RPI 5/10/2010

[GoneGolfin]

As expected, the Cajuns' Base RPI (and correspondingly its Adjusted RPI) dropped a small amount after Sunday's games (.0009). Also as expected, the Cajuns passed Western Kentucky and Stanford, while Texas State passed the Cajuns. I reported it would be close as to whether the Cajuns would remain ahead of Texas A&M. Well, they did in the Kislanko Adjusted RPI (by .0003) cited below (UL #37, A&M #38). But they did not in Boyd Nation's pRPI (A&M #36, UL #38). The simple reasons why are discussed here ... https://forumeus.com/sh...91&postcount=1

The weekend series against New Orleans was the catalyst for the Cajuns dropping .0033 in the RPI (.5752 to .5719 in Adjusted RPI, .5740 to .5707 in Base RPI). Not at all horrible. But a loss would have been devastating.

The Cajuns trail Pittsburgh at #36 by a mere .0001.

After falling out of the Base RPI Top 75 (very bad for UL), Troy moved back into the Base RPI Top 75 at #73 after Sunday's win over ULM.

Northwestern State is making me a little nervous (lost 2/3 to Lamar) as their Base RPI has fallen to #69. Should Northwestern State fall out of the Base RPI Top 75, two things would happen ...

1) Cajuns lose .0012 in RPI Bonus
2) Cajuns lose two wins against the Base RPI Top 75 (record would be 10-11)

Finally, it is important that Middle Tennessee continue to play well. The Blue Raiders have moved their Base RPI up to #59.

So ... Northwestern State, Troy, and Middle Tennessee are friends of the Cajuns.

Important measures:
- Record vs. Base RPI Top 75: 12-11
- Record vs. Base RPI Top 100: 14-14
- Record vs. Base RPI Top 50: 1-3
- Non-Conference RPI: .5657, #39
- Conference RPI ranking: #5
- Road RPI: .5532, #46
- SOS: .5411, #55

26 0.5874 Rice 27 0.5832
27 0.5865 Virginia Tech 26 0.5853
28 0.5815 Oregon 33 0.5755
29 0.5811 North Carolina 28 0.5811
30 0.5805 California 30 0.5769
31 0.5793 Oregon State 29 0.5793
32 0.5790 New Mexico 35 0.5718
33 0.5780 Florida Atlantic 31 0.5768
34 0.5762 College of Charleston 32 0.5762
35 0.5731 Texas State 37 0.5713
36 0.5720 Pittsburgh 34 0.5720
37 0.5719 Louisiana 38 0.5707
38 0.5716 Texas A&M 36 0.5716
39 0.5703 Stanford 39 0.5703
40 0.5695 Western Kentucky 44 0.5629
41 0.5678 The Citadel 41 0.5678
42 0.5671 Liberty 40 0.5683
43 0.5660 Kansas State 43 0.5630
44 0.5648 Texas Tech 42 0.5648
45 0.5647 Tennessee 45 0.5605
46 0.5645 Elon 49 0.5579
47 0.5635 Washington State 48 0.5581
48 0.5627 Boston College 46 0.5597
49 0.5598 Southeastern Louisiana 53 0.5556
50 0.5595 Baylor 47 0.5595

Brian

[raginsaints]

That is such a jampacked group of schools right there.

[GoneGolfin]

_ That is such a jampacked group of schools right there. _

Indeed. To give you a better idea ...

A loss to New Orleans yesterday would have resulted in a Cajun Adjusted RPI of .5654 ... a fall of an additional .0065 on top of the .0009 loss for simply playing and beating them. That would have been a fall to #44 ... sandwiched tightly between Kansas State and Texas Tech (there would be some small unaccounted for ripple effects in some other RPIs due to the UNO win, but not enough to change the RPI ranking here).

Brian

[cajunhawk]

Brian, Ill play devils advocate here. Why in gods good name is that purple mess to the east still in the top 15 rpi wise when they have cratered so hard lately? It makes no sense in reality but Im sure the rpi can reasonably explain it. Right? Right?

[AstroCajun]

_ Brian, Ill play devils advocate here. Why in gods good name is that purple mess to the east still in the top 15 rpi wise when they have cratered so hard lately? It makes no sense in reality but Im sure the rpi can reasonably explain it. Right? Right? _

But you really aren't playing the role of devil's advocate with your question.

The true devil's advocate points out that our success, (along with the upper echelon of the SEC, and to a lesser degree Tulane and SLU), are important elements that keep LSU's RPI strong.

I'll defer to Brian on the science, though.

[GoneGolfin]

_ Brian, Ill play devils advocate here. Why in gods good name is that purple mess to the east still in the top 15 rpi wise when they have cratered so hard lately? It makes no sense in reality but Im sure the rpi can reasonably explain it. Right? Right? _

Because the RPI formula is dominated by Strength of Schedule (SOS), teams that can assemble a strong schedule should fare well. 50% of the RPI is represented by the W-L% of your opponents (OWP). Then 25% of the RPI formula is represented by the W-L% of these same opponents' opponents (OOWP). Only 25% is represented by your W-L% (WP). Playing in a power conference allows you to have a solid OWP and OOWP. You can have an RPI in the Top 30, go 1-2 in a weekend series against a Top 20 opponent, and have your RPI hold, maybe go a little higher. Sweeping a series is not easy in college baseball. Taking one game in a three game series is considerably easier than taking two in a four game series ... especially if two of those games happen to be played as part of a double-header.

Another significant benefit ... if you are from a big conference, you can get away with playing a limited # of road games (LSU is 4-8 on the road this season, 30-7 at home). Remember that the Base RPI does not distinguish between home and road games. The Adjusted RPI (used by the NCAA selection committee) does award bonuses for road wins, but only non-conference road wins against the Top 75 Base RPI.

Recall that in 2007, the Cajuns played more road games than home games (nearly unheard of for an NCAA Tournament team). This is one of the many points I made in support of a #1 seed for the Cajuns. But I think the committee unfortunately did not notice this fact.

Now, it is difficult playing Top 40 teams week in, week out. And that should be taken into account. But I do think that in aggregate, the RPI benefits the big conferences and the East Coast ... but is a hindrance to the non-elite conferences and the West Coast. Implementing a formula chance to take into account all road games I think would make it a more level playing field.

I do think this year is a year where the Cajuns are in about the right spot RPI-wise ... mostly because they have not had to play nearly as many road games as in the past. Of course, they have five more coming (three more home).

Brian

[GoneGolfin]

_ But you really aren't playing the role of devil's advocate with your question.

The true devil's advocate points out that our success, (along with the upper echelon of the SEC, and to a lesser degree Tulane and SLU), are important elements that keep LSU's RPI strong.

I'll defer to Brian on the science, though. _

To some degree, yes (more in most years). Although LSU has only played the Cajuns, Tulane, and SLU once each this season ... so there is not much weighting here (they do play Tulane one more time). It is the SEC that gives them the SOS ... and the other teams in the SEC are able to load up on home games as well (providing them an RPI advantage that is fed to the rest of the SEC teams when they play each other). Meanwhile, LSU has not played a non-conference road game this season (they have played two neutral games ... though one was in Metairie (really a home game) ... and the other was in Shreveport). They do play at Tulane next week. But that will be the only non-conference road game they play this season. That is quite an advantage.

Brian

[CajunT]

_ Because the RPI formula is dominated by Strength of Schedule (SOS), teams that can assemble a strong schedule should fare well. 50% of the RPI is represented by the W-L% of your opponents (OWP). Then 25% of the RPI formula is represented by the W-L% of these same opponents' opponents (OOWP). Only 25% is represented by your W-L% (WP). Playing in a power conference allows you to have a solid OWP and OOWP. You can have an RPI in the Top 30, go 1-2 in a weekend series against a Top 20 opponent, and have your RPI hold, maybe go a little higher. Sweeping a series is not easy in college baseball. Taking one game in a three game series is considerably easier than taking two in a four game series ... especially if two of those games happen to be played as part of a double-header.

Another significant benefit ... if you are from a big conference, you can get away with playing a limited # of road games (LSU is 4-8 on the road this season, 30-7 at home). Remember that the Base RPI does not distinguish between home and road games. The Adjusted RPI (used by the NCAA selection committee) does award bonuses for road wins, but only non-conference road wins against the Top 75 Base RPI.

Recall that in 2007, the Cajuns played more road games than home games (nearly unheard of for an NCAA Tournament team). This is one of the many points I made in support of a #1 seed for the Cajuns. But I think the committee unfortunately did not notice this fact.

Now, it is difficult playing Top 40 teams week in, week out. And that should be taken into account. But I do think that in aggregate, the RPI benefits the big conferences and the East Coast ... but is a hindrance to the non-elite conferences and the West Coast. Implementing a formula chance to take into account all road games I think would make it a more level playing field.

I do think this year is a year where the Cajuns are in about the right spot RPI-wise ... mostly because they have not had to play nearly as many road games as in the past. Of course, they have five more coming (three more home).

Brian _

Formula needs to be changed in my opinion, similar to basketball.

[cajunhawk]

It seems like its just tailor made for the big conferences to basicly coast their way into a regional. Unlike Football and just like Basketball, the little guys still get their chance. Think Fresno State. Winning the conference tourney to get in and didnt look back. Thats the great thing about college baseball and basketball and everything thats wrong with college football. It's no secret the big money schools have it alot easier than smaller money schools from a lesser conference. This is today's NCAA. Its not about the kids, its about the money. This road game b.s. is what is really really wrong with the NCAA. If you're so damn good, if you play in the best conference in the nation, you play at the highest level, take your show on the road. Take the skirt off nancies!! Rich get richer poor get poorer. Diatribe on NCAA inequality over!!!

Another question Brian, how many SEC losses in a row would it take for LSU to drop the same amount the Cajuns would have dropped by losing one game to UNO this past weekend? I know thats an abstract question but losing three consecutive conference series can in no way qualify a team to be a top 15 team in the nation. Even the polls have given up on them.

[GoneGolfin]

_ Another question Brian, how many SEC losses in a row would it take for LSU to drop the same amount the Cajuns would have dropped by losing one game to UNO this past weekend?

Just wait until this weekend against Kentucky and you will find out. Of course, that would probably not be enough losses.

OK, that was uncalled for.

Answering your question would be very time consuming without writing software as I would need to re-calc OOWP several times. As a simple alternative, I am not going to use "SEC losses", but I will use the average team on LSU's schedule thus far. That is, I will keep LSU's OWP and OOWP the same. This is not the same as "SEC losses", as their non-conference schedule (25 games) RPI rank is #86. That is, their non-conference SOS (.5248) is considerably lower than their conference SOS (right about .6200).

The Cajuns would have dropped another .0065 with a single loss to New Orleans this weekend. .0074 in total with a loss to New Orleans.

Two more losses (of the opponent quality that is the average of LSU's schedule) would have dropped LSU .0068, about the equivalent of a UL loss to UNO.

For a more crude measure (not exact math), but still useful ... LSU dropped exactly .0040 in RPI with the two losses to Vanderbilt on Saturday and Sunday (this rate would diminish slightly with more losses). Of course, this depends on how their opponents performed and their opponents' opponents performed on Saturday and Sunday (LSU wins and losses are not the only variable). But you get the picture. About four losses to Vanderbilt would have dropped them a similar amount to UL losing a game to UNO. LSU beating Vanderbilt on Friday increased their RPI .0031.

Brian