Arch madness seeding

[BuBrave2006]

Tonight's WSU-UE game has some small seeding implications on a few more teams than just those involved.

With a WSU win:
*Evansville is eliminated from the #6 seed (previously 0.5% chance weighted, 4.7% unweighted), and locked into Thursday night
*Bradley is eliminated from the #6 seed (previously 0.3% chance weighted, 2.3% unweighted), and locked into Thursday night
*Missouri State is eliminated from the #7 seed (previously 0.8% chance weighted, 5.5% unweighted), and locked into Friday
*Loyola is eliminated from the #8 seed (previously 0.1% chance weighted, 3.1% unweighted)
*Drake and Loyola would be the only possible remaining Thursday-Friday flip (3.6% weighted chance, 12.5% unweighted) if Drake beats Loyola, Drake beats Bradley, and SIU beats Loyola

With a UE win:
*No teams are eliminated from potential seeds that are a current possibility, however many odds will change drastically due to the expected rarity of this outcome

Does this take into account potential changing of RPI tiebreaker? Consider the following scenario:

Bradley's current RPI is 226 and Missouri St's is 208. Assume Bradley goes 2-0 (Mo St and @Drake) and Missouri St goes 0-2 (@Bradley and Wichita). Both teams would finish 7-11 and would split the season series. If this scenario plays out, both teams would tie for 6th (likely would be the only teams at 7-11). I don't know enough about RPI to know that if it played out like this would make enough of a difference for Bradley to leapfrog Missouri St, but I would think it would be possible.

[uniftw]

Tonight's WSU-UE game has some small seeding implications on a few more teams than just those involved.

With a WSU win:
*Evansville is eliminated from the #6 seed (previously 0.5% chance weighted, 4.7% unweighted), and locked into Thursday night
*Bradley is eliminated from the #6 seed (previously 0.3% chance weighted, 2.3% unweighted), and locked into Thursday night
*Missouri State is eliminated from the #7 seed (previously 0.8% chance weighted, 5.5% unweighted), and locked into Friday
*Loyola is eliminated from the #8 seed (previously 0.1% chance weighted, 3.1% unweighted)
*Drake and Loyola would be the only possible remaining Thursday-Friday flip (3.6% weighted chance, 12.5% unweighted) if Drake beats Loyola, Drake beats Bradley, and SIU beats Loyola

With a UE win:
*No teams are eliminated from potential seeds that are a current possibility, however many odds will change drastically due to the expected rarity of this outcome

Does this take into account potential changing of RPI tiebreaker? Consider the following scenario:

Bradley's current RPI is 226 and Missouri St's is 208. Assume Bradley goes 2-0 (Mo St and @Drake) and Missouri St goes 0-2 (@Bradley and Wichita). Both teams would finish 7-11 and would split the season series. If this scenario plays out, both teams would tie for 6th (likely would be the only teams at 7-11). I don't know enough about RPI to know that if it played out like this would make enough of a difference for Bradley to leapfrog Missouri St, but I would think it would be possible.

Rough math..and I mean rough...would put MSU with an RPI of about 216 and Bradley at 220

Basically, close enough that there *could* be a shift to push Bradley over MSU. It's a small chance, but it's a chance.

[Hacksaw]

Does this take into account potential changing of RPI tiebreaker? Consider the following scenario:

Bradley's current RPI is 226 and Missouri St's is 208. Assume Bradley goes 2-0 (Mo St and @Drake) and Missouri St goes 0-2 (@Bradley and Wichita). Both teams would finish 7-11 and would split the season series. If this scenario plays out, both teams would tie for 6th (likely would be the only teams at 7-11). I don't know enough about RPI to know that if it played out like this would make enough of a difference for Bradley to leapfrog Missouri St, but I would think it would be possible.

This does take into account shifts in RPI, but no more so than what is projected from rpiforecast.com. According to that site, Bradley with two wins would be around a 224, and MSU with two losses would be around a 216. The projected RPI's I use for the scenarios are listed below. I suppose I could further sensitize RPI's in a range around those projections, but currently do not. uniftw is correct, there probably is a chance, but its quite small.

[BuBrave2006]

Does this take into account potential changing of RPI tiebreaker? Consider the following scenario:

Bradley's current RPI is 226 and Missouri St's is 208. Assume Bradley goes 2-0 (Mo St and @Drake) and Missouri St goes 0-2 (@Bradley and Wichita). Both teams would finish 7-11 and would split the season series. If this scenario plays out, both teams would tie for 6th (likely would be the only teams at 7-11). I don't know enough about RPI to know that if it played out like this would make enough of a difference for Bradley to leapfrog Missouri St, but I would think it would be possible.

This does take into account shifts in RPI, but no more so than what is projected from rpiforecast.com. According to that site, Bradley with two wins would be around a 224, and MSU with two losses would be around a 216. The projected RPI's I use for the scenarios are listed below. I suppose I could further sensitize RPI's in a range around those projections, but currently do not. uniftw is correct, there probably is a chance, but its quite small.

S**t

[chitown fanatic]

We REALLY need a TCU upset of Kansas tonight.........and a Nevada win would help too. The positive is we played tougher teams than the ears in the non con who play in power conferences. Any upset by the TCU's , Mississippis of the world could be enough to skew the numbers. I still think Elgin is an idiot for doing away with the old points system, which was perfect the way it was.(20 for beating 1st place on road, 19 at home, etc.)

[Redbirds4Life]

We REALLY need a TCU upset of Kansas tonight.........and a Nevada win would help too. The positive is we played tougher teams than the ears in the non con who play in power conferences. Any upset by the TCU's , Mississippis of the world could be enough to skew the numbers. I still think Elgin is an idiot for doing away with the old points system, which was perfect the way it was.(20 for beating 1st place on road, 19 at home, etc.)

What old points system is that? I don't remember that at all. And when was this used? and was it for just tiebreaking?

[Cdizzle]

We REALLY need a TCU upset of Kansas tonight.........and a Nevada win would help too. The positive is we played tougher teams than the ears in the non con who play in power conferences. Any upset by the TCU's , Mississippis of the world could be enough to skew the numbers. I still think Elgin is an idiot for doing away with the old points system, which was perfect the way it was.(20 for beating 1st place on road, 19 at home, etc.)

I like RPI better than those systems. I think NCSOS was even better, not sure why they went away from it.

The point system reminds me of the selection committee. "Well, you beat #1 at home, so you're clearly better. We'll just ignore those 3 losses to teams #8-10"

[BCPanther]

We REALLY need a TCU upset of Kansas tonight.........and a Nevada win would help too. The positive is we played tougher teams than the ears in the non con who play in power conferences. Any upset by the TCU's , Mississippis of the world could be enough to skew the numbers. I still think Elgin is an idiot for doing away with the old points system, which was perfect the way it was.(20 for beating 1st place on road, 19 at home, etc.)

I like RPI better than those systems. I think NCSOS was even better, not sure why they went away from it.

The point system reminds me of the selection committee. "Well, you beat #1 at home, so you're clearly better. We'll just ignore those 3 losses to teams #8-10"

Agreed. Going away from NonCon SOS was a mistake. Gave teams incentives for stepping up and scheduling well. RPI takes that into account as well, but not as well as the SOS numbers themselves.

[Cdizzle]

We REALLY need a TCU upset of Kansas tonight.........and a Nevada win would help too. The positive is we played tougher teams than the ears in the non con who play in power conferences. Any upset by the TCU's , Mississippis of the world could be enough to skew the numbers. I still think Elgin is an idiot for doing away with the old points system, which was perfect the way it was.(20 for beating 1st place on road, 19 at home, etc.)

I like RPI better than those systems. I think NCSOS was even better, not sure why they went away from it.

The point system reminds me of the selection committee. "Well, you beat #1 at home, so you're clearly better. We'll just ignore those 3 losses to teams #8-10"

Agreed. Going away from NonCon SOS was a mistake. Gave teams incentives for stepping up and scheduling well. RPI takes that into account as well, but not as well as the SOS numbers themselves.

I'd guess that fairly frequently RPI and NCSOS will be pretty close as far as in-league ranking. I just don't understand the change. At best, the RPI is just a watered down way to account for NCSOS. At worst, it gives undue benefit to teams that get to play teams that scheduled and performed well in the Non-Con.

[Hacksaw]