Selection Sunday Discussion

[stever20]

Bonnies were CO-CHAMPIONS of their League with a 29 rpi. They beat the A-10 tournament winner TWICE. Clearly there were some shenanigans with the Tulsa invite. 0fer 59 in bracket predictions? Fix was in.

and like they said the RPI isn't the sole determinant any longer. Remember the post that someone put on here a month or so ago.....

St Bonnie's did nothing OOC. Best OOC win was Ohio at #81. Tulsa had 2 wins better than that in Wichita and Iona, and they also beat Ohio themselves.

Quick question for you. Should Temple have gone to the tourney last year?

[hoops22]

Bonnies were CO-CHAMPIONS of their League with a 29 rpi. They beat the A-10 tournament winner TWICE. Clearly there were some shenanigans with the Tulsa invite. 0fer 59 in bracket predictions? Fix was in.

and like they said the RPI isn't the sole determinant any longer. Remember the post that someone put on here a month or so ago.....

St Bonnie's did nothing OOC. Best OOC win was Ohio at #81. Tulsa had 2 wins better than that in Wichita and Iona, and they also beat Ohio themselves.

Quick question for you. Should Temple have gone to the tourney last year?

If you're going to use the RPI's of a teams opponents as justification for your argument, shouldn't the RPI of the team itself play a major role? It's a joke the Bonnies didn't get in with an RPI of 29. As far as bad loses go, La salle wasn't a good loss, but Little Rock and Oral Roberts weren't good ones for Tulsa either. I honestly feel the determining factor was Tulsa had a friend on the commitee, and St Bona didn't. And that stinks.

[stever20]

Bonnies were CO-CHAMPIONS of their League with a 29 rpi. They beat the A-10 tournament winner TWICE. Clearly there were some shenanigans with the Tulsa invite. 0fer 59 in bracket predictions? Fix was in.

and like they said the RPI isn't the sole determinant any longer. Remember the post that someone put on here a month or so ago.....

St Bonnie's did nothing OOC. Best OOC win was Ohio at #81. Tulsa had 2 wins better than that in Wichita and Iona, and they also beat Ohio themselves.

Quick question for you. Should Temple have gone to the tourney last year?

If you're going to use the RPI's of a teams opponents as justification for your argument, shouldn't the RPI of the team itself play a major role? It's a joke the Bonnies didn't get in with an RPI of 29. As far as bad loses go, La salle wasn't a good loss, but Little Rock and Oral Roberts weren't good ones for Tulsa either. I honestly feel the determining factor was Tulsa had a friend on the commitee, and St Bona didn't. And that stinks.

Little Rock isn't a good loss? I'm sorry, but that's where the labels are moronic. You are trying to say how good the mid-majors are- but UALR is one of the best ones they had. You know what their KP rating was? 47, right in between Providence and Creighton. RPI is 42, 2 spots behind Providence. So I'm sorry but to say UALR is a bad loss is a total joke.

Why they don't use the RPI of the teams as much is because the teams/conferences can totally game the RPI.

[Jet915]

Bonnies were CO-CHAMPIONS of their League with a 29 rpi. They beat the A-10 tournament winner TWICE. Clearly there were some shenanigans with the Tulsa invite. 0fer 59 in bracket predictions? Fix was in.

Committee Chair is friends with Tulsa AD, that is your answer.

[herodotus]

ok but where was SB's buy game?

Also look at the difference between the AAC and A10 in terms of who they can get to come play them at home.

At Syracuse was their buy game. Bona is indeed in a tough spot. Temple, Nova, and St. Joe's all play each other, and those are solid RPI games most years. Duquesne gets Pitt on a neutral court every year, and can get return games from WVU, and Penn St. Syracuse won't come to Bona, and all of their other regional rivalries have Bona as the big dog.

[xbr1]

I don't mean to be mean to Bonas, but I doubt they have the money for that buy game. They are probably in one of the tougher situations in D1. Very small school, not a huge population base to draw fans from, too good to get paid for going on the road, not a big enough name to draw marquee home/homes. It is a really tough spot to be in. They recently had to put seats between their student section and the court in order to raise revenue, which I believe was supposed to make something like an extra $50k a year. That's maybe enough to afford a decent buy game.

Bonaventure has the money they got from Xavier when they left all those NCAA credits moving to the BE.

[herodotus]

I don't mean to be mean to Bonas, but I doubt they have the money for that buy game. They are probably in one of the tougher situations in D1. Very small school, not a huge population base to draw fans from, too good to get paid for going on the road, not a big enough name to draw marquee home/homes. It is a really tough spot to be in. They recently had to put seats between their student section and the court in order to raise revenue, which I believe was supposed to make something like an extra $50k a year. That's maybe enough to afford a decent buy game.

Bonaventure has the money they got from Xavier when they left all those NCAA credits moving to the BE.

You're misunderstanding what we mean by "buy games" with regard to Bona. We're not talking about Bona buying games. Bona can't buy the types of games that will help them get into the tournament. Bona can't even get home and homes against those teams. We're talking about Bona being the "bought" team. In other words, Bona going to Kentucky, with no return game. The problem Bona has, is that those are guaranteed losses, and you just can't book too many of them. If you step down to second tier P5 teams, they're only looking to schedule a handful of risky games, even at home. Teams like Bona, Creighton, and Tulsa are risky to schedule. No one will be impressed if Arizona beats those teams, but everyone will be criticizing if those teams pull an upset. Thus, the top teams will only schedule sure wins at home, along with a game or two vs a blue blood that won't really hurt them if they lose.

[XUFan09]

How can the committee value top 50 rpi wins but not value RPI rankings for teams. Makes no sense

RPI is not a precise measure and therefore shouldn't be used directly to rank a team. Hoeever, it's close enough that it can be used indirectly to say how a team fared against different tiers of opponents, as that allows for margin of error.

Ranking systems like Kenpom are more precise and really, I think one of them should be used. The advanced metrics do a better job of indicating how well a team is playing, so they better represent how tough it is to beat that team.

Can someone explain to me what makes RPI an imprecise metric? And how Kenpom (which rewards blowouts) over close wins would be better?

Simple. Because margin of victory is one of the best predictors of how likely a team is to win future games. It's not simply the dichotomy of close games vs. blowouts either, as you present it. It's the whole continuum. Beating a team by 10 rather than by 1 says something about the winner (and the loser). Extending the margin to 10 means that as the clock ticks down, the potential effect of luck approaches 0. If a team can extend the margin to 20, the potential effect of luck approaches 0 with even more time left on the clock. And so on. This is something the coaches are well aware of, too. They understand the concept of luck in close games, in a sport where teams only score on about 50% of their shots, and if they can help it, they don't want to be in a late game situation where luck is still a factor.

RPI doesn't account for this. Beating a team by 1 because you made a halfcourt shot versus beating a team by 21 because you controlled the entire game count for the same, even though we know that the 21-point victor is probably better than the 1-point victor. Because those two victory margins count for the same, there's a significant margin of error with RPI if we are to treat it like as a way or ranking teams. We have instances like Kansas beating San Diego State by 13 while Fresno State beats them by 1. I'm going to go out on a limb and suggest that Kansas is better than Fresno State, and not just that, that Kansas was better against that specific opponent than Fresno State was. In addition, teams can play the RPI, selectively scheduling potential conference champions from lesser conferences to boost their computer profile (e.g. Xavier scheduling Murray State last year). If they schedule a team like that, advanced metrics make it so they still have to soundly beat them too (which Xavier did).

Back to that halfcourt heave (or really, any buzzer-beating shot attempt). As the ball goes up, does the success or failure of that specific shot tell you anything significant about the two teams playing? Through the lens of the RPI, that's an enormous difference, a win versus a loss. Kenpom and other advanced metrics, however, hardly make distinctions based on one shot attempt. It only can slightly change the margin of victory, but everything else about those two teams has been laid out over the previous 40 minutes.

Edit: In the end, the problem with the RPI is it is binary, win or loss, with big changes in the number based off that. A great example of that is the end of the Georgetown-Marquette game near the end of the season. Georgetown took the lead off a shot from DSR with a few seconds left, then Marquette came down quickly and Luke Fischer drew a foul on a missed layup with a couple seconds left in the game. He made the shots, but then Georgetown almost made a shot at the buzzer. All in all, it was a tossup game, one where the foul could have not been called (end-of-game, it usually isn't), Fischer could have missed one or two free throws, or the buzzer-beater could have gone in. If Georgetown had won, though, their RPI would have been 95 instead of 106. That's 11 spots based off minute changes in the end-of-game scenario, changes that don't really say a lot about either team as a whole.

[stever20]

I don't mean to be mean to Bonas, but I doubt they have the money for that buy game. They are probably in one of the tougher situations in D1. Very small school, not a huge population base to draw fans from, too good to get paid for going on the road, not a big enough name to draw marquee home/homes. It is a really tough spot to be in. They recently had to put seats between their student section and the court in order to raise revenue, which I believe was supposed to make something like an extra $50k a year. That's maybe enough to afford a decent buy game.

Bonaventure has the money they got from Xavier when they left all those NCAA credits moving to the BE.

You're misunderstanding what we mean by "buy games" with regard to Bona. We're not talking about Bona buying games. Bona can't buy the types of games that will help them get into the tournament. Bona can't even get home and homes against those teams. We're talking about Bona being the "bought" team. In other words, Bona going to Kentucky, with no return game. The problem Bona has, is that those are guaranteed losses, and you just can't book too many of them. If you step down to second tier P5 teams, they're only looking to schedule a handful of risky games, even at home. Teams like Bona, Creighton, and Tulsa are risky to schedule. No one will be impressed if Arizona beats those teams, but everyone will be criticizing if those teams pull an upset. Thus, the top teams will only schedule sure wins at home, along with a game or two vs a blue blood that won't really hurt them if they lose.

Creighton and Tulsa seem to be scheduling pretty damn well... I mean, I know Creighton had some cupcakes this year, but even still got Arizona St playing there. Last year they got Oklahoma and St Mary's at home. 2 years ago, they got Tulsa, Nebraska, Cal there.

Tulsa got Wichita at their place 2 years ago. last year, they got Creighton and Oklahoma at their place. This year Wichita again.

[marquette]

I don't mean to be mean to Bonas, but I doubt they have the money for that buy game. They are probably in one of the tougher situations in D1. Very small school, not a huge population base to draw fans from, too good to get paid for going on the road, not a big enough name to draw marquee home/homes. It is a really tough spot to be in. They recently had to put seats between their student section and the court in order to raise revenue, which I believe was supposed to make something like an extra $50k a year. That's maybe enough to afford a decent buy game.

Bonaventure has the money they got from Xavier when they left all those NCAA credits moving to the BE.

You're misunderstanding what we mean by "buy games" with regard to Bona. We're not talking about Bona buying games. Bona can't buy the types of games that will help them get into the tournament. Bona can't even get home and homes against those teams. We're talking about Bona being the "bought" team. In other words, Bona going to Kentucky, with no return game. The problem Bona has, is that those are guaranteed losses, and you just can't book too many of them. If you step down to second tier P5 teams, they're only looking to schedule a handful of risky games, even at home. Teams like Bona, Creighton, and Tulsa are risky to schedule. No one will be impressed if Arizona beats those teams, but everyone will be criticizing if those teams pull an upset. Thus, the top teams will only schedule sure wins at home, along with a game or two vs a blue blood that won't really hurt them if they lose.

Yup my bad. Thanks for clearing that up. Guess my brain just couldn't process the idea of a team from a multi bid league being the "bought" team