forums

Canes #11/13 (Coach/AP) - 2.20.23

AmherstCane said:
It’s not actually our losses that are tanking our NET ranking. Of course they’re not helping, but it’s the wins against mediocre teams where we gave up a lot of points.

Neither net nor kenpom allows a ranking to be overly influenced by one or two games. Kenpom does weight early season games for offensive efficiency metrics and punish poor defensive performance early in the season. But our bad metrics are because we have not destroyed the mid teams we played… we beat them but we let them score. Thus a bad defensive efficiency ranking that is killing our NET and kenpom.

For example, our defensive efficiency ranking dropped like 10 spots after the 2nd Louisville game… and our kenpom ranking dropped by 5-6 slots.

I’m not defending the idiocy of using weighted margin of victory (which is ultimately what net efficiency metrics measure) over actual wins and losses… but that’s basically what they’re doing.
Click to expand...


And...again...I DON'T GIVE A ****.

I know exactly what the ****** metrics are doing. And it is wrong.

There is only one metric that matters. Only one. WINNING MOTHER****ING GAMES.

Coach L has 718 wins. Nobody gives a **** if his career point differential is 1,000 points or 10,000 points.

So, again, I'll make this simple, because you are telling repeating a lie (not YOUR lie, someone else's lie) by claiming that "one or two games" are not messing things up. Because if we HAD beaten Maryland and GaTech, and STILL had the 3-point loss to Pitt, and the 2-point losses to Duke and NC State, we'd be a Top 4 team with a #1 seed right now. With NO OTHER "metrics" changes against teams like "Louisville".

It's a ******* joke.

OOC margins of victory:

13
14
26
10 (Top 25 ranked team)
23
2
7
2
15
Game Cancelled

So 6 of our 9 OOC wins were double-digit wins. One of the three single-digit wins is against a tourney-team (as is our one OOC loss). And the cancelled game was against an opponent that is likely to make the tourney. And one of the 2-point victories is against a team that MIGHT win the Ivy and get the automatic bid. And the other 2-point win was against a team that was on the bubble until recently.

ACC margins of victory:

27
7 (NC State, probable tournament team)
2 (UVa, Top 10 ranked team)
11
16
4 (Syracuse, bubble team)
23
9
4 (Clemson, near certain tournament team)
22
8 (Loserville hits a 3 point shot just before the buzzer)
8 (North Carolina, bubble team)
9 (Wake Forest, bubble team)

5 of 13 games won by double-digits. Another 5 that become double-digit wins if one additional shot falls (or if the other team doesn't take a worthless shot with seconds left like Louisville did). Our only "bad" win is a 4-point win (since the 2-point win over UVa is quality, as is the 4-point win over Clemson).

And we're supposed to feel bad about "only" beating Louisville by 8 when we beat them by 27 in the first game?

Or we're supposed to feel bad that Louisville took a 3 with 6 seconds left on the clock to cut an 11 point Miami lead down to "only" an 8 point lead?

Come on. That's why Kenpom and NET and all the other metrics suck.

Win games. That's all that matters.
 
Last edited:
Advertisement
TheOriginalCane said:
And...again...I DON'T GIVE A ****.

I know exactly what the ****** metrics are doing. And it is wrong.

There is only one metric that matters. Only one. WINNING MOTHER****ING GAMES.

Coach L has 718 wins. Nobody gives a **** if his career point differential is 1,000 points or 10,000 points.

So, again, I'll make this simple, because you are telling repeating a lie (not YOUR lie, someone else's lie) by claiming that "one or two games" are not messing things up. Because if we HAD beaten Maryland and GaTech, and STILL had the 3-point loss to Pitt, and the 2-point losses to Duke and NC State, we'd be a Top 4 team with a #1 seed right now. With NO OTHER "metrics" changes against teams like "Louisville".

It's a ******* joke.

OOC margins of victory:

13
14
26
10 (Top 25 ranked team)
23
2
7
2
15
Game Cancelled

So 6 of our 9 OOC wins were double-digit wins. One of the three single-digit wins is against a tourney-team (as is our one OOC loss). And the cancelled game was against an opponent that is likely to make the tourney. And one of the 2-point victories is against a team that MIGHT win the Ivy and get the automatic bid. And the other 2-point win was against a team that was on the bubble until recently.

ACC margins of victory:

27
7 (NC State, probable tournament team)
2 (UVa, Top 10 ranked team)
11
16
4 (Syracuse, bubble team)
23
9
4 (Clemson, near certain tournament team)
22
8 (Loserville hits a 3 point shot just before the buzzer)
8 (North Carolina, bubble team)
9 (Wake Forest, bubble team)

5 of 13 games won by double-digits. Another 5 that become double-digit wins if one additional shot falls (or if the other team doesn't take a worthless shot with seconds left like Louisville did). Our only "bad" win is a 4-point win (since the 2-point win over UVa is quality, as is the 4-point win over Clemson).

And we're supposed to feel bad about "only" beating Louisville by 8 when we beat them by 27 in the first game?

Or we're supposed to feel bad that Louisville took a 3 with 6 seconds left on the clock to cut an 11 point Miami lead down to "only" an 8 point lead?

Come on. That's why Kenpom and NET and all the other metrics suck.

Win games. That's all that matters.
Click to expand...

Preach.

People were so enamored by these new fancy metrics they fell over themselves to adopt them without really understanding the deficiencies.

Really shame on the NCAA for throwing out the RPI for this garbage.
 
AmherstCane said:
Preach.

People were so enamored by these new fancy metrics they fell over themselves to adopt them without really understanding the deficiencies.

Really shame on the NCAA for throwing out the RPI for this garbage.
Click to expand...


And keep in mind, I'm not mad at you, as you are only citing what is actually being used. I just hate this concept of "but this is what the stats tell us", instead of using wins and common sense to rank teams.

I'm just not sure why the "eye-test" was so deficient. It just takes some reasonable analysis to figure out what is going on.
 
TheOriginalCane said:
And...again...I DON'T GIVE A ****.

I know exactly what the ****** metrics are doing. And it is wrong.

There is only one metric that matters. Only one. WINNING MOTHER****ING GAMES.

Coach L has 718 wins. Nobody gives a **** if his career point differential is 1,000 points or 10,000 points.

So, again, I'll make this simple, because you are telling repeating a lie (not YOUR lie, someone else's lie) by claiming that "one or two games" are not messing things up. Because if we HAD beaten Maryland and GaTech, and STILL had the 3-point loss to Pitt, and the 2-point losses to Duke and NC State, we'd be a Top 4 team with a #1 seed right now. With NO OTHER "metrics" changes against teams like "Louisville".

It's a ******* joke.

OOC margins of victory:

13
14
26
10 (Top 25 ranked team)
23
2
7
2
15
Game Cancelled

So 6 of our 9 OOC wins were double-digit wins. One of the three single-digit wins is against a tourney-team (as is our one OOC loss). And the cancelled game was against an opponent that is likely to make the tourney. And one of the 2-point victories is against a team that MIGHT win the Ivy and get the automatic bid. And the other 2-point win was against a team that was on the bubble until recently.

ACC margins of victory:

27
7 (NC State, probable tournament team)
2 (UVa, Top 10 ranked team)
11
16
4 (Syracuse, bubble team)
23
9
4 (Clemson, near certain tournament team)
22
8 (Loserville hits a 3 point shot just before the buzzer)
8 (North Carolina, bubble team)
9 (Wake Forest, bubble team)

5 of 13 games won by double-digits. Another 5 that become double-digit wins if one additional shot falls (or if the other team doesn't take a worthless shot with seconds left like Louisville did). Our only "bad" win is a 4-point win (since the 2-point win over UVa is quality, as is the 4-point win over Clemson).

And we're supposed to feel bad about "only" beating Louisville by 8 when we beat them by 27 in the first game?

Or we're supposed to feel bad that Louisville took a 3 with 6 seconds left on the clock to cut an 11 point Miami lead down to "only" an 8 point lead?

Come on. That's why Kenpom and NET and all the other metrics suck.

Win games. That's all that matters.
Click to expand...
One of the things that could also be hurting us is team like, I’m going to use TN as an example, have 3 wins over top 5 teams. We on the other hand have played 1 top 10 team all season. A lot of these other top 10 teams are playing each other and the ACC just doesn’t have that this year.
 
TheOriginalCane said:
And keep in mind, I'm not mad at you, as you are only citing what is actually being used. I just hate this concept of "but this is what the stats tell us", instead of using wins and common sense to rank teams.

I'm just not sure why the "eye-test" was so deficient. It just takes some reasonable analysis to figure out what is going on.
Click to expand...

Yeah the problem with the efficiency metric rankings is that there’s no appreciable difference between winning and losing if you play a top ranked team tight. It’s why Creighton was (and still is) ranked so highly despite losing so many games and why West Virginia is ranked NINE spots ahead of us in Kenpom despite sporting a 16-12 record.

Meanwhile Gonzaga is ranked so highly (13th in Kenpom) despite having a similar win/loss profile as us (similar record, similar strength of schedule, similar win distribution) because they have more 20+ point wins which boosts their offensive and defensive efficiency ratings.
 
Advertisement
Miami04 said:
One of the things that could also be hurting us is team like, I’m going to use TN as an example, have 3 wins over top 5 teams. We on the other hand have played 1 top 10 team all season. A lot of these other top 10 teams are playing each other and the ACC just doesn’t have that this year.
Click to expand...


I understand. But, again, this can become the human centipede of scheduling. All those Big 12 teams just keep getting "quality wins" by beating each other.

Look, we are in one of the best conferences in the country, year-in, year-out. When you have to construct your team to beat those kinds of teams, you shouldn't be penalized for a year in which a couple of ACC teams underperform (Duke, UNC) while giving way too much to a conference that is a bit better than normal.

Again, it you had teams ranked at #8 and #9, and you needed a mental tiebreaker, sure, look at some stats to help out. But you shouldn't be so tied to stats that you use them to overwhelm what is most important, which is winning games.

I've already given EIGHT teams which either have higher AP ranks and/or higher NCAA seedings than Miami. Maybe a couple of those would be acceptable. BUT EIGHT? When Miami just wins games?

Come on, now. I'm sick and tired of "this conference is up" or "this conference is down". That will play itself out in tournament berths and money, but it should not trump the winning of games.
 
Does anyone remember how many points a game miami from 2012-2013 (Larkin's team) and 2015-2016 teams had? These were our best teams in the last decade prior to this year. This team scores 79/pts a game. Pretty impressive I think.

If there is ever a chance to sneak in the final four as an underdog, this is the year to do it. There is no clear cut no 1 team.
 
CANEFAN83 said:
Does anyone remember how many points a game miami from 2012-2013 (Larkin's team) and 2015-2016 teams had? These were our best teams in the last decade prior to this year. This team scores 79/pts a game. Pretty impressive I think.

If there is ever a chance to sneak in the final four as an underdog, this is the year to do it. There is no clear cut no 1 team.
Click to expand...


2015-16 - 75.2 PPG
2012-13 - 69.7 PPG

I've said this before, 80 is our magic number. We've only lost one game where we scored 80 or more (the game at NC State).
 
Advertisement
CBurrr93 said:
Can moan about kenpom all we want. It’s been a pretty reliable metric for who wins the national championship over the years.
Click to expand...


Good. Then we can skip the NCAA Tournament. No need to have it. We just predict the winner using Kenpom, and everyone can save on travel costs.

Good lord, the crap that gets used as justification for Kenpom...

Nobody has yet posited a world where the entire NCAA Tournament would be RUINED by horrendous "mis-seeding" if we just took the 68 teams and ranked them strictly by record.

And that's the point. Once we get past the issue of "last four in" and "first four out" (and none of those teams ever actually WIN the Tournament), then WHO ******* CARES about the seeding? And I'm not talking about matchups. Yes, a "fading in March" team can get paired with a "hot in March" team, and upsets happen. I'm talking strictly seeding. If we ranked BY RECORD, who would be more than one or two seeding spots off of "Kenpom's seeding" (except Miami of course).

But instead, we license some of the most ridiculous "stats" and "metrics" to override what we all just watched for four months. Stupid, stupid, stupid.
 
Advertisement
1676998963486.png
 
TheOriginalCane said:
Good. Then we can skip the NCAA Tournament. No need to have it. We just predict the winner using Kenpom, and everyone can save on travel costs.

Good lord, the crap that gets used as justification for Kenpom...

Nobody has yet posited a world where the entire NCAA Tournament would be RUINED by horrendous "mis-seeding" if we just took the 68 teams and ranked them strictly by record.

And that's the point. Once we get past the issue of "last four in" and "first four out" (and none of those teams ever actually WIN the Tournament), then WHO ******* CARES about the seeding? And I'm not talking about matchups. Yes, a "fading in March" team can get paired with a "hot in March" team, and upsets happen. I'm talking strictly seeding. If we ranked BY RECORD, who would be more than one or two seeding spots off of "Kenpom's seeding" (except Miami of course).

But instead, we license some of the most ridiculous "stats" and "metrics" to override what we all just watched for four months. Stupid, stupid, stupid.
Click to expand...
They’re just metrics, man lol they’re not biased. The best overall team rarely wins it all. What makes it fun. I think we deserve a 4 at worst. I’m on your side. It just is what it is. Can’t do much about it other than beat who we play
 
Advertisement
CBurrr93 said:
Can moan about kenpom all we want. It’s been a pretty reliable metric for who wins the national championship over the years.
Click to expand...
National champ always has a off and def efficiency rating that is combined, under 50. Gotta be top 30 in each. Been that way for every winner almost the last 2 decades I believe.
 
CBurrr93 said:
They’re just metrics, man lol they’re not biased. The best overall team rarely wins it all. What makes it fun. I think we deserve a 4 at worst. I’m on your side. It just is what it is. Can’t do much about it other than beat who we play
Click to expand...


Look, the only way that PRIOR changes happened, whether it was with the BCS or the RPI or any other system, is by complaining about the current system and pointing out the flaws.

I did not say the metrics are biased, but they are certainly skewed. And if Miami doesn't intentionally gun for double-digit victories, then it is going to have a more significant IMPACT on Miami, whether there is intentional "bias" or not.

Seriously, I posted all of the victory margins for the entire season. It's not bad at all. It's certainly better than last year, when the "luck" factor was constantly being cited as the reason for Miami being underrated (before we ran to the Elite 8). 2 of our 3 most narrow conference margins of victories are against the top teams in the ACC (UVa and Clemson).

Look, we have ONE conference victory by "one possession", and it was to the #6 team in the country. We have THREE conference losses "by one possession". So if anything, "luck" has left the building, and all but a couple of non-con wins are by a legit margin (if not "double-digit" or "20 points").

Finally, just look back at two of Miami's most accomplished teams ever, and we are now SCORING MORE POINTS than ever. So what's the problem? We don't keep trying to score more late in games that we are going to win? I mean, just think about this, I remember way more games LAST YEAR that were "in doubt" than this year.

I just think it is wrong to evaluate everything on a metrics/statistics basis that says there is only one "way" to win a game, by a lopsided margin that is never in doubt. And I don't think it is right to allow teams with more losses to be "forgiven" because the losses were somehow more acceptable because of "offensive efficiency" or some other joke stat.

And that has nothing to do with "predictive" issues. Yes, teams that score more and hold teams to less have a better chance of winning the national championship. I didn't need a metric to tell me that.

It's time to point out that these metrics are FLAWED. That they favor only one style of play. That they incentivize running up the score. That they overwhelm the eye test and common sense.

MIAMI IS NOT THE ONLY ONE COMPLAINING. The UCLA coach had some choice words. If there are numerous head-scratching results, AND THERE ARE, then it's time to revisit the metrics.

Kenpom is in its infancy. NET is in its infancy. Either improve them or replace them.
 
Basketball, especially, is one sport where margin of victory can easily be skewed and shouldn't be a determining factor. You're up 20, you put in the 5'1" [keep the skin color out of it] walkons and the other team hits some 3s in the last two minutes and you win by 9. You're punished. Dumb.
 
Last edited by a moderator:
Advertisement
Back
Top