Predictive rankings don't like us as much as the record would indicate

[Millhouse]

It is because we aren't very good. I think our record will even itself out. We have no really good wins. The BC road loss is a bad loss even though the metrics say otherwise.

We will finish somewhere in the range of 17-12.

Generally I do think schedule adjusted scoring margin does predict well and it will be interesting to see if our beat downs are an outlier or if this are predictive too

 

[Millhouse]

Clemson beat TCU, Bama, and South Carolina. Alabama beat South Carolina, Auburn, and Mississippi ST.

All 6 of those wins are better than anything we have done. We also sht the bed against teams we should have blown out. They did not.

Beating Pitt and Oregon isn’t enough.

If teams with similar schedules have the same record but win and lose to different caliber teams, why are more good wins considered better than fewer bad losses?

 

[Cusefan0307]

If teams with similar schedules have the same record but win and lose to different caliber teams, why are more good wins considered better than fewer bad losses?

Bama has lost to 4 top 10 teams, Clemson and Ohio State. All were relatively close games. Our losses are worse.

Clemson has lost to 2 top 15 teams in close games, another team that's been in and out of the top 25 in Memphis in a close game, and 3 other close ACC losses including 2 on the road. I think there's an argument to be made that Clemson has similar losses as us, but they also have much better wins.

 

[tbonezone]

it would appear a win is a win no matter if it's by 2 in OT or 22 in regulation. OTOH a loss by 2 versus a loss by 22 carries a whole different grade of significance. getting blown out is never a good thing on the resume.

 

[Millhouse]

Bama has lost to 4 top 10 teams, Clemson and Ohio State. All were relatively close games. Our losses are worse.

Clemson has lost to 2 top 15 teams in close games, another team that's been in and out of the top 25 in Memphis in a close game, and 3 other close ACC losses including 2 on the road. I think there's an argument to be made that Clemson has similar losses as us, but they also have much better wins.

If 2 hypothetical teams have the same SOS and the same record, and one team has better wins, they also must have worse losses. Do you agree?

 

[Cusefan0307]

If 2 hypothetical teams have the same SOS and the same record, and one team has better wins, they also must have worse losses. Do you agree?

Mostly, but what if one team plays all top 10 teams and loses while another plays teams ranked 10-30 and goes .500. Theoretically the SOS averages could be the same, but the 2nd team had less variation in their schedule.

 

[Millhouse]

Mostly, but what if one team plays all top 10 teams and loses while another plays teams ranked 10-30 and goes .500. Theoretically the SOS averages could be the same, but the 2nd team had less variation in their schedule.

How could those teams have the same SOS?

 

[Cusefan0307]

We actually went up in the Net one spot last night. That loss was expected.

 

[orangenauburn]

I watch big 12 all the time since I live in Iowa. The big 12 is not that overrated. They have some really good teams.

 

[orangenauburn]

 

[Cstate]

As has been noted here correctly a bunch of times….margin of victory/defeat absolutely plays a key part in the NET, as well taking full advantage of doing well against your high ranking opponents. South Carolina moves up 10 spots after beating #5 Tennessee. On the flip side, Clemson drops 6 spots for only beating a bad Louisville (Quad 4) team by 5 points. They were leading the game by 21 at one point, so important to keep the leads big!

 

[FireballPhil]

I would agree their out of conference schedules are pretty weak. But have you watched them play? Especially defensively? It is a tough league. How many years did we have a weak non-con yet still turned out to be pretty good? You can't use a November schedule as your go all metric. It does game the rankings system for sure, a bunch of good teams blowing out much lesser teams. But they are doing what they should be-blowing them out. There may indeed be a couple teams that don't deserve to be ranked as high as they are, but just watch a few games.

 

[jncuse]

If 2 hypothetical teams have the same SOS and the same record, and one team has better wins, they also must have worse losses. Do you agree?

Not really in all examples - its why the SOS was so easily abused in the RPI era, or more often than that just accidentally really helped or screwed teams in the RPI era.

Here is an easy example.

Team A plays 3 games OOC = #4 in the country, #190 in the country, #349 in the country.
Team B plays 3 games OOC = #103 in the country, #120 in the country, #150 in the country.
What's the harder schedule. Under SOS its Team B. But is that really the harder schedule?

If your a bubble level team you are going to win 95%+ of your games against sub 100 teams at home. Here was a more common example in the RPI age - that could be planned or just accidental. You playing teams 125, 150, 175 at home OOC instead of 260, 300, 340 its going to make a massive difference in your SOS overall at the end of the year. But why should it matter when bubble level or tourney level teams are going to go 3-0 against each slate either way.

Using the hypothetical numbers above (and trust me those scenarios were extremely common in the RPI age (ask the MVC) you can really game or just get accidentally screwed by the SOS). Basically if you can game (or just get lucky or unlucky) the bottom 2/3rd's of the teams in the country during your OOC, you could play 2 really hard games instead of 6... and still end up with the same SOS as the team that played many harder teams.

Whether its RPI, KP, or SRS, the same concept exists.

 

[Millhouse]

Not really in all examples - its why the SOS was so easily abused in the RPI era, or more often than that just accidentally really helped or screwed teams in the RPI era.

Here is an easy example.

Team A plays 3 games OOC = #4 in the country, #190 in the country, #349 in the country.
Team B plays 3 games OOC = #103 in the country, #120 in the country, #150 in the country.
What's the harder schedule. Under SOS its Team B. But is that really the harder schedule?

If your a bubble level team you are going to win 95%+ of your games against sub 100 teams at home. Here was a more common example in the RPI age - that could be planned or just accidental. You playing teams 125, 150, 175 at home OOC instead of 260, 300, 340 its going to make a massive difference in your SOS overall at the end of the year. But why should it matter when bubble level or tourney level teams are going to go 3-0 against each slate either way.

Using the hypothetical numbers above (and trust me those scenarios were extremely common in the RPI age (ask the MVC) you can really game or just get accidentally screwed by the SOS). Basically if you can game (or just get lucky or unlucky) the bottom 2/3rd's of the teams in the country during your OOC, you could play 2 really hard games instead of 6... and still end up with the same SOS as the team that played many harder teams.

Whether its RPI, KP, or SRS, the same concept exists.

Your example rankings.
They have the same record 2-1
A loses to #4
B must have better wins than A, all the games are better than 150 where A's 2 wins are vs 190+
But that means B's loss is worse than A's because they don't play anyone as good as who A lost to

Zero sum