FB: Minnesota Intercollegiate Athletic Conference

[Pat Coleman]

So the game against rose Hulman is big-if they win this week, wouid they win the conference? Also, sju gets credit for mounts record as Rose Hulman played them in game 1.

SJU also gets credit for Defiance's (likely) 0-10 record a RH played them Week 9.

This is true that the entire HCAC will be part of SJU's OOWP. RHIT's OWP is just .407, and would get factored into SJU's OOWP.

https://www.d3football.com/seasons/2019/schedule?tmpl=sos-template

[sju56321]

Well never mind then

[wally_wabash]

SOS is absolutely used for Regional Rankings as a primary criteria.

The SOS methodology has been thoroughly critiqued already, but it's pretty clearly not a good indicator of actual schedule strength.

My other big beef with the regional rankings is the reliance on RR wins for future rankings. For one, it's a feedback loop, but the other issue, if it's given any significant weight (aside from as a tiebreaker, in which case I'm less bothered), why should a win over the #10 team (Martin Luther in this case... cough cough) count for significantly more than wins over the teams that would fall 11-15 (in this case, likely Gustavus and St John's or in the WIAC, LaCrosse and Platteville).

I'd certainly be in favor of extending the rankings beyond just 10, or using a once ranked, always ranked system where teams continue to get credit for results against ranked opponents regardless of which of the three public rankings they showed up in. 

But no matter where you draw the line, 10 or 12 or 15, you do have to draw the line somewhere.  Right now, that line is at #10. 

[jamtod]

SOS is absolutely used for Regional Rankings as a primary criteria.

The SOS methodology has been thoroughly critiqued already, but it's pretty clearly not a good indicator of actual schedule strength.

My other big beef with the regional rankings is the reliance on RR wins for future rankings. For one, it's a feedback loop, but the other issue, if it's given any significant weight (aside from as a tiebreaker, in which case I'm less bothered), why should a win over the #10 team (Martin Luther in this case... cough cough) count for significantly more than wins over the teams that would fall 11-15 (in this case, likely Gustavus and St John's or in the WIAC, LaCrosse and Platteville).

I'd certainly be in favor of extending the rankings beyond just 10, or using a once ranked, always ranked system where teams continue to get credit for results against ranked opponents regardless of which of the three public rankings they showed up in. 

But no matter where you draw the line, 10 or 12 or 15, you do have to draw the line somewhere.  Right now, that line is at #10.

And that's fine I guess if it's clear that a win over #10 means very little and a win over #5 (or wherever Bethel is) means a lot more. I understand that you have to draw the line somewhere, but there needs to be a clear weighting that winds down to almost nothing at the end.

[wally_wabash]

SOS is absolutely used for Regional Rankings as a primary criteria.

The SOS methodology has been thoroughly critiqued already, but it's pretty clearly not a good indicator of actual schedule strength.

My other big beef with the regional rankings is the reliance on RR wins for future rankings. For one, it's a feedback loop, but the other issue, if it's given any significant weight (aside from as a tiebreaker, in which case I'm less bothered), why should a win over the #10 team (Martin Luther in this case... cough cough) count for significantly more than wins over the teams that would fall 11-15 (in this case, likely Gustavus and St John's or in the WIAC, LaCrosse and Platteville).

I'd certainly be in favor of extending the rankings beyond just 10, or using a once ranked, always ranked system where teams continue to get credit for results against ranked opponents regardless of which of the three public rankings they showed up in. 

But no matter where you draw the line, 10 or 12 or 15, you do have to draw the line somewhere.  Right now, that line is at #10.

And that's fine I guess if it's clear that a win over #10 means very little and a win over #5 (or wherever Bethel is) means a lot more. I understand that you have to draw the line somewhere, but there needs to be a clear weighting that winds down to almost nothing at the end.

I think we've been given what amounts to confirmation that not all results and ranked opponents are equal.  Winning (or losing) against a team ranked first or second in their region is not identical to winning or losing to a team ranked ninth or tenth.  There's not a codified, perfectly graduated scale for that, but I think past chairs have been clear on that point.

[jamtod]

SOS is absolutely used for Regional Rankings as a primary criteria.

The SOS methodology has been thoroughly critiqued already, but it's pretty clearly not a good indicator of actual schedule strength.

My other big beef with the regional rankings is the reliance on RR wins for future rankings. For one, it's a feedback loop, but the other issue, if it's given any significant weight (aside from as a tiebreaker, in which case I'm less bothered), why should a win over the #10 team (Martin Luther in this case... cough cough) count for significantly more than wins over the teams that would fall 11-15 (in this case, likely Gustavus and St John's or in the WIAC, LaCrosse and Platteville).

I'd certainly be in favor of extending the rankings beyond just 10, or using a once ranked, always ranked system where teams continue to get credit for results against ranked opponents regardless of which of the three public rankings they showed up in. 

But no matter where you draw the line, 10 or 12 or 15, you do have to draw the line somewhere.  Right now, that line is at #10.

And that's fine I guess if it's clear that a win over #10 means very little and a win over #5 (or wherever Bethel is) means a lot more. I understand that you have to draw the line somewhere, but there needs to be a clear weighting that winds down to almost nothing at the end.

I think we've been given what amounts to confirmation that not all results and ranked opponents are equal.  Winning (or losing) against a team ranked first or second in their region is not identical to winning or losing to a team ranked ninth or tenth.  There's not a codified, perfectly graduated scale for that, but I think past chairs have been clear on that point.

Then I feel better about it. I just get annoyed to see the simplistic comparisons showing X team as 2-0 against RRO as if that is actually meaningful on its own.

[Redtooth]

It's too bad that each year the Regional Rankings committee decides to reorder the formula that gets used to determine where teams are ranked in the West Region...other than the default selection of the #1 team being from the WIAC.  The way strength of schedule is used is flawed as pointed out last week by Logan Hansen on Twitter....frankly the D3 selection committee should use the Hansen Ratings system to rank teams regionally and establish the playoff pods.....his math/algorithms are far cleaner and would provide much stronger playoff brackets.  The current #3 in the West is his #30.......something tells me Chapman as a #3 has ZERO bearing in reality.  Their best win is over Redlands that beat Linfield and somehow using the transitive property is better than SJU's wins over than #6 Bethel and #11 UST.  I have no doubt that SJU would have been ranked below Chapman even with a win against Concordia.

No other college football level has this level of playoff ambiguity...sad really

[Pat Coleman]

I have no doubt that SJU would have been ranked below Chapman even with a win against Concordia.

In no way is this true. St. John's advanced to the quarterfinals last year, and if they had beaten Concordia and remained unbeaten, that would have elevated them to the No. 2 spot in the West.

[sju56321]

Interesting-you know sju didn't get blown out by CC-A one point loss in OT and they go from 2 to 5?
I tend to think Red is right-this loss was bad in the eyes of the ncaa-and they would have dropped them with a one point win. Otherwise it makes no sense to say one point either way is a 3 point drop.

[sju56321]

We will have to agree to disagree then

[Pat Coleman]

Interesting-you know sju didn't get blown out by CC-A one point loss in OT and they go from 2 to 5?
I tend to think Red is right-this loss was bad in the eyes of the ncaa-and they would have dropped them with a one point win. Otherwise it makes no sense to say one point either way is a 3 point drop.

Nah, man, this is the way it works. It's not really new -- generally all of the unbeaten teams who have played reasonable competition are at the top of the ranking. And a loss is still a loss, even if it is by one point in overtime.

The bright spot for SJU is that Redlands plays the worst two teams in its league the next two weeks and its SOS will drop, probably substantially. Keep an eye on that and how those two teams shake out.

[bleedpurple]

Interesting-you know sju didn't get blown out by CC-A one point loss in OT and they go from 2 to 5?
I tend to think Red is right-this loss was bad in the eyes of the ncaa-and they would have dropped them with a one point win. Otherwise it makes no sense to say one point either way is a 3 point drop.

Disagree, a win and lose are night and day different in this situation regardless if it was one point or not. Win the game and they are still #2 in the west.

Agreed. And actually in any situation.  Win the game and you did what you showed up at the stadium to do. Lose and you failed. Period. Margin of victory is part of the hair splitting that happens in sorting out top seeds.  But when you go 8-0 instead of 7-1, hairs won't be split to the degree you are suggesting. Ya'll got dumped into the criteria bucket with the loss because there are a lot of 1 loss teams.

[wally_wabash]

I think we've been given what amounts to confirmation that not all results and ranked opponents are equal.  Winning (or losing) against a team ranked first or second in their region is not identical to winning or losing to a team ranked ninth or tenth.  There's not a codified, perfectly graduated scale for that, but I think past chairs have been clear on that point.

Then I feel better about it. I just get annoyed to see the simplistic comparisons showing X team as 2-0 against RRO as if that is actually meaningful on its own.

Maybe we don't do a good enough job of getting into the weeds when digging around in results against ranked opponents.  There is definitely more to it, at least I believe there should be, than simply counting wins and losses against other teams on the rankings lists. 

To whit:

It's too bad that each year the Regional Rankings committee decides to reorder the formula that gets used to determine where teams are ranked in the West Region...

There isn't a formula for this. It would be so much easier if there was!  The primary criteria are spelled out in the handbook, without any kind of ordered priority. 

Interesting-you know sju didn't get blown out by CC-A one point loss in OT and they go from 2 to 5?
I tend to think Red is right-this loss was bad in the eyes of the ncaa-and they would have dropped them with a one point win. Otherwise it makes no sense to say one point either way is a 3 point drop.

Win percentage is a primary criteria.  That loss, even by one point, is pretty damaging.  The St. John's situation demonstrates how little margin for error there is. 

[sju56321]

Hold on-I'm not saying sju should be 2 or 3 -the argument is vs another one loss team. I get that you are dumped in with other one loss teams-I think they question is wins over bethel and ust are worth more than a win over Linfield-based on where they sit. And is Hansen's metric better.
Pat made an interesting statement about the last two games for Redlands-but sju has hamline this week-bottom.

[Pat Coleman]

Hold on-I'm not saying sju should be 2 or 3 -the argument is vs another one loss team. I get that you are dumped in with other one loss teams-I think they question is wins over bethel and ust are worth more than a win over Linfield-based on where they sit. And is Hansen's metric better.
Pat made an interesting statement about the last two games for Redlands-but sju has hamline this week-bottom.

Good point about still having Hamline, but Redlands has two Hamlines.

The win vs. UST isn't worthy any extra beyond being a win against a 6-2 team at the moment because UST isn't ranked.