• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

Consolidated Bracketology Thread 3/12/23 updates

SkyDivingDeacon said:
This. I know there’s a simplified way of looking at NET that doesn’t make much sense on the surface and we need something in place to rank teams quantitatively, but in no way should a losing team be rewarded in a ranking system.
Click to expand...
So if the worst team in the country plays a road game at the best team and loses 104-103 in triple OT, they should just stay at the bottom?
 
One team Michigan State passed was Miami who lost at home to the 215th best team in the country. So I'd say in comparison, a six point road loss to a top 50 team looks better!
 
Wakeforest22890 said:
So if the worst team in the country plays a road game at the best team and loses 104-103 in triple OT, they should just stay at the bottom?
Click to expand...
That’s an extreme example. So is this - one team wins every game they play against inferior competition, but wins by less than they should and one team loses every game they play against superior competition but loses by less than they should. NET rewards the losing team for losing by less than they are supposed to and penalizes the winning team for winning by less than they should have. Make that make sense.
 
Has this been discussed…playing @ Ohio State is a Quad 1 opportunity for Illinois today when OSU has lost 14 of their last 15 games. Again, make it make sense.
 
DeacLaw05 said:
Wake Forest, at -3.5, beats Notre Dame by 8. We fall from 80 to 84 in the NET, and Irish jump up from 193 to 190. This system is pure trash.
Click to expand...
Where were we -3.5? Vegas line was -6.5 at the tip and Wake in the metrics (so not including either team's injuries) had Wake as an 11 point favorite. Notre Dame, since they've lost a lot of games - even though they've been close - is even worse in the NET (183rd now) than metrics (roughly in the 140s). So Wake would've needed to win by even more than 11 in all likelihood based on where the team are in the NET to break even (without considering teams moving around us)
 
Wakeforest22890 said:
One team Michigan State passed was Miami who lost at home to the 215th best team in the country. So I'd say in comparison, a six point road loss to a top 50 team looks better!
Click to expand...
The ACC is getting killed by 200ish and 300ish teams that have much more talent than their record. Does anybody think Maryland Eastern Shore goes into Miami and wins last night?
 
SkyDivingDeacon said:
That’s an extreme example. So is this - one team wins every game they play against inferior competition, but wins by less than they should and one team loses every game they play against superior competition but loses by less than they should. NET rewards the losing team for losing by less than they are supposed to and penalizes the winning team for winning by less than they should have. Make that make sense.
Click to expand...
How does that not make sense to you? Simplifying it a little, but if you're "supposed" to win a game by X that is based on your ranking in the system. So if you're "supposed" to win by 10 as the 25th ranked team and win by 2 then you're going to drop a little because you didn't play like the 25th ranked team "should have" played at that point.
 
Last edited:
ChrisL68 said:
The ACC is getting killed by 200ish and 300ish teams that have much more talent than their record. Does anybody think Maryland Eastern Shore goes into Miami and wins last night?
Click to expand...
I mean we're at sample size almost 30 now. These ACC teams have had ample opportunity to prove they aren't where their rating are and they've largely failed to do so. I mean I think after this week the bottom five teams in the ACC (not counting VT since their metrics are much better) have a total of 6 wins against the top 10 in the ACC.
 
Argue all you want about the application of NET, but I genuinely do not understand people’s confusion about how it works.

The system ranks teams and attempts to place a numerical value on how good you are. When you play a game, it predicts the score based on its ranking of both teams (maybe a different score prediction than Vegas!). If you outperform the prediction, your numerical value assessing how good you are will improve, and if you don’t it will go down.

After each day, teams get re-ranked. Some teams might perform poorly, but teams ranked around them did worse. Some teams might win and perform well, but other teams around them did better. So movement within the rankings is not as simple as outperforming the projection = moving up.
 
Wakeforest22890 said:
How does that not make sense to you? Simplifying it a little, but if you're "supposed" to win a game by X that is based on your ranking in the system. So if you're "supposed" to win by 10 and as the 25th ranked team and win by 2 then you're going to drop a little because you didn't play like the 25th ranked team "should have" played at that point.
Click to expand...
It’s not so much that it doesn’t make sense…I mean I get it in theory, but there’s way too much of a ranking bias regarding the margin of victory and/or defeat in NET. It should be a data point, but NET swings way too wide the common sense mark for me to take it seriously.

We’re really at a point in athletics where we reward the losing team for losing by less than they should have and are penalizing the team that actually wins the game only because they won by less than they should have? Huh? That means you can have, all things being equal, a team with a losing record in a conference be ranked higher in NET than a team with a winning record all because of the point spread with their respective wins/losses? That’s ridiculous and I assure you, that anyone who thinks differently is in the vast minority.
 
Wakeforest22890 said:
I mean we're at sample size almost 30 now. These ACC teams have had ample opportunity to prove they aren't where their rating are and they've largely failed to do so. I mean I think after this week the bottom five teams in the ACC (not counting VT since their metrics are much better) have a total of 6 wins against the top 10 in the ACC.
Click to expand...
UMES and Florida State neutral court pick em you have to bet $100 who are you placing the bet on? How about FSU -3?

Same for Louisville and Charleston Southern.

The distribution of the wins in the ACC seems pretty standard for most leagues.
 
SkyDivingDeacon said:
It’s not so much that it doesn’t make sense…I mean I get it in theory, but there’s way too much of a ranking bias regarding the margin of victory and/or defeat in NET. It should be a data point, but NET swings way too wide the common sense mark for me to take it seriously.

We’re really at a point in athletics where we reward the losing team for losing by less than they should have and are penalizing the team that actually wins the game only because they won by less than they should have? Huh? That means you can have, all things being equal, a team with a losing record in a conference be ranked higher in NET than a team with a winning record all because of the point spread with their respective wins/losses? That’s ridiculous and I assure you, that anyone who thinks differently is in the vast minority.
Click to expand...
I mean teams play vastly different schedules. Alcorn State is 12-3 in the SWAC and West Virginia is 5-11 in the Big 12. Nobody thinks Alcorn State should be ranked higher just because they have a better conference record.
 
ChrisL68 said:
UMES and Florida State neutral court pick em you have to beat $100 who are you placing the bet on? How about FSU -3?

Same for Louisville and Charleston Southern.
Click to expand...
I mean not taking these specific teams into account because the spreads would be different than a pick em or anything (but also taking the spirit of your question at face value) I'll say that I'd probably bet on the more talented team personally, but that at the same time if you think there's value to be made overall in these types of matchups go ahead and find some and fire off in Vegas. I doubt we've discovered some big secret here on the message boards.

If you could take Louisville or Bellarmine straight up today in Louisville who would you take? Or Louisville-App State, or Louisville-Wright State, or Louisville-Lipscomb?
 
Also, everyone is going to bitch about a system that seems to unfairly rank teams and I know we need something to help put wins/losses into perspective. If NET used a weighted average of a variety of data points in ranking teams, I’d be good with it, but there’s way too much emphasis on ‘playing like you’re supposed to’. Again, if Illinois beats OSU today, it’s a Quad 1 win. OSU has lost 14 of their last 15 games and is 13th in the Big 10. The system may not be completely ‘broken’ but it sure as hell isn’t a good system to have in place when it’s used to determine which teams play in the tourney and which teams stay home.
 
Wakeforest22890 said:
I mean teams play vastly different schedules. Alcorn State is 12-3 in the SWAC and West Virginia is 5-11 in the Big 12. Nobody thinks Alcorn State should be ranked higher just because they have a better conference record.
Click to expand...
Of course not. But, to use your example, should a team in the same conference be ranked higher than another team if they have more losses but simply lost by less than they were supposed to in those losses and the team with a winning record won by less than they were supposed to?
 
The NET uses two things to rank teams:

1. Adjusted efficiency (how many points per possession you scored in a game adjusted for the strength of your opponent) and
2. Something called Team Value Index which gives you additional points for winning games based on opponent and location

I don't really know where this fits in the "there's too much playing like you're supposed to" involved in the ratings but just figured I'd repost that.

Most of the metric out there just focus on number 1, so at least in the NET you get a boost for actually winning games
 
SkyDivingDeacon said:
Of course not. But, to use your example, should a team in the same conference be ranked higher than another team if they have more losses but simply lost by less than they were supposed to in those losses and the team with a winning record won by less than they were supposed to?
Click to expand...
I mean it all just depends on what we're ranking the teams for. Are we saying who has the best chance to win today if these two teams played on a neutral court or are we ranking for a tournament selection? Since the NET is the latter, I think there should be tweaks made than JUST straight KP or Torvik or Sagarin, but there are. Also, the committee uses the NET as a guideline, not as a bright line rule. That's why Rutgers in the 80s in NET made it last year
 
Wakeforest22890 said:
Where were we -3.5? Vegas line was -6.5 at the tip and Wake in the metrics (so not including either team's injuries) had Wake as an 11 point favorite. Notre Dame, since they've lost a lot of games - even though they've been close - is even worse in the NET (183rd now) than metrics (roughly in the 140s). So Wake would've needed to win by even more than 11 in all likelihood based on where the team are in the NET to break even (without considering teams moving around us)
Click to expand...

I’ll simplify my argument. Any system that punishes the team that won the actual game while rewarding the team that lost the actual game is fucking retarded.
 
You must log in or register to reply here.
Back
Top