This week's Gridiron Power Index

[WTFCollegefootballfan]

http://www.collegesportingnews.com/articles/20151110gpi.html

[CappinHard]

http://www.collegesportingnews.com/articles/20151110gpi.html

According to this index, here are the breakdowns of SDSU, NDSU, and Ill St. wins and losses:

http://i909.photobucket.com/albums/ac299/benpierson/Playoffs_zpsjy0myi0s.jpg (http://s909.photobucket.com/user/benpierson/media/Playoffs_zpsjy0myi0s.jpg.html)

Wins - All 3 are comparable (I found it interesting how much the numbers evened out when factoring in the last 2 games)
Losses - NDSU has the clear disadvantage, Ill. St. has slight advantage with only 1 FCS loss compared to 2 FCS losses for SDSU and NDSU
Head to head - NDSU over SDSU, SDSU over Ill. St.
FBS - SDSU has advantage with win over Kansas (mediocre win), Ill. St. lost to Iowa (good loss)


When it comes down to it, the cases each team can make are:

NDSU - head to head win over SDSU
SDSU - head to head win over Ill St., better losses than NDSU, FBS win
Ill St. - only 1 FCS loss, better losses than NDSU

If the committee stresses head to head, the seeding should go: NDSU, SDSU, Ill. St.
If the committee stresses losses, the seeding should go: Ill. St., SDSU, NDSU
If the committee takes a combined approach, the seeding should go: SDSU, Ill St., NDSU

Clear as mud.

[F'N Hawks]

Question: How much human influence/bias is there in GPI?

[BisonTru]

Question: How much human influence/bias is there in GPI?

They use the Stats and Coaches poll and seven computer polls, which the high and low of the computers are dropped. So my understanding the human polls account for 2/7th of the GPI.

Someone can correct me if I'm not quite right, but that is how I understand it.

[troianus]

They use the Stats and Coaches poll and seven computer polls, which the high and low of the computers are dropped. So my understanding the human polls account for 2/7th of the GPI.

Someone can correct me if I'm not quite right, but that is how I understand it.
I believe that you are correct.

[dbackjon]

What a joke.

[jacksfan29]

What a joke.

You played a D2 and lost to Davis. Where would you expect to be? Your going to get slaughtered in the computer polls simply on the fact that your SOS is terrible but that loss to Davis is a killer.

[dbackjon]

You played a D2 and lost to Davis. Where would you expect to be? Your going to get slaughtered in the computer polls simply on the fact that your SOS is terrible but that loss to Davis is a killer.

It is not just that.

Harvard #5, Dartmouth #13?

NAU went to Cheney and beat EWU by 22. Yet are 11 spots lower.

Computer polls are very bad at spotting trends (like NAU on a roll) or factors like NAU has a freshman QB that is improving every week.

[BisonTru]

It is not just that.

Harvard #5, Dartmouth #13?

NAU went to Cheney and beat EWU by 22. Yet are 11 spots lower.

Computer polls are very bad at spotting trends (like NAU on a roll) or factors like NAU has a freshman QB that is improving every week.

The computer polls have NAU a lot closer to EWU than the human polls. One puter poll has you guys 17 spots higher than EWU. You could make the argument computer polls are better at spotting trends than the human polls.

And the Harvard argument gets old. Bottom line, we don't know how good they are and we aren't going to find out. They haven't lost in two years and the IVY isn't as weak as some believe.

[bisonboone11]

It is not just that.

Harvard #5, Dartmouth #13?

NAU went to Cheney and beat EWU by 22. Yet are 11 spots lower.

Computer polls are very bad at spotting trends (like NAU on a roll) or factors like NAU has a freshman QB that is improving every week.
Do computer rating systems even attempt to look at trends? Seems like they would be pretty complicated without trying to take that into consideration. I don't think they look at injuries either. Obviously there are flaws in any system. Tough to account for everything.

[Mayville Bison]

Wins - All 3 are comparable (I found it interesting how much the numbers evened out when factoring in the last 2 games)
Losses - NDSU has the clear disadvantage, Ill. St. has slight advantage with only 1 FCS loss compared to 2 FCS losses for SDSU and NDSU
Head to head - NDSU over SDSU, SDSU over Ill. St.
FBS - SDSU has advantage with win over Kansas (mediocre win), Ill. St. lost to Iowa (good loss)


When it comes down to it, the cases each team can make are:

NDSU - head to head win over SDSU
SDSU - head to head win over Ill St., better losses than NDSU, FBS win
Ill St. - only 1 FCS loss, better losses than NDSU

If the committee stresses head to head, the seeding should go: NDSU, SDSU, Ill. St.
If the committee stresses losses, the seeding should go: Ill. St., SDSU, NDSU
If the committee takes a combined approach, the seeding should go: SDSU, Ill St., NDSU

Clear as mud.

Couple things to remember/add.
- SDSU played SUU - if they win the Big Sky, that's going to play a big role in determining a seed.
- If UND, Weber, and Montana all win out (going to assume USD drops both since that's the assumption you were going with), you are looking at a pretty tough OOC schedule with all three having GPIs <40.
- I'm usually not the one to discredit an FBS win (look at NDSU's over the years), but Kansas isn't even using FCS levels of scholarships this year. I'd put them middle of the pack Big Sky and I'm not sure if they win a game in the MVFC except MSU. A win is a win though.
- Somewhere in there you have to put the negative side to your cases - especially if you want to use the combined approach you are hoping for. For instance
NDSU - Loss to a likely 5-6 team at home (bad loss to a bottom half Valley team)
SDSU - Two home losses - one of which was dominated by NDSU
ISU - Best win is to a 4th place MVFC team at home. (no wins against the top end of the Valley)

The committee showed last year that they stressed losses over wins by giving UNH the top seed over NDSU despite NDSU's FBS win. I don't think we've seen a head-to-head situation though before with 3 teams. I personally value the H2H and who you beat, but I can understand why each team needs to make a statement the last two weeks in hopes to get the higher seed.

[Mayville Bison]

It is not just that.

Harvard #5, Dartmouth #13?

NAU went to Cheney and beat EWU by 22. Yet are 11 spots lower.

Computer polls are very bad at spotting trends (like NAU on a roll) or factors like NAU has a freshman QB that is improving every week.

Computer polls basically have you dead even with a slight nod given to EWU. It's the human polls that are killing NAU when compared to EWU. That is all name recognition and the fact that they are the last team not named NDSU to win it all.

[DetroitFlyer]

I think the entire Ivy League is ranked above Dayton....

[CappinHard]

Couple things to remember/add.
- SDSU played SUU - if they win the Big Sky, that's going to play a big role in determining a seed.
- If UND, Weber, and Montana all win out (going to assume USD drops both since that's the assumption you were going with), you are looking at a pretty tough OOC schedule with all three having GPIs <40.
- I'm usually not the one to discredit an FBS win (look at NDSU's over the years), but Kansas isn't even using FCS levels of scholarships this year. I'd put them middle of the pack Big Sky and I'm not sure if they win a game in the MVFC except MSU. A win is a win though.
- Somewhere in there you have to put the negative side to your cases - especially if you want to use the combined approach you are hoping for. For instance
NDSU - Loss to a likely 5-6 team at home (bad loss to a bottom half Valley team)
SDSU - Two home losses - one of which was dominated by NDSU
ISU - Best win is to a 4th place MVFC team at home. (no wins against the top end of the Valley)

The committee showed last year that they stressed losses over wins by giving UNH the top seed over NDSU despite NDSU's FBS win. I don't think we've seen a head-to-head situation though before with 3 teams. I personally value the H2H and who you beat, but I can understand why each team needs to make a statement the last two weeks in hopes to get the higher seed.

All valid points, except for saying that Kansas would get beat by every MVFC team except MSU is a stretch. Doesn't really matter either way, I just don't think that would be true.

I hope the scenario that all 3 win out comes to fruition, because I'll be happy with whatever SDSU gets in that case (assuming we get a seed), but it will be fun/interesting to see what the selection committee comes up with.

[bulldog10jw]

I think the entire Ivy League is ranked above Dayton....

Even Cornell? xeyebrowx

[wow]

Couple things to remember/add.
- SDSU played SUU - if they win the Big Sky, that's going to play a big role in determining a seed.
- If UND, Weber, and Montana all win out (going to assume USD drops both since that's the assumption you were going with), you are looking at a pretty tough OOC schedule with all three having GPIs <40.
- I'm usually not the one to discredit an FBS win (look at NDSU's over the years), but Kansas isn't even using FCS levels of scholarships this year. I'd put them middle of the pack Big Sky and I'm not sure if they win a game in the MVFC except MSU. A win is a win though.
- Somewhere in there you have to put the negative side to your cases - especially if you want to use the combined approach you are hoping for. For instance
NDSU - Loss to a likely 5-6 team at home (bad loss to a bottom half Valley team)
SDSU - Two home losses - one of which was dominated by NDSU
ISU - Best win is to a 4th place MVFC team at home. (no wins against the top end of the Valley)

The committee showed last year that they stressed losses over wins by giving UNH the top seed over NDSU despite NDSU's FBS win. I don't think we've seen a head-to-head situation though before with 3 teams. I personally value the H2H and who you beat, but I can understand why each team needs to make a statement the last two weeks in hopes to get the higher seed.

I'm not saying Kansas is a good FBS team, but you are selling them short, IMO. A few things to consider:

-Iowa State beat UNI by 24 and Kansas by 25. UNI has been pretty good in the MVFC.
-SDSU crushed SUU (who currently leads the Big Sky), and barely squeaked by Kansas. Kansas would likely do quite well in the Big Sky depending on how the schedule worked out.
-Kansas has also played 4 top 25 teams, which makes them look worse than they are.

Again, not saying SDSU/Kansas is App/Michigan, but any FBS win is impressive and should be considered, IMO.

[Mayville Bison]

I'm not saying Kansas is a good FBS team, but you are selling them short, IMO. A few things to consider:

-Iowa State beat UNI by 24 and Kansas by 25. UNI has been pretty good in the MVFC.
-SDSU crushed SUU (who currently leads the Big Sky), and barely squeaked by Kansas. Kansas would likely do quite well in the Big Sky depending on how the schedule worked out.
-Kansas has also played 4 top 25 teams, which makes them look worse than they are.

Again, not saying SDSU/Kansas is App/Michigan, but any FBS win is impressive and should be considered, IMO.

Looks like I was off a little. They are using 64 scholarships as of summer ball, so they were right on par with FCS teams. http://www.kansas.com/sports/college/big-12/university-of-kansas/article26998060.html

The bolded is where I'll disagree. Portland State beating North Texas by itself doesn't show me really anything about PSU. PSU beating UNT by 59 is an impressive win. SDSU beating a still winless Kansas team by 3 doesn't show me really anything about SDSU. Had they won by 17 like they were ahead at the half, that's an impressive win.

That all being said, it's still a helluva lot better win than a loss to USD at home. xbangx

[bluehenbillk]

The GPI has been outperformed by the AGS Poll for 6 consecutive years and counting so there's that....

[dbackjon]

Computer polls basically have you dead even with a slight nod given to EWU. It's the human polls that are killing NAU when compared to EWU. That is all name recognition and the fact that they are the last team not named NDSU to win it all.

OK - then you voters get your stuff together... :)

[dbackjon]

The computer polls have NAU a lot closer to EWU than the human polls. One puter poll has you guys 17 spots higher than EWU. You could make the argument computer polls are better at spotting trends than the human polls.

And the Harvard argument gets old. Bottom line, we don't know how good they are and we aren't going to find out. They haven't lost in two years and the IVY isn't as weak as some believe.

It's easy to win when you don't play anyone.

[Herder]

Gasp!! How could anything challenge Jax St as #1 in the land!!

[ursus arctos horribilis]

Gasp!! How could anything challenge Jax St as #1 in the land!!

What is this non sequitur all about? Are people advocating it is a lock that can't be challenged?

[BisonTru]

It's easy to win when you don't play anyone.

I know when a poll doesn't have your team as high as you think they are, you need to somehow discredit them. But the Harvard argument is dumb, even in the human polls the IVY is ranked higher than the OVC. But Jacksonville St is receiving almost every single vote for the top team in the country, rightfully so. Why? Because they are doing what you would expect the top team in the country to do with their weak schedule. Harvard is doing the same outside of a close victory to Dartmouth which looks like a damn good squad as well. They might be a paper tiger, you don't know, and I don't know, and the same argument could be made for JSU. Unfortunately we won't find out if Harvard is for real, but we will find out about JSU.

Find a new argument why the computers are wrong.

[dbackjon]

I know when a poll doesn't have your team as high as you think they are, you need to somehow discredit them. But the Harvard argument is dumb, even in the human polls the IVY is ranked higher than the OVC. But Jacksonville St is receiving almost every single vote for the top team in the country, rightfully so. Why? Because they are doing what you would expect the top team in the country to do with their weak schedule. Harvard is doing the same outside of a close victory to Dartmouth which looks like a damn good squad as well. They might be a paper tiger, you don't know, and I don't know, and the same argument could be made for JSU. Unfortunately we won't find out if Harvard is for real, but we will find out about JSU.

Find a new argument why the computers are wrong.

Harvard and Dartmouth don't pass the eye test - neither would finish in the top half of either the Big Sky or the MVFC.
Jax State passes the eye test.

I don't need a new argument - since I am right.

[Herder]

Why would the Ivy care about rankings? Isn't it insulting for them to be associated with anything FCS? I wouldn't even rank them, unless they land in the AP Top 25 :)

You don't want to play anyone outside of your bubble, so what's the use of ranking you outside your 8 team bubble. I suggest the HICU poll, for the historically IVY colleges and Universities.

[BisonTru]

Harvard and Dartmouth don't pass the eye test - neither would finish in the top half of either the Big Sky or the MVFC.
Jax State passes the eye test.

I don't need a new argument - since I am right.

How many IVY games do you devote your undivided attention to weekly?

Must be quite a few considering how much faith you have in the "Jon's Eye Test."

[dbackjon]

How many IVY games do you devote your undivided attention to weekly?

Must be quite a few considering how much faith you have in the "Jon's Eye Test."

Saw the Harvard/Dartmouth game. the "battle of the best" Not impressed.

[FordhamFan]

Saw the Harvard/Dartmouth game. the "battle of the best" Not impressed.

Come onnnn man. I agree NAU is a little low in this index but the whole crap on the Ivy's thing is a terrible argument. You simply can't win it. They'll never let you. They'll never play anyone that will either shut you up or prove you right. So, maybe every single poll/index/whatever ranking Harvard ahead of NAU is a sign that maybe you're alone on this one.

P.S. Harvard would 1000% finish in the top half of the Big Sky.

[dbackjon]

Come onnnn man. I agree NAU is a little low in this index but the whole crap on the Ivy's thing is a terrible argument. You simply can't win it. They'll never let you. They'll never play anyone that will either shut you up or prove you right. So, maybe every single poll/index/whatever ranking Harvard ahead of NAU is a sign that maybe you're alone on this one.

P.S. Harvard would 1000% finish in the top half of the Big Sky.


Since they can never prove it, why bother ranking them?

I could even say that Harvard would finish last in the Big Sky, and no way of proving me wrong...

[eiu1999]

Since they can never prove it, why bother ranking them?

I could even say that Harvard would finish last in the Big Sky, and no way of proving me wrong...

or right

[FordhamFan]

Since they can never prove it, why bother ranking them?

I could even say that Harvard would finish last in the Big Sky, and no way of proving me wrong...

I would be cool if people didn't rank Ivy's. But they do. Higher than NAU. So obviously a lot of people see SOMETHING.

(And you're right, they can't prove either of us right or wrong, I just think they would easily compete in the Big Sky)

[BisonTru]

Saw the Harvard/Dartmouth game. the "battle of the best" Not impressed.

Do you have a link of the replay of the UC D/NAU?

I'd like to form an opinion. You know since we're using robust sample sizes and picking suspect performances.