Initial release of women's RPI was incorrect...used men's formula...Tech at 29

[RealityCheck]

Reality,
If WKU were to end up as 13 in RPI, do you think the committee would hold true to form and give them a 4 seed? That would be remarkable.

If WKU finishes around 13, I think a 4 or maybe a 5 seed is certainly possible. WKU has three Top 50 OOC home wins (Tech, Charlotte, and Louisville) and all of its five losses have been to Top 100 RPI teams on the road. Those Top 50 wins are RPI bonuses using the secret RPI adjustments that the men have done away with.

As you pointed out in a earlier post, the biggest thing missing from Tech's resume is a Top 50 RPI win. At least Tech doesn't have any losses other than its four to Top 100 RPI teams on the road.

[The Historian]

If I’m understanding this correctly, the men’s weighted formula was released in its entirety. Therefore, there is no “secret formula” on the men’s side. The public gets to see what the committee members are seeing.


On the women’s side, the unweighted formula is still being used by the committee, but there is still a “secret formula” the committee uses that is not released to the public.


If that’s correct, I’m assuming its some type of weighted formula, similar to the men’s. Given Palm's initial weighted index, that could help us.

[dogtrooper]

This thread is giving me a head ache.

[RealityCheck]

If I’m understanding this correctly, the men’s weighted formula was released in its entirety. Therefore, there is no “secret formula” on the men’s side. The public gets to see what the committee members are seeing.


On the women’s side, the unweighted formula is still being used by the committee, but there is still a “secret formula” the committee uses that is not released to the public.


If that’s correct, I’m assuming its some type of weighted formula, similar to the men’s. Given Palm's initial weighted index, that could help us.

The only difference between the men's and women's RPI is in the 25% team winning percentage factor. The women's RPI weights each game equally but the men's RPI weights away wins and home losses 1.4, home wins and away losses 0.6, and neutral games 1.0. Weightings aren't used in the calculation of the SOS factors that make up the other 75% of either RPI formula.

Example...8-2 team is 5-0 at home and 3-2 away.
Women's RPI uses 25% of the .800 winning percentage for the 8-2 record.
Men's RPI refigures the record as 7.2-1.2 (.8571) for that 25% of the RPI total.
The men's RPI would be .0143 higher with everything else being equal for this team than what it would be using the women's RPI.

The "secret adjustments" still used after the basic RPI is calculated for the women reward good OOC schedules (over 50% OOC against the Top 50 RPI), penalize bad OOC schedules (over 50% OOC against RPIs worse than 150), reward Top 50 RPI wins, and penalize losses to teams below 100 or 150 in RPI. How much these adjustments are worth has never been made public, and they have nothing to do with the weighted RPI now being used for the men.

[champion110]

The only difference between the men's and women's RPI is in the 25% team winning percentage factor. The women's RPI weights each game equally but the men's RPI weights away wins and home losses 1.4, home wins and away losses 0.6, and neutral games 1.0.

Example...8-2 team is 5-0 at home and 3-2 away.
Women's RPI uses 25% of the .800 winning percentage for the 8-2 record.
Men's RPI refigures the record as 7.2-1.2 (.8571) for that 25% of the RPI total.

The "secret adjustments" still used after the basic RPI is calculated for the women reward good OOC schedules (over 50% OOC against the Top 50 RPI), penalize bad OOC schedules (over 50% OOC against RPIs worse than 150), reward Top 50 RPI wins, and penalize losses to teams below 100 or 150 in RPI. How much these adjustments are worth has never been made public, and they have nothing to do with the weighted RPI now being used for the men.

It could really help us, then, if Iowa keeps winning. Of all our victories, they stand the best chance at finishing below 50 rpi. That would work in our favor. I don't think Ark. State and Mississippi can make it down to below 50 before the end of the regular season. I guess Mississippi COULD, but not likely.

[dawg80]

Wrong...big time.

Western Kentucky is projected to win its last 6 SBC regular season games with 4 of them over teams now in the Top 100 RPI. The WKU schedule has home games against North Texas (93), UALR (124), Arkansas State (71), and MTSU (72), with road games at Denver (158) and FIU (73). Sweeping those six would cause the jump from 30 RPI now to the projected 13.

The SBC is 9th in overall RPI while the WAC is 14. CUSA is 13 BTW.

Until the women go to the weighted RPI like the men have, game sites are not factored into a team's RPI record counting for 25% of overall RPI.

Welllllllllllll....excuse me! :icon_wink

(thanks for the info)

[champion110]

Dang, our rpi has dropped to 44 on collegerpi.com. I hope the NCAA one has us in a better position. I think it is due out today.

Rutgers beat UConn last night. Since Mississippi beat Rutgers, that should help us some.

[RealityCheck]

Dang, our rpi has dropped to 44 on collegerpi.com. I hope the NCAA one has us in a better position. I think it is due out today.

Rutgers beat UConn last night. Since Mississippi beat Rutgers, that should help us some.

RPI is RPI...there is only one way to calculate it if all of the results are input correctly.

Tech was at 45 through Monday's games on the NCAA list, so Tech gained one spot with Tuesday's action.