MVC RPI (March 17, 2008)

[The Mad Hatter]

what happened to evansville?
the past two or so years they were pretty decent.
did they just lose alot of talent?
or do they just all of a sudden suck for no normal reason?

I believe their coach moved on to Notre Dame.

[notquiteBWing]

what happened to evansville?
the past two or so years they were pretty decent.
did they just lose alot of talent?
or do they just all of a sudden suck for no normal reason?

I believe their coach moved on to Notre Dame.

ahhh.
I see.
now it makes sense.

[Shocker85]

These games hurt us even if we win all three games, right?

My recollection of RPI come from old discussions on rec.sport.college.baseball. That was a long time ago and I can't find direct reference to those discussions right now. Here is another detailed discussion of RPI: http://sebaseball.rivals.com/content.asp?CID=310988

The first component in the RPI formula is team winning percentage. Even if you play low RPI teams, sweeping them ALWAYS helps that component.

The next component is opponent winning percentage which accounts for half the RPI. Games vs. the opponent are omitted and rest of the schedule is averaged. In reality, games vs these teams don't help, but the fact that they're low RPI in itself doesn't matter - their winning percentage does.

The final component is opponents opponents winning percentage, which averages in games the low RPI teams lost, so in a way they help. Each game becomes a miniscule fraction of the total, however.

I don't believe games against bad teams kill the RPI, just so long as you win them. Those wins count directly into the winning percentage component. Their low RPI is most likely due to their own bad winning percentage, which gets averaged into the total, thus becoming a small fraction of the total.

WSU's RPI went up a lot this week, while the teams they played have RPI lower than WSU's number. This leads me to believe that the wins are by far the most important factor. In a way, NOT playing these teams which are guaranteed wins would hurt more than not playing them at all.

Makes sense?

--'85.

[SB Shock]

Makes sense?

--'85.

Thats a good link on explaining some of the issue surround the RPI (and how it can be manipulated).

The key to the RPI is this:

Schedule the best teams that you know you can beat

That is what Missouri State did in 2006. They were 8-1 in non-conference, but there best wins were RPI 126 N. Illinois and No. 52 Wisc-Milw (Bracket Buster).

Their non-confernce wins were against 126, 229, 159, 201, 130, 178, 52. This all equaled a 21 RPI and 15-0 record vs 100+ competition. But they were at risk with having 5-8 record vs Top 100 competition. When the politics got involved, the selection committee had their reason for excluding them regardless of their RPI.

Back to WSU - WSU only has 6 games versus Top 50 competition left on their non-conference schedule. They need to do well in these games. Right now they are 4-3 vs Top 50.

There are 9 games vs Conference teams that in the Mid-50 to Mid-60's that will be key for the Shockers to do well.

The rest of the schedule WSU just need to win (losses against 200+ RPI teams never will look good on the resume).

I don't believe games against bad teams kill the RPI, just so long as you win them.

Last year WSU on 4/24 was ranked 13th in ISR. The played N. Colorado (222) and UNI (174) and won 5 of 6. After the week was over they dropped 9 places down to 22nd.

[Shocker85]

Last year WSU on 4/24 was ranked 13th in ISR. The played N. Colorado (222) and UNI (174) and won 5 of 6. After the week was over they dropped 9 places down to 22nd.

Yes, but while the ISR is a better predictor of relative team strength, it doesn't necessarily move in the same way as the RPI, does it? Does the NCAA look at ISR, or is RPI their tool of choice?

Also, if you compare the two ISR from last year and look at the ISR rating scores, there a component of other teams moving UP as well which caused some of the displacement. WSU's ISR rating was at the bottom of a pack of teams very close in rating value (around 116-117). Even at the midpoint in the season, there can be enough deviation week-to-week in either direction to cause these rankings movements.

One note of concern, the RPI last year did not deviate much from the mid-20's from late March to 5/29 (the start of post-season play), which coincides with the conference schedule. WSU was 7-8 in non-conference games during that period. To improve the RPI while playing the poor RPI MVC teams, there's little doubt that taking care of business in conference and winning most non-conference games would help.

The good thing (or bad thing), is that the Selection Committee can use any criteria they desire for seeding. Based purely on ISR or RPI or any stetistical tool available, there's no way WSU should have hosted a Regional, much less a Super Regional in 2007. That said, however, being in the top 10 would be much better than being in the mid-20's.

--'85.

[SB Shock]

Last year WSU on 4/24 was ranked 13th in ISR. The played N. Colorado (222) and UNI (174) and won 5 of 6. After the week was over they dropped 9 places down to 22nd.

Yes, but while the ISR is a better predictor of relative team strength, it doesn't necessarily move in the same way as the RPI, does it?

I didn't realize it this morning, but Boyd has his RPI by by weeks (otherwise I would have listed it in the first place).

WSU RPI on 4/24: 14th
WSU RPI on 5/8: 21st

This period includes additionally 3xEU (61st) and 1 game versus Ark

So, I guess in at least this case the answer would be yes, RPI and ISR showed similiar trends.

The good thing (or bad thing), is that the Selection Committee can use any criteria they desire for seeding. Based purely on ISR or RPI or any stetistical tool available, there's no way WSU should have hosted a Regional, much less a Super Regional in 2007. That said, however, being in the top 10 would be much better than being in the mid-20's.

--'85.

That's ultimultely the crutch of the matter. Selection committee can do whatever they please. With WSU, due to their scheduling issue and conference affilitation will never be in position to be a no-brainer national seed, because there will be some parameter holding them back.

All they can do in the end is be the best they can be and just win and win and win.

[Shocker85]

All they can do in the end is be the best they can be and just win and win and win.

And the schedule this season sets-up interestingly for a closing the season well.

What appear to be the two toughest MVC teams, IllSt and Creighton, are early in the conference season, although BOTH are roadies. The Shox close with NIU, EU, and BU, with Evansville (currently 1-15) on the road. Also in that stretch are ORU and Arizona State at home. If the Shox can win those non-con games, they could run off a long streak, or at the very worst finish something like 10-2 or 11-1.

I guess we shouldn't get ahead of ourselves, but this sets-up interestingly.

--'85.

[SpanglerFan316]

Ratings Percentage Index (RPI)
Records Last Updated: Sunday, March 23, 2008 12:35 PM ET

Long Beach State 16 3 0.8421 10 0.6853 1 0.6330 3 7 6
Florida State 19 2 0.9048 3 0.6812 2 0.6067 10 3 3
UC Irvine 16 2 0.8889 7 0.6680 3 0.5944 17 5 7
Arizona State 22 1 0.9565 1 0.6614 4 0.5630 49 1 1
Miami 17 2 0.8947 5 0.6495 5 0.5678 44 10 5
Stanford 9 5 0.6429 71 0.6452 6 0.6460 1 19 18
North Carolina 18 4 0.8182 14 0.6446 7 0.5867 23 6 4
Missouri 18 2 0.9000 4 0.6411 8 0.5548 58 8 9
Florida 17 4 0.8095 16 0.6313 9 0.5719 38 29 33
Wichita State 12 3 0.8000 18 0.6300 10 0.5734 37 24 16

[KC Shox]

We need to go 5-0 this week to even have a chance staying in the top 10 in RPI. Binghamton and NDSU won't do us any favors as far as SOS is concerned.

[haysdb]

We need to go 5-0 this week to even have a chance staying in the top 10 in RPI. Binghamton and NDSU won't do us any favors as far as SOS is concerned.

Agreed. Our schedule gets much stronger after this week.