2016-17 RPI Rankings

[ArmyVet]

It seems to me that Butler is way overvalued in the RPI with two really bad losses and a third loss that wasn't even a contest.

[ConnersvilleBulldog]

It seems to me that Butler is way overvalued in the RPI with two really bad losses and a third loss that wasn't even a contest.

Fair. But wins against Nova, Arizona (neutral court), Cincinnati, Xavier, IU (neutral court), Utah (away), & Northwestern more than make up for 2 really bad road losses (by a total of 4 pts).

[stever20]

It seems to me that Butler is way overvalued in the RPI with two really bad losses and a third loss that wasn't even a contest.

that's what happens when you have 13 of your games vs top 100 teams. #9 SOS.

They might be slightly overvalued in RPI but they are 16 in KP. It'll be interesting to see what seed they get. If they finish strong, they'll get a good one.

[sciencejay]

It seems to me that Butler is way overvalued in the RPI with two really bad losses and a third loss that wasn't even a contest.

Fair. But wins against Nova, Arizona (neutral court), Cincinnati, Xavier, IU (neutral court), Utah (away), & Northwestern more than make up for 2 really bad road losses (by a total of 4 pts).

I think these posts bring into sharp relief the confusion and main complaints over selections to the Dance and seeding. Butler has great wins and a great record, yet they also have a couple of questionable/bad losses. ArmyVet chose to focus on the losses, and ConnersvilleBulldog focused on the great wins. There will likely be different committee members with those same positions when it comes time to determine seeding. I'm hoping the committee member focusing on Butler's great wins is more convincing than the other one so that the Dawgs get as good a seed/location as possible.

[DudeAnon]

People can cherry pick all they want. Butler has the 3rd best RPI in the country for a reason. I don't see anyone bitching about Kansas or Kentucky being in the top 10, but Butler is "overvalued" somehow? If RPI were the only metric, Butler would be a 1 seed right now. They are having an unreal season and any argument to the contrary is bullsh--.

[kmacker69]

16 wins against teams in top 100 in the RPI, 6 top 50, 4 top 20 and 2 top 10. That's what overcomes the two very close road losses to +100 teams. RPI isn't a perfect metric by any means, but it does favor beating lots of really good teams and not losing to horrible ones. (200+ and Ind St is getting close to dropping us a few spots just by losing to other horrible teams.) Our schedule is built to be overvalued by RPI even with a few more losses.

[Fieldhouse Flyer]

RPI Forecast – Teams # 1 to # 351 . . . RPI Forecast - Big East Conference

Big East Conference Projections - through games of Saturday Jan. 14, 2017.

Overall - Expected RPI Rank – Team – Projected Regular-season W-L (Projected Conference W-L)

4.0 - Villanova - 27-4 (15-3)
10.0 - Butler - 23-7 (12-6)
15.5 - Creighton - 25-5 (13-5)
20.9 - Xavier - 21-10 (11-7)

58.3 - Marquette - 19-11 (10-8)
59.5 - Seton Hall - 19-11 (9-9)
74.4 - Georgetown - 16-15 (7-11)
96.2 - Providence - 16-15 (6-12)
155.0 - St. John's - 12-19 (6-12)
223.5 - DePaul - 10-21 (3-15)

I’m posting the projections below as a volatility sensitivity study for one-day changes after last night’s three Big East games:
--------------------------------------------------------------------------------------------------------------------------------------------------
Big East Conference Projections - through games of Tuesday Jan. 24, 2017:

Expected RPI Rank – Team – Projected Regular-season W-L (Projected Conference W-L)

4.2 - Villanova - 27-4 (14-4)
7.6 - Butler - 23-7 (12-6)
16.6 - Creighton - 24-6 (12-6)
20.7 - Xavier - 21-10 (11-7)
44.9 - Marquette - 20-10 (11-7)

59.0 - Seton Hall - 18-12 (8-10)
76.9 - Providence - 17-14 (7-11)
88.3 - Georgetown - 15-16 (6-12)
160.7 - St. John's - 12-19 (6-12)
223.9 - DePaul - 9-22 (2-16)
--------------------------------------------------------------------------------------------------------------------------------------------------
Big East Conference Results - games on Wednesday Jan. 25, 2017:

St John's 91, Providence 86
Georgetown 71, Creighton 51
Butler 61, Seton Hall 54
--------------------------------------------------------------------------------------------------------------------------------------------------
Big East Conference Projections - through games of Wednesday Jan. 25, 2017:

Expected RPI Rank – Team – Projected Regular-season W-L (Projected Conference W-L)

4.3 - Villanova - 27-4 (14-4)
6.2 - Butler - 24-6 (13-5)
19.5 - Creighton - 24-6 (12-6)
19.8 - Xavier - 21-10 (11-7)
46.7 - Marquette - 20-10 (11-7)

66.2 - Seton Hall - 17-13 (7-11)
77.0 - Georgetown - 16-15 (7-11)
91.3 - Providence - 16-15 (6-12)
145.9 - St. John's - 12-19 (6-12)
224.0 - DePaul - 9-22 (2-16)

Seton Hall is projected to fall on the wrong side of the bubble, despite appearing in 76 of 82 Brackets as of last night.
--------------------------------------------------------------------------------------------------------------------------------------------------
Bracket Matrix – last updated Wednesday Jan. 25, 2017 - 10:03 PM

Seed No. – Overall Ranking – Team (No. of Brackets out of 82)

# 1 - 1 – Villanova (82)
# 3 – 10 - Butler (82)
# 4 – 14 - Creighton (82)
# 6 – 23 - Xavier (82)
# 9 - 36 - Marquette (78)
# 10 – 40 - Seton Hall (76)

[Fieldhouse Flyer]

UDPride RPI Rankings – Monday morning January 23rd

2 – Villanova
4 - Butler
12 - Creighton
15 - Xavier
36 - Seton Hall
50 - Providence

53 - Marquette
93 – Georgetown

154 - St. John's
212 – DePaul

UDPride RPI Rankings – Friday morning January 27th

2 – Villanova
3- Butler
14 - Creighton
16 – Xavier
41 - Marquette
46 - Seton Hall

74 - Providence
81 – Georgetown

143 - St. John's
214 – DePaul

Another good thing about the RPI is that it does not penalize teams for losing a road game to a high-quality opponent.

[Fieldhouse Flyer]

----------------------------------------------------------------------------------------------------------
Big East Conference Projections - through games of Wednesday Jan. 25, 2017:

Expected RPI Rank – Team – Projected Regular-season W-L (Projected Conference W-L)

19.8 - Xavier - 21-10 (11-7)
----------------------------------------------------------------------------------------------------------
Big East Conference Projections - through games of Thursday Jan. 26, 2017:

Expected RPI Rank – Team – Projected Regular-season W-L (Projected Conference W-L)

20.9 - Xavier - 21-10 (11-7)
----------------------------------------------------------------------------------------------------------

RPI Forecast projected a Xavier loss at Cincinnati and consequently, it had a minimal effect on Xavier’s projected numbers.

On Selection Day, There Are Metrics, and Then There Are Advanced Metrics – Marc Tracy, New York Times - March 11, 2016

“The common metrics most of us use are KenPom, Sagarin, L.R.M.C., B.P.I., and K.P.I.,” the committee chairman, Oklahoma Athletic Director Joe Castiglione, told reporters.

For the purpose of formulating my own opinions concerning the relative quality of basketball teams, I personally favour: RPI Forecast, Bracket Matrix, and the current RPI Rankings (in that order) over the AP Poll, Coaches Poll, and the advanced metrics listed above.

RPI Forecast is my favorite, because it takes into consideration the statistically probable outcomes of future games, which is an important factor when pondering what a team’s likely W-L Record, RPI Ranking, and NCAA Tournament seeding will be on Selection Sunday. In late January, all of February, and early March, RPI Forecast is a valuable evaluation tool.

Bracket Matrix is also a valuable evaluation tool, because of its large sample size (82 Brackets) to which the principles of Normal Distribution apply.

RPI Rankings are the official metric published by the NCAA, and used by the NCAA Tournament Selection Committee (to an unspecified extent).

The AP and Coaches Polls are fixated on W-L records. If you win, you stay the same or move up a few spots. If you lose, you move down a few spots. On Monday, Xavier will likely drop out of the Top 25 in both the AP and Coaches Polls, while remaining in the Top 25 in every other metric. If an AP # 25 team loses an away game to the AP # 1 team, the # 25 team will drop out of the Top 25 in the polls, despite losing a game it clearly should have lost. Does that make it a ‘worse’ team? No – it doesn’t.

I have no time for the so-called ‘advanced metrics’, because I do not know what criteria are being evaluated, how those criteria are being evaluated, or what relative weights are being assigned to each criteria used in the formulae. I cannot agree with, nor disagree with an unknown methodology.

[Bill Marsh]

----------------------------------------------------------------------------------------------------------
Big East Conference Projections - through games of Wednesday Jan. 25, 2017:

Expected RPI Rank – Team – Projected Regular-season W-L (Projected Conference W-L)

19.8 - Xavier - 21-10 (11-7)
----------------------------------------------------------------------------------------------------------
Big East Conference Projections - through games of Thursday Jan. 26, 2017:

Expected RPI Rank – Team – Projected Regular-season W-L (Projected Conference W-L)

20.9 - Xavier - 21-10 (11-7)
----------------------------------------------------------------------------------------------------------

RPI Forecast projected a Xavier loss at Cincinnati and consequently, it had a minimal effect on Xavier’s projected numbers.

On Selection Day, There Are Metrics, and Then There Are Advanced Metrics – Marc Tracy, New York Times - March 11, 2016

“The common metrics most of us use are KenPom, Sagarin, L.R.M.C., B.P.I., and K.P.I.,” the committee chairman, Oklahoma Athletic Director Joe Castiglione, told reporters.

For the purpose of formulating my own opinions concerning the relative quality of basketball teams, I personally favour: RPI Forecast, Bracket Matrix, and the current RPI Rankings (in that order) over the AP Poll, Coaches Poll, and the advanced metrics listed above.

RPI Forecast is my favorite, because it takes into consideration the statistically probable outcomes of future games, which is an important factor when pondering what a team’s likely W-L Record, RPI Ranking, and NCAA Tournament seeding will be on Selection Sunday. In late January, all of February, and early March, RPI Forecast is a valuable evaluation tool.

Bracket Matrix is also a valuable evaluation tool, because of its large sample size (82 Brackets) to which the principles of Normal Distribution apply.

RPI Rankings are the official metric published by the NCAA, and used by the NCAA Tournament Selection Committee (to an unspecified extent).

The AP and Coaches Polls are fixated on W-L records. If you win, you stay the same or move up a few spots. If you lose, you move down a few spots. On Monday, Xavier will likely drop out of the Top 25 in both the AP and Coaches Polls, while remaining in the Top 25 in every other metric. If an AP # 25 team loses an away game to the AP # 1 team, the # 25 team will drop out of the Top 25 in the polls, despite losing a game it clearly should have lost. Does that make it a ‘worse’ team? No – it doesn’t.

I have no time for the so-called ‘advanced metrics’, because I do not know what criteria are being evaluated, how those criteria are being evaluated, or what relative weights are being assigned to each criteria used in the formulae. I cannot agree with, nor disagree with an unknown methodology.

You're a wise man.

BTW, when you use RPI Forecast, you're actually using advanced metrics because he uses Sagarin to handicap the remaining games.