10/13/08 Gridiron Power Index (GPI), James Madison Continues as No. 1

[CSN-info]

http://www.championshipsubdivisionnews.com/skins/andreas_01/img/GPI.JPG

The Gridiron Power Index (GPI), the index ranking for the NCAA Division I FCS and a top indicator of at-large playoff selection continues with James Madison in the top spot again this week.

The Colonial Athletic Association, the largest league in the FCS has seven teams in the top 25; the Southern Conference has five; the Missouri Valley Football Conference placed four; the Big Sky and the Southland Conferences have three each; the Great West Football Conference notched a pair; and the Big South Conference plus the Ivy League have one each.

10/13/2008 GPI Top 25
1. James Madison (1.00)
2. Montana (4.75)
3T. Cal Poly (5.00)
3T. Appalachian St (5.00)
5. Villanova (5.88)
6. Richmond (6.38)
7. Wofford (7.63)
8. Weber St (8.50)
9. Massachusetts (8.63)
10. Elon (8.88)
11. Northern Iowa (10.38)
12. N Arizona (13.00)
13. New Hampshire (13.50)
14. W Illinois (14.25)
15. Cent Arkansas (14.38)
16. McNeese St (17.63)
17. William & Mary (18.63)
18. Furman (19.75)
19. S Illinois (19.88)
20. Liberty (20.00)
21. Ga Southern (23.00)
22. Harvard (23.38)
23. S Dakota St (23.88)
24. Sam Houston St (24.00)
25T. UC Davis (24.25)
25T. Northeastern (24.25)

Conference Ranking:
Rank, League, Total Average
1. Colonial Athletic Association (23.34)
2. Southern Conference (29.07)
3. Great West Football Conference (30.65)
4. Big Sky Conference (31.15)
5. Southland Conference (33.46)
6. Missouri Valley Football Conference (34.91)
7. Ivy League (47.32)
8. Ohio Valley Conference (49.84)
9. Big South Conference (50.99)
10. Patriot League (52.65)
11. Mid-Eastern Athletic Conference (62.44)
12. Northeast Conference (66.63)
13. Southwestern Athletic Conference (69.13)
14. Pioneer Football League (74.38)
15. Independents (80.3)

Read more ... http://www.championshipsubdivisionnews.com/index.php?blog=5&title=10-13-08-gridiron-power-index-gpi-james--1&more=1&c=1&tb=1&pb=1

[dbackjon]

Montana, Weber and NAU all top 12, like I was saying...

[kirkblitz]

Big south is moven up!!!!!!!!!!!!!!!

[Syntax Error]

Let's try this again. A little analysis because the polls and the computers are at odds right now.


Team GPI Polls GPI Computer ARC
James Madison 1 1.00 1.00
Montana 2 7.33 3.20
Cal Poly 3 3.67 5.80
Appalachian St 3 2.00 6.80
Villanova 5 8.00 4.60
Richmond 6 8.67 5.00
Wofford 7 4.67 9.40
Weber St 8 18.67 2.40
Massachusetts 9 9.00 8.40
Elon 10 5.67 10.80
Northern Iowa 11 5.00 13.60
N Arizona 12 17.67 10.20
New Hampshire 13 11.67 14.60
W Illinois 14 16.00 13.20
Cent Arkansas 15 13.33 15.00
McNeese St 16 11.67 21.20
William & Mary 17 25.67 14.40
Furman 18 18.33 20.60
S Illinois 19 14.00 23.40
Liberty 20 15.00 23.00
Ga Southern 21 24.67 22.00
S Dakota St 23 21.33 25.40
Jacksonville St 28 24.00 27.20
N Dakota St 29 18.67 32.00
Tennessee St 32 22.33 35.60
The Citadel 35 25.33 37.20
Hampton 53 23.67 55.00
Only Top 25 poll teams are shown.

[Tod]

A little analysis because the polls and the computers are at odds right now.


Team GPI Polls GPI Polls Difference
James Madison 1 1.00 0.00
Montana 2 7.33 -5.33
Cal Poly 3 3.67 -0.67
Appalachian St 3 2.00 1.00
Villanova 5 8.00 -3.00
Richmond 6 8.67 -2.67
Wofford 7 4.67 2.33
Weber St 8 18.67 -10.67
Massachusetts 9 9.00 0.00
Elon 10 5.67 4.33
Northern Iowa 11 5.00 6.00
N Arizona 12 17.67 -5.67
New Hampshire 13 11.67 1.33
W Illinois 14 16.00 -2.00
Cent Arkansas 15 13.33 1.67
McNeese St 16 11.67 4.33
William & Mary 17 25.67 -8.67
Furman 18 18.33 -0.33
S Illinois 19 14.00 5.00
Liberty 20 15.00 5.00
Ga Southern 21 24.67 -3.67
S Dakota St 23 21.33 1.67
Jacksonville St 28 24.00 4.00
N Dakota St 29 18.67 10.33
Tennessee St 32 22.33 9.67
The Citadel 35 25.33 9.67
Hampton 53 23.67 29.33
Average 1.96
Positive number means the polls have them ranked higher than the GPI.

Only Top 25 poll teams are shown.

Once again, the Big Sky is under-appreciated.

xsmhx xsmhx xsmhx

[Syntax Error]

Team Polls GPI ARC Difference
James Madison 1.00 1.00 0.00
Montana 7.33 3.20 4.13
Cal Poly 3.67 5.80 -2.13
Appalachian St 2.00 6.80 -4.80
Villanova 8.00 4.60 3.40
Richmond 8.67 5.00 3.67
Wofford 4.67 9.40 -4.73
Weber St 18.67 2.40 16.27
Massachusetts 9.00 8.40 0.60
Elon 5.67 10.80 -5.13
Northern Iowa 5.00 13.60 -8.60
N Arizona 17.67 10.20 7.47
New Hampshire 11.67 14.60 -2.93
W Illinois 16.00 13.20 2.80
Cent Arkansas 13.33 15.00 -1.67
McNeese St 11.67 21.20 -9.53
William & Mary 25.67 14.40 11.27
Furman 18.33 20.60 -2.27
S Illinois 14.00 23.40 -9.40
Liberty 15.00 23.00 -8.00
Ga Southern 24.67 22.00 2.67
S Dakota St 21.33 25.40 -4.07
Jacksonville St 24.00 27.20 -3.20
N Dakota St 18.67 32.00 -13.33
Tennessee St 22.33 35.60 -13.27
The Citadel 25.33 37.20 -11.87
Hampton 23.67 55.00 -31.33
Average -3.11
Positive number means computers have the team ranked higher than the polls.

[Syntax Error]

So generally you could say the polls are over-rating teams compared to the GPI ARC and GPI.

[Dukie95]

Looks like the Big Sky and CAA are underrated by the humans and the SoCon is overrated.

I guess you can't convince a computer to forget about the Elon/UR game.

[ccd494]

Last team out? Maine. Wow.

[BeauFoster]

Montana...#2? Must have put a virus on the disk or something.

[gophoenix]

Looks like the Big Sky and CAA are underrated by the humans and the SoCon is overrated.

I guess you can't convince a computer to forget about the Elon/UR game.

This is an idiotic statement. Computers and people rank differently. Computers rank cumulatively no matter what, which is still biased. People rank biased also but more upon what you've done lately that what you've done cumulatively. Both have their own merits and downfalls. So, depending on what you are measuring both types are accurate. Computers also ignore conditions, like the hurricanes that games were played in which affect point spreads.

[Syntax Error]

Polls Rankings vs. GPI

Over Team
-29.33 Hampton
-10.33 N Dakota St
-9.67 Tennessee St
-9.67 The Citadel
-6.00 Northern Iowa
-5.00 S Illinois
-5.00 Liberty
-4.33 Elon
-4.33 McNeese St
-4.00 Jacksonville St
-2.33 Wofford
-1.67 Cent Arkansas
-1.67 S Dakota St
-1.33 New Hampshire
-1.00 Appalachian St

Under Team
10.67 Weber St
8.67 William & Mary
5.67 N Arizona
5.33 Montana
3.67 Ga Southern
3.00 Villanova
2.67 Richmond
2.00 W Illinois
0.67 Cal Poly
0.33 Furman
0.00 James Madison
0.00 Massachusetts

Polls Rankings vs. GPI ARC

Over Team
-31.33 Hampton
-13.33 N Dakota St
-13.27 Tennessee St
-11.87 The Citadel
-9.53 McNeese St
-9.40 S Illinois
-8.60 Northern Iowa
-8.00 Liberty
-5.13 Elon
-4.80 Appalachian St
-4.73 Wofford
-4.07 S Dakota St
-3.20 Jacksonville St
-2.93 New Hampshire
-2.27 Furman
-2.13 Cal Poly
-1.67 Cent Arkansas

Under Team
16.27 Weber St
11.27 William & Mary
7.47 N Arizona
4.13 Montana
3.67 Richmond
3.40 Villanova
2.80 W Illinois
2.67 Ga Southern
0.60 Massachusetts
0.00 James Madison

The numbers are numbers over and under ranked. Example: -29 means they are ranked 29 spots too high. 16 means they are ranked 16 spots too low.

[Syntax Error]

definitely looks like some teams are being seriously

over-ranked:
Hampton
N Dakota St
Tennessee St
The Citadel

and under-ranked:
Weber St
William & Mary
N Arizona

At least at this point in the season by the polls.

THIS IS NOT SMACK.

[GolfingGriz]

According to the GPI, JMU and UM have the best FCS wins and ranked #1 and #2. Rematch of 04???? Haha long, long ways away.

[uofmman1122]

Montana...#2? Must have put a virus on the disk or something.Oh, and I imagine ASU rightfully deserves the #2 spot, right?

Remind me again, who has ASU beaten this year? xrolleyesx

[Appstate29]

Oh, and I imagine ASU rightfully deserves the #2 spot, right?

Remind me again, who has ASU beaten this year? xrolleyesx

who has UM beaten this year?

[GolfingGriz]

who has UM beaten this year?

Cal Poly...#3...on the road...

[BeauFoster]

Oh, and I imagine ASU rightfully deserves the #2 spot, right?

Remind me again, who has ASU beaten this year? xrolleyesx

Notice I never said that ASU deserved to be #2...all I said is that Montana didn't deserve it.

I know that computer rankings aren't perfect. But lets just take a look at the Sauceda poll. Not even counting where the FCS teams rank, this poll has Duke ranked 23. Duke. Ranked 23. Mike Tyson would say that is ludacwis. Sure, the Blue Devils aren't bad this year, but they aren't a top 25 team.

Massey has North Carolina ranked above Penn State. I've watched both teams play more than once, it's my opinion that PSU is better (it's the opinion of virtually every human polster in existence, too).

The Self Poll has Ohio State ranked above USC, the same Trojan team that dismantled OSU just a few weeks ago.

The point is, computer rankings just don't work...especially this early in the season. There are plenty more examples, and if I had the time, I could list them all here. The post would be 4,000 lines long.

[uofmman1122]

Notice I never said that ASU deserved to be #2...all I said is that Montana didn't deserve it.Oh, really?

And where would you say ASU should be ranked? I mean, if Montana doesn't deserve to be #2 with more quality wins than ASU, SURELY you had ASU ranked lower than Montana. :D

[BeauFoster]

Oh, really?

And where would you say ASU should be ranked? I mean, if Montana doesn't deserve to be #2 with more quality wins than ASU, SURELY you had ASU ranked lower than Montana. :D

I will say that ASU, with losses to the #1 team in FCS and a top 15 FBS team, is very good. Probably better than the 13 that one of the computer polls has (Massey). And ASU hasn't nearly been beaten by a D-II school this season, nor were they beaten by a team ranked 18 in both the AGS and Sports Network polls (ranked 20 in the coaches poll).

[uofmman1122]

I will say that ASU, with losses to the #1 team in FCS and a top 15 FBS team, is very good. Probably better than the 13 that one of the computer polls has (Massey). And ASU hasn't nearly been beaten by a D-II school this season, nor were they beaten by a team ranked 18 in both the AGS and Sports Network polls (ranked 20 in the coaches poll).Despite all that, Montana has still actually beaten a top 5 team, on the road, no less. xlolxxwhistlex

I'm just playing devil's advocate here. xsmiley_wix

FWIW, I actually had Montana at #9 and ASU at #2, but I also think any team in the top 15 could beat any other team in the top 15. I think Montana has the capacity to be #1, but only when they start playing the football they're capable of. Our defense was dominating last week. Had we played Weber with that same defensive effort, we would be undefeated.

On the same token, ASU has not looked as stellar as I thought they would. Cruising ahead in the first half, and then slacking off is going to bite them in the arse eventually, like it did against JMU. Make a statement win at GSU this week, and you become my #2 (or #1, if JMU slips).

There are definitely cases to be made for both. :p

[OL FU]

definitely looks like some teams are being seriously

over-ranked:
Hampton
N Dakota St
Tennessee St
The Citadel

and under-ranked:
Weber St
William & Mary
N Arizona

At least at this point in the season by the polls.

THIS IS NOT SMACK.

I understand that it is not smack and that the numbers are the numbers but it does assume that the GPI is dead on accurate and if it is dead on accurate it should not change one bit from now until the end of the year except for possible injuries and personnel changesxnodx But like all rankings it will change as experience tells it. That isn't a bad thing it is a good thing but it does imply that dead on accuracy is not possible.


I would say that people should certainly consider the big number swings from the polls to the GPI, but considering the underrated or overrated numbers posted previously five or less, maybe ten or less, are probably meaningless.xtwocentsx

[BeauFoster]

Despite all that, Montana has still actually beaten a top 5 team, on the road, no less. xlolxxwhistlex

I'm just playing devil's advocate here. xsmiley_wix

FWIW, I actually had Montana at #9 and ASU at #2, but I also think any team in the top 15 could beat any other team in the top 15. I think Montana has the capacity to be #1, but only when they start playing the football they're capable of. Our defense was dominating last week. Had we played Weber with that same defensive effort, we would be undefeated.

On the same token, ASU has not looked as stellar as I thought they would. Cruising ahead in the first half, and then slacking off is going to bite them in the arse eventually, like it did against JMU. Make a statement win at GSU this week, and you become my #2 (or #1, if JMU slips).

There are definitely cases to be made for both. :p

I read ya! You know me, I like to stir the pot, too. Truth is, I haven't seen UM play, in person or on TV. All I go on is what Wolfman says, and he thinks that you guys should be ranked around 110, or so xsmiley_wix

I agree that not playing a full 4 qtrs will eventually bite ASU in the butt (it already has), but if they put it together, watch out. There's no points for potential (Tommy Bowden, anyone), but ASU has it. There are plenty of other teams that have it, too. It all depends on who puts it together in December.

[appfan2008]

Everyday there is more reinforcing evidence that the big sky needs to be ranked higher next week... I for one will certainly recognize that in my poll next monday...

[Grizalltheway]

What's that? The big fluff is ranked ahead of the MVFC? Hmm...xreadx

[89Hen]

definitely looks like some teams are being seriously

over-ranked:

and under-ranked:

At least at this point in the season by the polls.

THIS IS NOT SMACK.
You are making one BIG assumption... that the computers are the ones that are right. xlolx xlolx xlolx

Massey Top 25
1. James Madison
2. Montana
3. Villanova
4. Weber St
5. Massachusetts
6. Elon
7. Wofford
8. Cal Poly
9. Richmond
10. Cent Arkansas
11. N Arizona
12. Furman
13. Appalachian St
14. William & Mary
15. Harvard
16. New Hampshire
17. Liberty
18. Cornell
19. Maine
20. Northern Iowa
21. W Illinois
22. Ga Southern
23. McNeese St
24. Northwestern St
25. So Illinois

[WrenFGun]

I don't mean to promulgate the UNH bashing lead by me in more threads on this board, but I'm extremely surprised UNH is so high in the GPI. It looks like W&M may wind up being a "quality loss," but they must count the FBS win quite a bit, because the other three wins look pretty ugly.

[Syntax Error]

I understand that it is not smack and that the numbers are the numbers but it does assume that the GPI is dead on accurate...No, just analyzing what the numbers are. xnodx

[bluehenbillk]

If there is one thing that's as predictable as the sun will always set in the west, it's the self-proclaimed PR guy for this crappy index is all over this thread week after week playing spin doctor.

The best line I read was assuming the unthinkable, that this system was dead on accurate. I may send AGS my doctor's bill, I fell out of my chair and bruised my elbow in laughter from that comment.

[Syntax Error]

The best line I read was assuming the unthinkable, that this system was dead on accurate. I may send AGS my doctor's bill, I fell out of my chair and bruised my elbow in laughter from that comment.Must be a hearing aid or eyeglasses invoice because I did not read that here. xrolleyesx

Get this, when numbers come out they will be analyzed. The numbers are what they are, right or wrong. I would not spend time analyzing them if they were not pretty accurate in the past. Sorry if it bugs you. xcoffeex

thanks for finally chiming in!

[mcveyrl]

On the same token, ASU has not looked as stellar as I thought they would. Cruising ahead in the first half, and then slacking off is going to bite them in the arse eventually, like it did against JMU. Make a statement win at GSU this week, and you become my #2 (or #1, if JMU slips).

There are definitely cases to be made for both. :p

While I do fear a let down after the UR game...I'm confident we'll handle BYE in the 4th.

[89Hen]

I would not spend time analyzing them if they were not pretty accurate in the past.
The AGS Poll was more "accurate" than the GPI. Seems like a lot of extra work to make something less accurate. xpeacex

[Syntax Error]

The AGS Poll was more "accurate" than the GPI. Seems like a lot of extra work to make something less accurate. Hey, my job is data analyzing and what better way to do it by analyzing the GPI? The GPI has historically been the best indicator and includes the major polls and computer systems. The AGS Poll has been the best poll since it started (six years after the GPI started and it did better one time). So "less accurate" is bad terminology. xcoffeex

[89Hen]

The GPI has historically been the best indicator and includes the major polls and computer systems. The AGS Poll has been the best poll since it started (six years after the GPI started and it did better one time). So "less accurate" is bad terminology. xcoffeex
xnonox Since the AGS Poll started, it has been more accurate than the GPI. Yes, it was only better once, but that's one more than the GPI has been to the AGSP. xnodx Again, the GPI takes the AGS and makes it less accurate.

[Syntax Error]

The AGS Poll was more "accurate" than the GPI.xnonox ONCE.

No need to overstate it.

Bottom line is it is more fun for me to analyze a bunch of numbers. Sorry if it bugs you (notice the two people bashing are fans of the same team?). xcoffeex

[CSUBUCDAD]

Big south is moven up!!!!!!!!!!!!!!!

Good to see it too. Anybody surprised to see Chuck South rated above Liberty? I am.

[bluehenbillk]

89 you're dead on again. The AGS Poll has blown away the crappy index a few years running now, not one time as stated by the PR department. The crappy index has been off the mark on a regular basis for multiple years now, put the dog to bed, take it out back & just shoot it, thanks.

[OL FU]

If there is one thing that's as predictable as the sun will always set in the west, it's the self-proclaimed PR guy for this crappy index is all over this thread week after week playing spin doctor.

The best line I read was assuming the unthinkable, that this system was dead on accurate. I may send AGS my doctor's bill, I fell out of my chair and bruised my elbow in laughter from that comment.

No I have to take the blame for that. Syn did not say it was dead on accurate. I said that the only way most of the under rated or overrated numbers really mean anything is if you believe the GPI is dead on accurate. syn said he was simply analyzing the numbers as over or under based on the GPI which of course gets back to my point which is most of the analysis doesn't mean a lot unless you think the GPI is dead on accurate. xrotatehx xrotatehx

My head hurts.

[bluehenbillk]

unless you think the GPI is dead on accurate.


You ever see the circus where they see how many clowns will fit in a car?? Kinda like how many people would agree with that statement cramming into a broom closet.

[Syntax Error]

... The AGS Poll has blown away the crappy index a few years running now...No it hasn't. That is crazy talk. Oops, billk talking crazy about the GPI??? xlolx

Here you go 89Hen and billk, http://masseyratings.com/cf/compare1aa.htm

35 computers and polls rating FCS. Guess which has been at the very top EVERY single time? That is what a good index does. xcoffeex

[bluehenbillk]

No it hasn't. That is crazy talk. Oops, billk talking crazy about the GPI??? xlolx

Here you go 89Hen and billk, http://masseyratings.com/cf/compare1aa.htm

35 computers and polls rating FCS. Guess which has been at the very top EVERY single time? That is what a good index does. xcoffeex

Direct question: True or False, the AGS Poll did a better job of predicting at-large playoff teams the past two years than your crappy index.

Let's have a direct one word answer smart guy....

[OL FU]

Three things you can't avoid. Death, Taxes and Billk and Syntax arguing the GPIxsmiley_wix :D

[GannonFan]

Hey, wait a minute, we're also forgetting that I've outperformed the GPI ever since it's inception, and I've routed it in the past 2 years. Where's the love???? xlolx xlolx

Hey Hey
Ho Ho
This GPI's Gotta Go!!!!!!

xpeacex

[bluehenbillk]

Sorry OLFU, I gave them a pass last week, but when the PR machine is gonna throw blatant mistruths out there someone has to call it on the carpet. The crappy index has surely not been the pre-eminent playoff predictor or whatever the heck the verbiage is they throw out there. It's a conglomeration of compilated numbers they throw against the wall & try to make it stick in hopes that it would gain the credibility of the other three letter index that rules FBS that they just copied.

Does the real Roy Kramer get royalties from this thing? God forbid I refer to anyone on here as him, it'd just get me into trouble again.

I asked a simple question, let's see the reply....

[bluehenbillk]

Hey it's Gannon Fan, 89Hen & me, the gang's all here. The 3 of us should just pool our top 25 votes and create our own index & see how it stacks up to this crappy one. We'll send it out to college SID's & see if we can make it legitimate too, whatdya guys think? Maybe the PR department will answer my direct question directly....

[GannonFan]

Hey it's Gannon Fan, 89Hen & me, the gang's all here. The 3 of us should just pool our top 25 votes and create our own index & see how it stacks up to this crappy one. We'll send it out to college SID's & see if we can make it legitimate too, whatdya guys think?


Well, we're basically doing that with the AGS Poll anyway. You see, it takes that many more, sane people to vote in order to have enough sway to correct the many faulty computer models that get thrown into the mix and only serve to make the GPI less accurate and reliable than the AGS Poll. But hey, if they need us to step up even more so, then so be it!!!

xpeacex

[Syntax Error]

Direct question: True or False, the AGS Poll did a better job of predicting at-large playoff teams the past two years than your crappy index.
Let's have a direct one word answer smart guy....FALSE.

BTW, it is not my index, I was just analyzing it before you came in attacking me.

THE GPI ORIGINATED FROM A CAA FAN. xnodx

[bluehenbillk]

False? False? I feel like Jim Mora saying playoffs, but False?

The spin behind this crappy index has now achieved an all-time high. Will it stop at nothing to try to mask ineptness?

You know the AGS Poll has outperformed the crappy index in predicting more playoff teams. You think people just forget about it?? Believe me there's 3 people here that haven't forgot, that direct answer is direct "false"hood my friend.

[Syntax Error]

Yes, FALSE.

The AGS Poll has not done a better job of predicting at-large playoff teams the last two years than the GPI. It did just one year in the AGS Poll history. Sorry.

The fun is analyzing the GPI numbers, computers and polls (which includes the AGS Poll, the best poll around). Geez. xoopsx

[KAUMASS]

Yes, FALSE.

The AGS Poll has not done a better job of predicting at-large playoff teams the last two years than the GPI. It did just one year in the AGS Poll history. Sorry.

The fun is analyzing the GPI numbers, computers and polls (which includes the AGS Poll, the best poll around). Geez. xoopsx


I did some research on the GPI and the playoffs over the last ten years last year. I see if I can dig it up.

Update: Found it..

Here is a thread I started at playoff time for at large projections last year.

GPI and At Large Selections History since 2000 - 11-14-2007, 05:13 PM

--------------------------------------------------------------------------------

Here is the history since 2000 with eligible at large selections in the GPI.

The GPI has correctly picked: The committee took:
2000-8/8 of the at large participants.
2001-7/8-took Western Kentucky(19)(8-3) over Northwestern St(12)(8-3)
2002-7/8-took Northwestern St(16)(9-3) over Wofford(12)(9-3)
2003-7/8-took Bethune-Cookman(48)(9-2) over Lehigh(22)(8-3)
2004-7/8-took Lehigh(17)(9-2) over Cal Poly(12)(9-2)
2005-7/8-took Lafayette(41)(8-3) over Youngstown(12)(8-3)
2006-5/8-took Coastal Carolina(20)(9-2),Montana St(22)(7-4) and E. Illinois(26)(8-4) over Portland St(10)(7-4), San Diego(13)(10-0) and Wofford(19)(7-4).

Notes:

Since 2000, Montana St. is the only team to win a game that was ranked lower than a team in the GPI not taken as an at large participant(2006)

There was no committe bias as I looked up the members for those years, except maybe for regionalization.

Looks like there were some make up games given as well, Northwestern taken in 2002(make up for 2001 ). Lehigh in 2004(make up for 2003).

Teams owed in the future??: Portland St,(2006) Youngstown St(2005), Wofford(2002).
San Diego? Not sure with strengh of schedule, but they were ranked 13 in the GPI. Arguable either way.

Some I agree with, some I don't. Some higher ranked teams that did not ge in might have dropped their last two games while another may have won 7-8 in a row.(I didn't want to research, this took me long enough)

I thought you guys might like this info as it is interesting to see with regards to the upcoming at large playoff selections.


I'll have to look at 2007 numbers.

[boonegoon]

You are making one BIG assumption... that the computers are the ones that are right. xlolx xlolx xlolx

Massey Top 25
1. James Madison
2. Montana
3. Villanova
4. Weber St
5. Massachusetts
6. Elon
7. Wofford
8. Cal Poly
9. Richmond
10. Cent Arkansas
11. N Arizona
12. Furman
13. Appalachian St
14. William & Mary
15. Harvard
16. New Hampshire
17. Liberty
18. Cornell
19. Maine
20. Northern Iowa
21. W Illinois
22. Ga Southern
23. McNeese St
24. Northwestern St
25. So Illinois

I know that this is homerish but App at 13? Their 2 losses were to top teams on the road and even though some of their wins were lackluster they all were by double digits. I sense a conspiracy. Not really, I just hope that Massey is wrong.

[Syntax Error]

I know that this is homerish but App at 13?Don't worry, 89Hen did not post what the GPI uses from Massey. Check http://www.collegesportingnews.com/stats/writer/GPI/20081013gpi.html for the real GPI numbers. App St got 7 6 4 7 7 7 9 ratings from the computers. xcoffeex

[ngineer]

Computers don't see the games, either.

[bluehenbillk]

KAUMASS, thanks for the info proving my point over the last 2 years that the crappy index has produced equally crappy results. At least accuracy still counts for something nowadays. Some people will just try to re-write history thinking people won't go through the trouble of actually looking it up.

95 percent or more of the posters on here are more accurate than some crappy BCS wannabe system.

[Syntax Error]

Computers don't see the games, either.Yes, but they don't forget bad games like poll voters do. Good thing the GPI is the best index of polls and computers together. xthumbsupx

[Syntax Error]

xrolleyesx

[OL FU]

Sorry OLFU, ....

I just assumed you were busyxnodx xsmiley_wix

[89Hen]

xnonox ONCE.

No need to overstate it.

Bottom line is it is more fun for me to analyze a bunch of numbers. Sorry if it bugs you (notice the two people bashing are fans of the same team?). xcoffeex
Notice that the GPI routinely ranks the Hens higher than any of the Hen fans... your point? xconfusedx

I didn't overstate it in any way. The GPI takes the AGS Poll, combines it with two other polls and a bunch of computers and spits out a result that has been less accurate than the AGS Poll. That begs the question, why go through all that to be less accurate?

[89Hen]

Don't worry, 89Hen did not post what the GPI uses from Massey.

App St got 7 6 4 7 7 7 9 ratings from the computers. xcoffeex
6. NAU
7. AppSt

Good luck convincing anyone besides Jon on that.

[Syntax Error]

xlolx

[89Hen]

xlolx
Great reply. xthumbsupx

[SuperJon]

Good to see it too. Anybody surprised to see Chuck South rated above Liberty? I am.

CSU is fourth in the Big South. They're 69th while Liberty is 20th.

[IaaScribe]

Yeah, I was wondering what he was talking about there.

[KAUMASS]

Does anybody have a copy of the GPI from last year prior to the playoffs?

CSN archives are all mixed up and any old links are expired.

Thanks..

[Syntax Error]

Does anybody have a copy of the GPI from last year prior to the playoffs?

CSN archives are all mixed up and any old links are expired.

Thanks..Final 2007 GPI Playoff Indicator
Rank, Team, (Total)

Auto-Qualifiers:
1. Northern Iowa (1.00)
2. McNeese St (3.13)
5. Massachusetts (5.00)
8. Montana (7.50)
11. Wofford (11.63)
19. E Kentucky (19.50)
22. Delaware St (22.50)
43. Fordham (35.63)

At-Large:
4. S Illinois (4.88)
6. Appalachian St (5.38)
7. Richmond (6.13)
9. James Madison (9.50)
10. Delaware (10.25)
14. New Hampshire (15.75)
15. Villanova (17.50) <--- not selected
20. E Washington (19.75)

Bubble:
21. Hofstra (21.38)
30. Holy Cross (28.00)
35. S Carolina St (32.38)
36. Albany (32.50)
37. E Illinois (33.75) <--- selected

[WileECoyote06]

Why is the Ivy League rated higher than the Big South or the OVC when both have better OOC records?

Any explanation?

[GannonFan]

Notice that the GPI routinely ranks the Hens higher than any of the Hen fans... your point? xconfusedx

I didn't overstate it in any way. The GPI takes the AGS Poll, combines it with two other polls and a bunch of computers and spits out a result that has been less accurate than the AGS Poll. That begs the question, why go through all that to be less accurate?

Bingo!!! It's like buying a GPS device that's accurate to a certain distance, opening it up, tweaking with the insides of it, putting it back together and then using it, only to find that it's now less accurate, and then saying that you think it's better that way. Just sheer craziness. xpeacex

[danefan]

Why is the Ivy League rated higher than the Big South or the OVC when both have better OOC records?

Any explanation?

because they don't play anyone but PL and rarely CAA teams.

[WileECoyote06]

because they don't play anyone but PL and rarely CAA teams.

I noticed that last year. I consistently see them rated higher than some other conferences; but I think back to the Hampton-Princeton game last year. Hampton was the #4 team from the MEAC and handily beat the #4 team from the Ivy League.

[danefan]

I noticed that last year. I consistently see them rated higher than some other conferences; but I think back to the Hampton-Princeton game last year. Hampton was the #4 team from the MEAC and handily beat the #4 team from the Ivy League.

BTW, I read your question wrong.

I don't really know why they get rated higher. They shouldn't be. But people are free to vote for them in the top 25 without fear of reprise. They'll never really be tested.

[WileECoyote06]

BTW, I read your question wrong.

I don't really know why they get rated higher. They shouldn't be. But people are free to vote for them in the top 25 without fear of reprise. They'll never really be tested.

Voting individual schools is okay; but I don't see how the conference rates higher than certain other conferences.

[KAUMASS]

Final 2007 GPI Playoff Indicator
Rank, Team, (Total)

Auto-Qualifiers:
1. Northern Iowa (1.00)
2. McNeese St (3.13)
5. Massachusetts (5.00)
8. Montana (7.50)
11. Wofford (11.63)
19. E Kentucky (19.50)
22. Delaware St (22.50)
43. Fordham (35.63)

At-Large:
4. S Illinois (4.88)
6. Appalachian St (5.38)
7. Richmond (6.13)
9. James Madison (9.50)
10. Delaware (10.25)
14. New Hampshire (15.75)
15. Villanova (17.50) <--- not selected
20. E Washington (19.75)

Bubble:
21. Hofstra (21.38)
30. Holy Cross (28.00)
35. S Carolina St (32.38)
36. Albany (32.50)
37. E Illinois (33.75) <--- selected

Hey Syntax,

Thanks for the info. What I am trying to find out is did the GPI predict 6 out of the 8 at large according to rankings? I think NDSU was 13, (not eligible), 16-20 was G. Southern, Youngstown and I forget who else. But I remember that some of those teams from 16-20 only had 6 D1 wins or had 4-5 losses or something like that.
What stands out for me is that historically when the committee picking an at large team that has a lower ranking in the GPI-(E. Illinois in 2007), that team has never gotten past the first round.

Thanks,

KA

[Syntax Error]

Hey Syntax,
Thanks for the info. What I am trying to find out is did the GPI predict 6 out of the 8 at large according to rankings?7 out of 8. So your listing would be:

1998-(IIRC it was 8/8 of the at large participants)
1999-(IIRC it was 8/8 of the at large participants)
2000-8/8 of the at large participants.
2001-7/8-took Western Kentucky(19)(8-3) over Northwestern St(12)(8-3)
2002-7/8-took Northwestern St(16)(9-3) over Wofford(12)(9-3)
2003-7/8-took Bethune-Cookman(48)(9-2) over Lehigh(22)(8-3)
2004-7/8-took Lehigh(17)(9-2) over Cal Poly(12)(9-2)
2005-7/8-took Lafayette(41)(8-3) over Youngstown(12)(8-3)
2006-5/8-took Coastal Carolina(20)(9-2),Montana St(22)(7-4) and E. Illinois(26)(8-4) over Portland St(10)(7-4), San Diego(13)(10-0) and Wofford(19)(7-4).
2007-7/8-took E. Illinois(37) over Villanova(15)

Note: The team(s) not selected ("Woofed") usually make the playoffs the next year.

[IaaScribe]

Why is the Ivy League rated higher than the Big South or the OVC when both have better OOC records?

Any explanation?

Big South still plays a lot of DIIs:

Big South has:
Liberty vs. North Greenville, Glenville State
CSU vs. North Greenville, Edward Waters
VMI vs. Chowan
Gardner-Webb vs. Tusculum

I think SBU and Coastal were the only ones to avoid DIIs this year.

The OVC isn't as bad, but UTM and SEMO at least both have DIIs on the slate.

[WileECoyote06]

Big South still plays a lot of DIIs:

Big South has:
Liberty vs. North Greenville, Glenville State
CSU vs. North Greenville, Edward Waters
VMI vs. Chowan
Gardner-Webb vs. Tusculum

I think SBU and Coastal were the only ones to avoid DIIs this year.

The OVC isn't as bad, but UTM and SEMO at least both have DIIs on the slate.

That makes sense, only if everyone else is penalized for DII games. SoCon seems to play quite a few as well.

BTW, six of the Ivy League's 8 FCS OOC wins came against the same three teams.

[MacThor]

Final 2007 GPI Playoff Indicator
Rank, Team, (Total)

Auto-Qualifiers:
1. Northern Iowa (1.00)
2. McNeese St (3.13)
5. Massachusetts (5.00)
8. Montana (7.50)
11. Wofford (11.63)
19. E Kentucky (19.50)
22. Delaware St (22.50)
43. Fordham (35.63)

At-Large:
4. S Illinois (4.88)
6. Appalachian St (5.38)
7. Richmond (6.13)
9. James Madison (9.50)
10. Delaware (10.25)
14. New Hampshire (15.75)
15. Villanova (17.50) <--- not selected
20. E Washington (19.75)

Bubble:
21. Hofstra (21.38)
30. Holy Cross (28.00)
35. S Carolina St (32.38)
36. Albany (32.50)
37. E Illinois (33.75) <--- selected

Thanks for that! I hadn't realized that all four semi-finalists last year were at-large teams.

[lizrdgizrd]

Final 2007 GPI Playoff Indicator
Rank, Team, (Total)

Auto-Qualifiers:
1. Northern Iowa (1.00)
2. McNeese St (3.13)
5. Massachusetts (5.00)
8. Montana (7.50)
11. Wofford (11.63)
19. E Kentucky (19.50)
22. Delaware St (22.50)
43. Fordham (35.63)

At-Large:
4. S Illinois (4.88)
6. Appalachian St (5.38)
7. Richmond (6.13)
9. James Madison (9.50)
10. Delaware (10.25)
14. New Hampshire (15.75)
15. Villanova (17.50) <--- not selected
20. E Washington (19.75)

Bubble:
21. Hofstra (21.38)
30. Holy Cross (28.00)
35. S Carolina St (32.38)
36. Albany (32.50)
37. E Illinois (33.75) <--- selected

A few gaps in the numbers. Who were #3, 12, 13, 16-18?

[KAUMASS]

7 out of 8. So your listing would be:

1998-(IIRC it was 8/8 of the at large participants)
1999-(IIRC it was 8/8 of the at large participants)
2000-8/8 of the at large participants.
2001-7/8-took Western Kentucky(19)(8-3) over Northwestern St(12)(8-3)
2002-7/8-took Northwestern St(16)(9-3) over Wofford(12)(9-3)
2003-7/8-took Bethune-Cookman(48)(9-2) over Lehigh(22)(8-3)
2004-7/8-took Lehigh(17)(9-2) over Cal Poly(12)(9-2)
2005-7/8-took Lafayette(41)(8-3) over Youngstown(12)(8-3)
2006-5/8-took Coastal Carolina(20)(9-2),Montana St(22)(7-4) and E. Illinois(26)(8-4) over Portland St(10)(7-4), San Diego(13)(10-0) and Wofford(19)(7-4).
2007-7/8-took E. Illinois(37) over Villanova(15)

Note: The team(s) not selected ("Woofed") usually make the playoffs the next year.

Thanks for the update. So folks, here you have it. Since 1998, the GPI has predicted 88% accuracy of the at large spots. Bottom line is that by the end of the season, the GPI is pretty accurate. Right now, it looks like it's got some misfires, but it should get better by the end of the season.

When the committee picks a team that is not next in line in the GPI that is available to be in the playoffs, that team has never gotton past the first round. Can someone bring this up to the committee of the NCAA?

Now, you may also disagree with the way the GPI is formulated, which is a whole different story!!!

[MacThor]

A few gaps in the numbers. Who were #3, 12, 13, 16-18?

I can't find the archive but I believe those teams were playoff-ineligible. #3 might have been transitional South Dakota St?

[89Hen]

So folks, here you have it. Since 1998, the GPI has predicted 88% accuracy of the at large spots. Bottom line is that by the end of the season, the GPI is pretty accurate.
It is. But the AGS Poll (one component of the GPI) has a higher percentage. xnodx xthumbsupx

[lizrdgizrd]

I can't find the archive but I believe those teams were playoff-ineligible. #3 might have been transitional South Dakota St?
I found it! Thanks NDSU Bison archives! xthumbsupx


Gridiron Power Index Top 25 (Nov. 20, 2007)
1. Northern Iowa (1.00)
2. McNeese St (3.13)
3. N Dakota St (4.75)
4. S Illinois (4.88)
5. Massachusetts (5.00)
6. Appalachian St (5.38)
7. Richmond (6.13)
8. Montana (7.50)
9. James Madison (9.50)
10. Delaware (10.25)
11. Wofford (11.63)
12. Youngstown St (14.75)
13. S Dakota St (15.13)
14. New Hampshire (15.75)
15. Villanova (17.50)
16T. Elon (18.75)
16T. Ga Southern (18.75)
18. Yale (19.25)
19. E Kentucky (19.50)
20. E Washington (19.75)
21. Hofstra (21.38)
22. Delaware St (22.50)
23. Grambling (22.63)
24. The Citadel (22.75)
25. W Illinois (22.88)


A few gaps in the numbers. Who were #3, 12, 13, 16-18?
3. NDSU - transition
12. YSU - ???
13. SDSU - transition
16. Elon - only 6 DI wins
17. GSU - only 6 DI wins
18. Yale - too good for playoffs xrolleyesx

[GannonFan]

It is. But the AGS Poll (one component of the GPI) has a higher percentage. xnodx xthumbsupx

Yup. Just today, I had to add 10 plus 10. Now, I first got 20 when I did that, but then I thought, for a hoot, that I'd whip up a fancy computer model to simulate counting and, well, I know I need to tweak it a bit more (I can tweak in season, right, computer models seem to do that all the time when they don't think they're getting the right answer) but the computer model came out with 15 as the answer. I figure I'll just index the computer's answer with my answer and now I've come to the conclusion that 10 plus 10 is somewhere close to 17.5. Aw, good thing I took the right answer and indexed it over a model that may or may not give me the right answer. It's not any simpler (or any more correct) but gosh, it's fun to look at the numbers. xpeacex

[lizrdgizrd]

Any Given Saturday Poll Top 25
11/19/2007
(First place votes in parenthesis)
1. Northern Iowa (55) 11-0
2. McNeese St. (1) 11-0
3. Montana (3) 11-0
4. Southern Illinois 10-1
5. Appalachian St. 9-2
6. North Dakota St. 10-1 (not eligible)
7. Richmond 9-2
8. Massachusetts 9-2
9. Delaware St. 10-1
10. Eastern Kentucky 9-2
11. Wofford 8-3
12. James Madison 8-3
13. Delaware 8-3
14. Eastern Washington 8-3
15. Eastern Illinois 8-3
16. Youngstown St. 7-4
17. New Hampshire 7-4
18. South Dakota St. 7-4
19. Grambling St. 8-2
20. Yale 9-1
21. Georgia Southern 7-4
22. Cal Poly 7-4
23. Villanova 7-4
24. Harvard 8-2
25. Fordham 8-3

So AGS Poll was 8/8 on at-large playoff teams. xchinscratchx

[KAUMASS]

Anybody feel like doing some research and compare the AGS poll the last 10 years (if that many done) to at large selections?

[RabidRabbit]

I found it! Thanks NDSU Bison archives! xthumbsupx




3. NDSU - transition
12. YSU - ???
13. SDSU - transition
16. Elon - only 6 DI wins
17. GSU - only 6 DI wins
18. Yale - too good for playoffs xrolleyesx

YSU, I believe also was 7-4 with 1 D-II win, ie only 6 DI wins. Finished 3rd in 7 team Gateway.

[lizrdgizrd]

1. Appalachian St (41) - auto
2. Montana (2) - auto
3. Massachusetts (3) - auto
4. North Dakota St (2) - transition
5. Youngstown St - auto
6. James Madison
7. Furman
8. New Hampshire
9. Illinois St
10. Southern Illinois
11. Hampton - auto
12. Tenn Martin - auto
13. Coastal Carolina
14. Princeton - DNP
15. Northern Iowa
16. Eastern Illinois
17. Cal Poly
18. Portland St
19. Montana St
20. San Diego
21. South Dakota St
22. Monmouth
23. McNeese St - auto
24. Yale
25. Wofford

AGS Poll = 7/8 for at-large in 2006.

[KAUMASS]

AGS Poll = 7/8 for at-large in 2006.

If the numbers are similar the last few years or so, maybe we should be the selection committee or just have them use a combo of the GPI and AGS!! pollxthumbsupx xnodx

[lizrdgizrd]

AGS Poll (11/21/2005)

Rank, School, First Place Votes, Movement from Last Poll
1. New Hampshire (34) NC - auto
2. Appalachian State (2) +1 - auto
3. Furman +2
4. Hampton (2) NC - auto
5. Texas State +1
6. Northern Iowa +3 - auto
7. Georgia Southern +1
8. Montana -6
9. Southern Illinois +4
10. Cal Poly NC
11. Richmond +3
12. Eastern Washington +5 - auto
13. Youngstown State +2
14. Eastern Illinois +5 - auto
15. Brown +3
16. Montana State +6
17. Nicholls State +6 - auto
18. Grambling +2
19. Massachusetts -12
20. South Carolina State +1
21. Illinois State +3
22. Colgate NR
23. Lehigh -12
24. Lafayette NR
25. North Dakota State -9

So AGS Poll 7/8 in 2005.