Change in RPI formula

[TrueHoyaBlue]

Interesting announcement from the NCAA that the RPI formula has changed this year (quote is from Jerry Palm's CollegeRPI.com).

----------------------------------

RPI Change
Dec 18
The NCAA announced yesterday that the basic RPI formula has been changed. This is different than the announcement a couple of months ago that the secret adjustments had been changed.

They announced that the new formula will still use the 25-50-25 ratio as it has since 1994, but that all road wins are treated as 1.4 wins, all road losses are treated as .6 losses, all home wins are worth .6 wins and all home losses are valued at 1.4 losses. Neutral site games are still worth 1.

It also said that the secret adjustments no longer exist.

I will be bugging the NCAA for some clarification next week, but in the meantime, I'm going to assume it is as simple as it sounds.


---------------------------------------


Interestingly, this change has dropped G'town's RPI from the 30-50 range, to 102 at the moment.

[CAHoya07]

Strange, very strange. Based on what it did to our RPI, I like the old system better.

But I assume this should even itself out further into the season?

[TrueHoyaBlue]

Yeah, and I know that we'll definitely have more road games in next year's non-conference, just in looking at our commitments to Illinois and SJSU.

[Cambridge]

Wait a second...so they are killing the "bonus" for playing top 20 teams? WTF

[bmartin]

What happened to Georgetown's RPI:
Traditional record: 4 wins, 2 losses, .6667
New RPI record: 3.2 wins, 2.8 losses, .5333
3 home wins @ 0.6 = 1.8
1 road win @ 1.4 = 1.4
2 homes losses @ 1.4 = 2.8

The road win bonus/home loss penalty is too big.
This punishes home losses severely.

[SFHoya99]

The RPI is stupid to begin with, but giving a 40% bonus to going on the road? Are you kidding me?

Just use Sagarin or Pomeroy -- this is why we shouldn't have a selection committee made up of people who can't do math. A little statistics is a dangerous thing.

[HoyaChris]

It's not a 40% bonus for going on the road, it's a 133% bonus. RPI is really worthless anyway, but this has unbelievable implications. Apparently the NCAA thinks that we should be playing at Howard and Duke should be playing at Elon - all the time.

[SFHoya99]

Thanks for the correction.

I think this means that we can make the NCAA tournament merely by scheduling all road games.

[JimmyHoya]

BRILLIANT!

We begin playing all the local teams which would provide a tough OOC schedule, plus a semi-home advantage, despite us being "away".

Then, when we absolutely must play at home for Big East stuff, we jam people into McDonough and show UConn, ND, 'Cuse, etc. the craziness and win those games too.

To hell with the MCI Center!

33-0 and an astronomical 16.83458 RPI rating here we come~! ;D

Sure, the alums might be a little Editedy, but when we're #1, they'll be more than willing to give money and hten we'll have that on campus arena we covet so much.

[GUHoya07]

haha, you wish. Too bad we're not allowed to play Big East games at McDonough because it doesn't satisfy the conference minimum capacity requirements.

[hoyadrummer]

I don't think the selection commitee could have hurt Georgetown worse if they had tried. As we have no real "home court" our continued losses in unfriendly MCI are going to KILL us if this continues to be the RPI formula in future years. This is REAL problem.

[Jack]

I guess this new formula does some damage to BC's "No. 1" RPI ranking as well, given the fact they hardly leave Conte in the non-conference. It does seem like a very unfair alteration, the sort of thing Dick Vitale might come up with to force the big guys to give his beloved "mid-majors" more home games.

[Cambridge]

I can understand giving a bonus for an away win...but to penalize so severely for a home loss....

This will only destroy major conferences. Stupid move. I'm sorry but a loss to a top twenty UConn at home should not be much worse RPI wise than a win at LaSalle...stupid, stupid, ridiculous move.

Without the bonus for top 20 teams, there is no correction for hosting a good team. In the past, I'm pretty sure the "corrections" were a .2 swing for home-away PLUS a .2 bonus for playing a top 20 RPI team.

This meant a loss at home to a top 20 would equal 1 loss rather than 1.2 losses. And the win would equal 1 win rather than .8 of a win.

[hoyainspirit]

Pomeroy has a nice blog on the subject. kenpom.typepad.com/ha/2004/12/the_new_and_imp.html

[hoya4ever]

if we were to have a bonus for playing ranked teams does that mean that we would be punished for playing bad teams? or does it only count when we play good teams?

[TrueHoyaBlue]

Random tidbit -- as of 12/29 (post Norfolk State, pre Howard) the Hoyas' RPI rank stands at 94, whereas under the old ranking system, it would have been 60.

Guess we do get hurt pretty badly by the new system.

[Fan Of The Game]

I don't mind the change.

While our scheduling seems to be on the rise, I'm all for a system that encourages us to play more tough games on the road than MEAC cellar dwellers at home.

[FOTP]

I like the change a lot. Great for college basketball and the Hoyas.

Our schedule is definitely getting better. However, let's be careful here. The Illinois game was not a game that we went after. They called us at the last minute because someone else bailed on them.

[hoyarooter]

The old regime would probably have turned down the Illinois offer in order to schedule Elon.

[FOTP]

Probably true...probably true.

JTIII gets a pass on the schedule for obvious reasons.