More Validity than RPI

[dawg80]

Common opponents:

Louisiana Tech and Duke have faced three common opponents- Stephen F. Austin, Tennessee and Tulsa. Duke defeated Stephen F. Austin (119-53) and Tulsa (90-36), but lost to Tennessee (72-69). Louisiana Tech knocked off Stephen F. Austin (109-60) and Tulsa (63-60 and 81-57), but lost to Tennessee (85-65).


And even that must be taken with a grain of salt, because tomorrow is a different day, with a totally different set of variables then when Tech and Duke played those other teams.

Still, that provides more insight into the "relative" strength of teams. A LOT more than that silly RPI.

And, no, I am not arguing that Duke did not deserve a higher seed than Tech. Perhaps a #1 vs #3....

[FriscoDawg]

Still, that provides more insight into the "relative" strength of teams. A LOT more than that silly RPI.

RPI does not measure strengths of teams, "relative" or otherwise. It uses the 3 sets of team records to measure performance, not strength. Strength is a subjective measure based on potential that may or may not be realized. Performance can be objectively measured based on documented history.

And, no, I am not arguing that Duke did not deserve a higher seed than Tech. Perhaps a #1 vs #3....

And if seeding was based on only on RPI, Tech would be a #6 and couldn't have played a #1 seed in the regional semis.

[TechsasDawg]

FriscoDawg, if I award you a tripple gold star for knowing more about RPI than everybody else in the world, will you please let it rest? Nobody is questioning your expertise in RPI, where it comes from, where it goes, what it is used for, or how it is computed! But most of us are really getting tired of you showing off your surpurb knowledge everytime anybody uses the letters: "rpi". Would you please just gloat in your untouchable position as the Worlds greatest RPI Gruru, and let some of us lowly posters on this board have a few conversations withot interruption from an expert?

Thanks,

TD
~~

[FriscoDawg]

FriscoDawg, if I award you a tripple gold star for knowing more about RPI than everybody else in the world, will you please let it rest? Nobody is questioning your expertise in RPI, where it comes from, where it goes, what it is used for, or how it is computed! But most of us are really getting tired of you showing off your surpurb knowledge everytime anybody uses the letters: "rpi". Would you please just gloat in your untouchable position as the Worlds greatest RPI Gruru, and let some of us lowly posters on this board have a few conversations withot interruption from an expert?

Thanks,

TD
~~

Since the brackets were announced I haven't started a single thread regarding RPI or seeding. You are preaching to the choir here, but when jabs at RPI and seeding continue to rain down from the peanut gallery (or I am specifically mentioned by name) I have every justification to respond.

[markay714]

FriscoDawg, if I award you a tripple gold star for knowing more about RPI than everybody else in the world, will you please let it rest? Nobody is questioning your expertise in RPI, where it comes from, where it goes, what it is used for, or how it is computed! But most of us are really getting tired of you showing off your surpurb knowledge everytime anybody uses the letters: "rpi". Would you please just gloat in your untouchable position as the Worlds greatest RPI Gruru, and let some of us lowly posters on this board have a few conversations withot interruption from an expert?

Thanks,

TD
~~

And, the crowd cheers loudly and says AMEN!!!!!!!!!!

[dawg80]

UCSB, a 11th seed, just pushed the #2 UCONN, to the wire in the Sweet 16.

I knew UCSB was a better team than that stupid RPI placed them in seeding. Which was why it was easy to choose them as my darkhorse in the tourney. It was a real no-brainer, actually.

RPI says UCSB is between the 41st and 44th best team in the nation. I knew they are a legit Top 20 at least, probably Top 15, as their trip to the final 16 proved.

I don't know why it's so hard for some people to understand that teams like Colorado garnering 6th seed is a complete joke, because they got there NOT on their own merit, but rather by playing other, over-rated teams.

UCSB and Colorado can play to Kingdom Come, and UCSB will smoke them everytime. Tech and Texas Tech can play a 10-game series and our Techsters will win at least 9, perhaps all 10 times! But, T Tech has a higher RPI...don't ya know.

[FriscoDawg]

RPI doesn't give any indication of the "best" team, otherwise why play a tournament--award the trophy to Tennessee every year.

UCSB was RPI #55 and 47th (not 41st-44th) among teams in the field of 64. That's a 12 seed--not an 11--using only RPI. If RPIs are going to be quoted, at least use the correct figures.

Tech played only two regular season games against teams still alive entering the Sweet 16. Texas Tech played 7 games (2 OOC) against Sweet 16 teams, and Colorado played 4 (2 OOC).

And the NCAA tournament is a one-and-done deal, not a series. It could be argued that Duke should beat Tech 9 times out of 10, but it only takes once.

[markay714]

Can you please just go ahead and repeat your message please, TechsasDawg? I want to say Amen again.

[champion110]

Ughghghghgghghhhh

I respect you, Frisco - I REALLY do. You are an expert on the rpi and a fan of it. If I ever have any questions about the rpi, you will be the first person that I contact.

However, please do not give us a lesson again about rpi not measuring strength, but who "deserves" a higher seed instead. We know this. The fact is that rpi gives a higher seed to teams that do NOT deserve higher seeds. It favors the big conferences (the same conferences that are bcs in football). That is UNFAIR. It will always be unfair. When you are seeded lower, your chances of getting knocked out earlier in the tournament will be higher. That is because you have to face the #1 seed earlier. The advantage in the tournament goes to the higher seeds and as long as we are in the WAC, we will never get a #1 seed again. That should tell you something, especially since we have one of the most consistent top programs in the country over the last 25 years.

[FriscoDawg]

You are an expert on the rpi and a fan of it.

I have never said I was a fan of or in favor of RPI. But it is a main component of the system that is and will continue to be used for the forseeable future, so I try to understand it.

Tech has no chance of a 1 seed before it ever steps on the floor against WAC teams due to its weak OOC schedules.

And C110, your statement about lower seeds meeting #1s earlier is not correct in Tech's case this year--a 6 seed would have kept Tech away from Duke longer than the 5 seed did. That's why I prefer a 6 seed to a 4 or 5 if a team isn't in the top 3.

[TechsasDawg]

OK, everybody. Let's just use ABC when we really mean "***", and maybe FriscoDawg won't notice what we're talking about. Don't anybody tell him what we're doing, now. That would blow the whole thing. As an example:

Earlier, dawg80 indicated that actual head to head scores against common opponents provides more insight into the "relative" strength of teams than does ABC. See how easy that is? Maybe we fooled him this time.

TD
~~

[champion110]

:-)

[american]

RPI doesn't give any indication of the "best" team ...

Amen.

Why does the selection committee use it then? As you mention, the best team is determined on the court, but predicting which team is better (which all concede should be based on performance) could be done much better than it is now. The rpi is a crude system to rank teams based on criteria that favors the super conferences. Many believe it is objective, but it is far from it.

It's not completely without merit, however. If one team is ranked about 75 places ahead of another in the rpi, then it's probably a better team. If one team is a few hundredths or thousandths of an rpi score ahead of another, on the other hand, then it is by no means superior. In fact, Frisco, it has not even performed better than the other team, unless you define performance by one's rpi score.

Expert human judgment should be used for seeding untill the NCAA adopts a ratings index that also takes into account margin of victory and whether you play at home or on the road. In my opinion (and I'm sure many others) the index should also weight a team's actual winning percentage a bit more heavily. Those enamored of the rpi, however, will insist that its arbitrary assignment of a .25 weight to a team's own winning percentage is the only way to measure a team's performance.

[FriscoDawg]

OK, everybody. Let's just use ABC when we really mean "***", and maybe FriscoDawg won't notice what we're talking about. Don't anybody tell him what we're doing, now. That would blow the whole thing. As an example:

Earlier, dawg80 indicated that actual head to head scores against common opponents provides more insight into the "relative" strength of teams than does ABC. See how easy that is? Maybe we fooled him this time.

TD
~~

Head-to-head scores have a definite entertainment value. But what's wrong with using just wins and losses and not limiting the comparison to selected head-to-head games--include every game of every team. Oops, I think I just described key parts of the ABC calculation. :wink:

[american]

I now throw in the towel.