Way to go RPI!!!! ( sarcasm )

[dcZONAfan]

Nope.

RPI uses your win%, your opponent's win% and their opponent's win%.

So beating Duke (19-5) is significantly better than beating LA Tech (19-5)

(and the Duke win would also be better than the Toledo win)

ahhh, gotcha. Sometimes you make too much sense trolly

 

[mr.hockey4242]

I definitely understand the point of RPI and it has a good use but it really does overrate some of these smaller teams.

UMass and George Washington would be great A10 examples. Hell, probably St. Louis too.

 

[TrollyMcTroller]

A road win by 1 over Toledo would be a "better" win than a 25 point home drubbing of Duke yes... I get WHY they have an RPI, I just dont understand why its the PRIMARY metric. If we are really looking for the 68 best teams then that makes no sense and if we arent looking for the 68 best teams and we are only looking at resumes then we should just take the 34 highest eligible teams in the RPI for the at larges after the auto bids are handed out. Its when the #62 RPI team gets in ahead of the number 35 RPI team that I get confused. In some instances it seems to matter a lot and in others not at all. Very strange.

False.

A road win by 1 over Toledo would be weighted more than a home win by 25 over Duke. The Duke win would still be a better win.

And for the most part, the top available RPI teams do get the at large bids. Yes there are excpetions, but wouldn't you want there to be exceptions? Isn't that what the committee is for? They use the RPI as tool. They also use common sense and the eyeball test. They look at things like player injuries and other variables that are harder to quantify. Yes, the BPI does take player injuries into account but it is unaware of the circumstances. The BPI doesn't know if a player is out for a game, or out for the season. But the committee does. So they factor that stuff in.

 

[element1286]

Nope.

RPI uses your win%, your opponent's win% and their opponent's win%.

So beating Duke (19-5) is significantly better than beating LA Tech (19-5)

(and the Duke win would also be better than the Toledo win)

It's broken down 50/25/25 right? I still don't understand how this is better than anything involving MOV?

There are still holes in the measure than can be corrected, I don't understand why one wouldn't want to correct them and use something better. There is nothing RPI can tell you that a MOV measure can't tell you.

 

[TrollyMcTroller]

Depends on where the win was at Trolly. But you know that

Nope.

The location determines the weighting of the win, not the quality. Beating #8 will always be better than beating #31, regardless if it's in Raleigh, Toledo, or on Neptune.

 

[rmilia1]

I definitely understand the point of RPI and it has a good use but it really does overrate some of these smaller teams.

UMass and George Washington would be great A10 examples. Hell, probably St. Louis too.

Mid major schools get a big bump in RPI because they tend to play more road/neutral court games OOC than major schools. Those wins count at a much higher rate than home wins so teams that play more of them ( even if its against just average competition ) get a bigger bump especially when they are all lumped together in 1 conference as they are in the A10 this year where you see 5 A10 teams in the top 44 in the RPI. There are 15 mid majors in the top 50 of the RPI which is 3-5 more mids than in any other metric. Road/neutral value is the reason

 

[rmilia1]

Nope.

The location determines the weighting of the win, not the quality. Beating #8 will always be better than beating #31, regardless if it's in Raleigh, Toledo, or on Neptune.

How can that be right??? Not disagreeing I guess just makes no sense mathematically. If a team is given "extra" valuation for beating a team on the road it is going to automatically increase the quality of the win correct?? So I dont see how a win over say NDSU at #46 on the road wouldnt be better than beating #25 Colorado at home... Guess Im not sure of the initial values awarded to each win based off of their rank but its have to be a HUGE spread to compensate for the extra .4 weight given to the win.

 

[TrollyMcTroller]

It's broken down 50/25/25 right? I still don't understand how this is better than anything involving MOV?

There are still holes in the measure than can be corrected, I don't understand why one wouldn't want to correct them and use something better. There is nothing RPI can tell you that a MOV measure can't tell you.

It's 25/50/25 (W%/OP W%/ OP OP W%)

You're trying to turn the RPI into something it's not. If you want to measure relative team strength, there are already dozens of rating systems that do that. That's not what the RPI is for.

The RPI measures resumes. A resume is Who you played, where you played, and who won. Thats it.

Your resume is where you worked, what position and how long you worked there. That's it. It isn't how many TPS reports you filed while you were working for Inatech. It isn't the number of inter-office memos you sent while you were at Intertrode. It's where you worked, what position, and how long. If that stuff is good enough they'll bring you in for an interview and you can tell them all about your TPS reports.

If your teams resume is good enough, then the committee will take a closer look and you can tell them all about how you kicked the crap out of Duke by 25. But that isn't what goes on a resume.

 

[TrollyMcTroller]

How can that be right??? Not disagreeing I guess just makes no sense mathematically. If a team is given "extra" valuation for beating a team on the road it is going to automatically increase the quality of the win correct?? So I dont see how a win over say NDSU at #46 on the road wouldnt be better than beating #25 Colorado at home... Guess Im not sure of the initial values awarded to each win based off of their rank but its have to be a HUGE spread to compensate for the extra .4 weight given to the win.

Okay, I see what you're saying.

I'm looking at the individual win, and you're looking at the affect that win has on your overall RPI. And in that case, yes, the NDSU win would improve your RPI more than Colorado win. That said, we agree that they put too much weighting on the home/away aspect.

But if you factor it in at all, then you're going to have anomalies like that. They might not be as pronounced but it is still a possibility.

 

[rmilia1]

It's 25/50/25 (W%/OP W%/ OP OP W%)

You're trying to turn the RPI into something it's not. If you want to measure relative team strength, there are already dozens of rating systems that do that. That's not what the RPI is for.

The RPI measures resumes. A resume is Who you played, where you played, and who won. Thats it.

Your resume is where you worked, what position and how long you worked there. That's it. It isn't how many TPS reports you filed while you were working for Inatech. It isn't the number of inter-[/B]office memos you sent while you were at Intertrode. It's where you worked, what position, and how long. If that stuff is good enough they'll bring you in for an interview and you can tell them all about your TPS reports.

If your teams resume is good enough, then the committee will take a closer look and you can tell them all about how you kicked the crap out of Duke by 25. But that isn't what goes on a resume.

So youre saying 2 chicks at the same time???

 

[rmilia1]

Okay, I see what you're saying.

I'm looking at the individual win, and you're looking at the affect that win has on your overall RPI. And in that case, yes, the NDSU win would improve your RPI more than Colorado win. That said, we agree that they put too much weighting on the home/away aspect.

But if you factor it in at all, then you're going to have anomalies like that. They might not be as pronounced but it is still a possibility.

Word. OK. Makes sense, we were just talking about 2 different things. Its like Im married again... LOL

 

[dcZONAfan]

It's 25/50/25 (W%/OP W%/ OP OP W%)

You're trying to turn the RPI into something it's not. If you want to measure relative team strength, there are already dozens of rating systems that do that. That's not what the RPI is for.

The RPI measures resumes. A resume is Who you played, where you played, and who won. Thats it.

Your resume is where you worked, what position and how long you worked there. That's it. It isn't how many TPS reports you filed while you were working for Inatech. It isn't the number of inter-office memos you sent while you were at Intertrode. It's where you worked, what position, and how long. If that stuff is good enough they'll bring you in for an interview and you can tell them all about your TPS reports.

If your teams resume is good enough, then the committee will take a closer look and you can tell them all about how you kicked the crap out of Duke by 25. But that isn't what goes on a resume.

It's a JUMP-to-conclusions Mat

 

[TrollyMcTroller]

So youre saying 2 chicks at the same time???

"Damn straight. I always wanted to do that, man. And I think if I were a millionaire I could hook that up, too, 'cause chicks dig dudes with money."

"Well, not all chicks."

"Well, the type of chicks that'd double up on a dude like me do."

 

[dcZONAfan]

"Damn straight. I always wanted to do that, man. And I think if I were a millionaire I could hook that up, too, 'cause chicks dig dudes with money."

"Well, not all chicks."

"Well, the type of chicks that'd double up on a dude like me do."

I watched it a couple weeks ago. Came on at like 1 am, I said I would just watch a little bit. Nope, stayed up for the whole thing. So fucking good

 

[TrollyMcTroller]

And for the record I think the RPI does a good job at it's intended purpose, but if I was going to change it...


I'd keep the formula and criteria the same. I'd change the weighting of the home/away to 0.8/1.2 and I'd change the weighting of the W%/OP W%/ OP OP W% to 35%/45%/20%

If you wanted to get really fancy for resume measuring, you could use the same basic concept of the RPI, but instead of 75% being opponent's w% and their opponents w%, you could replace that with the just the opponent's kenpom or sagarin rating. that way you'd have a more accurate measure of how good your opponents are (which is relevant to resume) plus in a roundabout sort of way it would include MoV for you guys that insist on shoehorning it on somewhere.

 

[Destroydacre]

I was told a few years back that when the committee is discussing seeding and at large bids (say Syracuse is 3rd in the RPI) that they don't actually see that ranking. What they do see is that Syracuse was 5-2 vs RPI top 25, 10-3 vs RPI top 50, 7-0 vs RPI 200+ etc etc. Does anyone know if that's actually the case?

 

[rmilia1]

And for the record I think the RPI does a good job at it's intended purpose, but if I was going to change it...


I'd keep the formula and criteria the same. I'd change the weighting of the home/away to 0.8/1.2 and I'd change the weighting of the W%/OP W%/ OP OP W% to 35%/45%/20%

If you wanted to get really fancy for resume measuring, you could use the same basic concept of the RPI, but instead of 75% being opponent's w% and their opponents w%, you could replace that with the just the opponent's kenpom or sagarin rating. that way you'd have a more accurate measure of how good your opponents are (which is relevant to resume) plus in a roundabout sort of way it would include MoV for you guys that insist on shoehorning it on somewhere.

Damn man, we just became best friends. I love that idea.

 

[TrollyMcTroller]

I was told a few years back that when the committee is discussing seeding and at large bids (say Syracuse is 3rd in the RPI) that they don't actually see that ranking. What they do see is that Syracuse was 5-2 vs RPI top 25, 10-3 vs RPI top 50, 7-0 vs RPI 200+ etc etc. Does anyone know if that's actually the case?

My understand is that they have boatloads of information at their disposal. I've sen a lot of speculation as to which stuff they pay more attention too, but I think it actually depends on where a given team is in the rpi.

If I'm on the committee, and I'm seeding the top teams, I'm probably looking at the "vs. top-n" more so than the overall rating/ranking. It's already been established that (I'll go with your example) Syracuse is a good team. Then the question is how did they do against other upper echelon teams.

If I'm looking at bubble teams, I"m more concerned with getting an overall picture of what they did. But that's just me.

For all I know, they've got some interns in there filling the brackets out and the committee members are snorting blow of hookers' asses on the conference table.

 

[TrollyMcTroller]

Damn man, we just became best friends. I love that idea.

Tell ya what. I'll work on that tonight, and get back to you tomorrow.

One thing I can tell you for sure... next season I'm starting from day-1 and putting all the season results in a database every night, so I don't have to go digging through a hundred web pages to get the data I want... write a query, execute, boom. Answer!

 

[MI Nightmare]

Soon to be 6 loss Kansas can do no wrong in the RPI....