make the tournament.  But if Seton Hall misses the tournament, they have only themselves to blame for losing to some very mediocre teams.

[bilsu]

 I do not think teams with losing conference records should get in, because they won games in November.

[CountryRoads]

I do not think teams with losing conference records should get in, because they won games in November.

I disagree. New Mexico has played zero games against a P6 opponent this year. Seton Hall has beaten a 1 seed and a 2 seed. Personally, I think it's ridiculous that they are in lock step on bubble right now as it lacks common sense.

[Newsdreams]

I disagree. New Mexico has played zero games against a P6 opponent this year. Seton Hall has beaten a 1 seed and a 2 seed. Personally, I think it's ridiculous that they are in lock step on bubble right now as it lacks common sense.

It depends on the traffic jam