PDA

View Full Version : Inaugural 2010 GPI Released: 10/12/2010



Lehigh Football Nation
October 13th, 2010, 08:42 AM
Didn' see this posted anywhere yet.

http://www.collegesportingnews.com/content.php?261-Gridiron-Power-Index-Released


CHICAGO — The Gridiron Power Index (GPI), the index ranking for the NCAA Division I FCS and a top indicator of at-large playoff selection begins its inaugural rankings for the 2010 season with Delaware of the Colonial Athletic Association at the No. 1 spot.

The Colonial Athletic Association, the largest league in the FCS has five teams in the top 25 (and all five happen to also be in the Top Ten); the Big Sky Conference and Southern Conferences also have five Top 25 teams; and the Missouri Valley Football has four teams. Two Ohio Valley Conference teams currently reside in the Top 25, as do the Southland Conference has two; and the Big South, Great West, Southland, and Mid-Eastern Athletic conferences have one each.

Delaware is in the No. 1 spot. The Patriot League, well... xwhistlex xnonono2x

theasushow
October 13th, 2010, 08:46 AM
Didn' see this posted anywhere yet.

http://www.collegesportingnews.com/content.php?261-Gridiron-Power-Index-Released



Delaware is in the No. 1 spot. The Patriot League, well... xwhistlex xnonono2x


This is the first ranking we see that really factors in strength of schedule....wow look at South Carolina State, barely even cracked the top 25, and Jacksonville State is also effected.

PhoenixSupreme
October 13th, 2010, 09:03 AM
This is the first ranking we see that really factors in strength of schedule....wow look at South Carolina State, barely even cracked the top 25, and Jacksonville State is also effected.

It really gives Western Illinois and Sacramento St. a huge boost as well because of that

darell1976
October 13th, 2010, 09:20 AM
#50!!!! I didn't know there was 124 FCS teams.

Gil Dobie
October 13th, 2010, 09:27 AM
10/11/2010 GPI Top 25

1. Delaware (1.25)
2. Appalachian St (3.00)
3. James Madison (4.625)
4. Stephen F. Austin (4.75)
5. William & Mary (5.75)
T6. Montana St (6.625)
T6. Jacksonville St (6.625)
8. Massachusetts (7.125)
9. Eastern Washington (10.25)
10. Villanova (11.25)
T11. Wofford (13.625)
T11. Western Illinois (13.625)
13. Montana (13.75)
14. North Dakota State (14.500)
15. Cal Poly SLO (15.25)
16. Georgia Southern (19.625)
17. Northern Arizona (20.25)
18. Northern Iowa (20.375)
19. Southern Illinois (20.625)
20. Sacramento State (20.75)
T21. Southeast Missouri State (21.75)
T21. Chattanooga (21.75)
T23. Furman (23.625)
T23. South Carolina State (23.625)
25. Liberty (23.75)

RabidRabbit
October 13th, 2010, 09:27 AM
And a 1-4 SDSU is in at 28th! Apparently Jacks did face as a tough a start as record indicates.

Sycamore51
October 13th, 2010, 09:34 AM
Indiana St. is 35th, and we've been blown out by WIU, beat by 33 by an average at best Cinci team, played 2 Non-D1 teams. They must count ppg or something pretty heavy in this.

danefan
October 13th, 2010, 09:42 AM
Deserves the obligatory..........xwhistlex

Will the modified GPI ever be made public?

Dane96
October 13th, 2010, 09:48 AM
Hold on...I am in the bathroom. Anyone have TP available? Can I borrow the paper the GPI is printed on?

WMTribe90
October 13th, 2010, 09:49 AM
I think this looks better than the TSN poll. GPI has Chatt and Furman in the top 25 over schools like UR, Elon and UNH, which I agree with.

SC St should be higher than 25, but it is telling that their SOS is so poor. This is why I have a hard time puting any MEAC team in the top 10 until the MEAC wins some playoff games.

Dane96
October 13th, 2010, 09:53 AM
Only prob with your statement is this is an index...an index that INCLUDES the TSN poll to come up with the final numbers.

Thus, it is inherently flawed on that basis alone, but there are a number of reasons why the GPI is nonsense.

WMTribe90
October 13th, 2010, 11:17 AM
A final product can be better than the sum of its parts. The GPI can be and is better IMO than the TSN poll. Realize people love it or hate it and don't want to rehash the whole debate. To me, its just a tool and one of many things to consider. I would never want playoff selections to be based off this type of BCS style model, but on occassion it does illuminate disparities in some of the straight human poles (see my original post).

OL FU
October 13th, 2010, 11:41 AM
Just curious, anybody notice where Elon and Harvard are rankedxconfusedx















xwhistlex

danefan
October 13th, 2010, 12:53 PM
Just curious, anybody notice where Elon and Harvard are rankedxconfusedx

xwhistlex

How about Penn?

Just really proves what we've been saying about the ciruclar effect the PL/IVY scheduling arrangement has had on the rankings and polls for the last few years. They prop each other up without any reference to the outside world. Unfortunately for them its come back to bite them in arse this year. At least the Ivy. The play in the PL has certainly dropped off, but I'm not sure the Ivy has. The Ivy is just being destroyed by the fact that they don't beat anyone OOC except the PL teams.

WMTribe90
October 13th, 2010, 01:58 PM
How about Penn?

Just really proves what we've been saying about the ciruclar effect the PL/IVY scheduling arrangement has had on the rankings and polls for the last few years. They prop each other up without any reference to the outside world. Unfortunately for them its come back to bite them in arse this year. At least the Ivy. The play in the PL has certainly dropped off, but I'm not sure the Ivy has. The Ivy is just being destroyed by the fact that they don't beat anyone OOC except the PL teams.

Certainly agree with this. PL is down considerably from the late 90's and first part of this decade. Ivy is holding steady or improving somewhat and has surpassed thePL, but there's no way to confirm the relative strength of the Ivy League becuase, with few exceptions (Penn v Villanova), they don't play any barometer games against the power conferences.

houtexan
October 13th, 2010, 02:18 PM
Just curious, anybody notice where Elon and Harvard are rankedxconfusedxlex

48. Elon
59. Harvard

OL FU
October 13th, 2010, 02:51 PM
48. Elon
59. Harvard

I was irritating an old buddy. and I am sure he is tired of it by now:o:D

blukeys
October 13th, 2010, 02:54 PM
And a 1-4 SDSU is in at 28th! Apparently Jacks did face as a tough a start as record indicates.

So tell us who is tougher, Delaware or Nebraska?????

Squealofthepig
October 13th, 2010, 03:55 PM
Here's another interesting way to look at this, comparing the GPI rank with the AGS vote for this week:

1. Delaware (1.25) -
2. Appalachian St (3.00) -
3. James Madison (4.625) +5
4. Stephen F. Austin (4.75) +2
5. William & Mary (5.75) -
T6. Montana St (6.625) +3
T6. Jacksonville St (6.625) -3
8. Massachusetts (7.125) -4
9. Eastern Washington (10.25) +2
10. Villanova (11.25) -3
T11. Wofford (13.625) +3
T11. Western Illinois (13.625 +11)
13. Montana (13.75) -1
14. North Dakota State (14.500) +1
15. Cal Poly SLO (15.25) -2
16. Georgia Southern (19.625) +4
17. Northern Arizona (20.25) unranked
18. Northern Iowa (20.375) +5
19. Southern Illinois (20.625) +1
20. Sacramento State (20.75) unranked
T21. Southeast Missouri State (21.75) -3
T21. Chattanooga (21.75) unranked
T23. Furman (23.625) -2
T23. South Carolina State (23.625) -13
25. Liberty (23.75) unranked

UAalum72
October 13th, 2010, 04:19 PM
A final product can be better than the sum of its parts. The GPI can be and is better IMO than the TSN poll.But this is not the sum of all parts, it's the average of all parts. The GPI is probably better than TSN, but the presence of TSN make the GPI worse than a (GP minus TSN) index.

slostang
October 13th, 2010, 04:42 PM
I really have to questioned how Montana is ranked higher than Cal Poly. Cal Poly beat them in head to head and both have 4-2 records. Montana does not have a win against a team with a winning record. Sac State is their best win and they are 3-3. Montana wins: 0-6 DII Western State, Sac State, Idaho State and Northern Colorado. Cal Poly's wins: 5-1 DII Humbolt State, Montana, McNeese State and Old Dominion. Cal Poly's losses: Texas State and Fresno State. Montana's losses: Cal Poly and Eastern Washington. It does not add up to me.

RabidRabbit
October 13th, 2010, 04:53 PM
Sagarin's (which is component of the GPI) uses the components of your opponent's opponents as a weight factor also. SDSU is so high because in addition to Delaware (who beat a FCS school that beat a top 25 BCS team), we also have Nebraska, and UNI (SFA/NDSU/IA St) as big counters.

Cal Poly's ranking should go up playing two FBS killers (USD & UC-D). Not certain if Montana can increase due competitors' competitors from here on out, where as Cal Poly's future games include several 2 FBS playing FCS teams.

jmufan999
October 13th, 2010, 10:11 PM
Here's another interesting way to look at this, comparing the GPI rank with the AGS vote for this week:

1. Delaware (1.25) -
2. Appalachian St (3.00) -
3. James Madison (4.625) +5
4. Stephen F. Austin (4.75) +2
5. William & Mary (5.75) -
T6. Montana St (6.625) +3
T6. Jacksonville St (6.625) -3
8. Massachusetts (7.125) -4
9. Eastern Washington (10.25) +2
10. Villanova (11.25) -3
T11. Wofford (13.625) +3
T11. Western Illinois (13.625 +11)
13. Montana (13.75) -1
14. North Dakota State (14.500) +1
15. Cal Poly SLO (15.25) -2
16. Georgia Southern (19.625) +4
17. Northern Arizona (20.25) unranked
18. Northern Iowa (20.375) +5
19. Southern Illinois (20.625) +1
20. Sacramento State (20.75) unranked
T21. Southeast Missouri State (21.75) -3
T21. Chattanooga (21.75) unranked
T23. Furman (23.625) -2
T23. South Carolina State (23.625) -13
25. Liberty (23.75) unranked

that's a really interesting way to look at it. i guess i don't really have anything to add, just thought it was interesting.

WileECoyote06
October 14th, 2010, 06:53 AM
I think this looks better than the TSN poll. GPI has Chatt and Furman in the top 25 over schools like UR, Elon and UNH, which I agree with.

SC St should be higher than 25, but it is telling that their SOS is so poor. This is why I have a hard time puting any MEAC team in the top 10 until the MEAC wins some playoff games.

Nothing wrong with that philosophy, SCSU has yet to play the other teams with winning records in their conference (Hampton, Bethune-Cookman). BCC will also receive a spike from playing SCSU. It also doesn't help that their conference currently has three teams with 1 win or less (DSU, A & T, Howard) or that FAMU lost to Tenn. State.

danefan
October 14th, 2010, 06:53 AM
that's a really interesting way to look at it. i guess i don't really have anything to add, just thought it was interesting.

But is it really comparing the GPI considering the AGS poll is a component of the GPI.

I guess you can look at it and say the GPI does nothing but make the AGS poll worse (or better if your a Western Illinois fan I guess).

Take a look at Massey's Composite ratings. http://www.masseyratings.com/cf/compare1aa.htm

Some of the same computer rankings are used in the GPI. There are some real messed up results. For example, there is a computer ranking on the Massey Composite that has Penn as #5.

Cocky
October 14th, 2010, 07:14 AM
Computers only compute what a human puts in. So these polls are just opinions like the others.

Lehigh Football Nation
October 14th, 2010, 09:17 AM
Some of the same computer rankings are used in the GPI. There are some real messed up results. For example, there is a computer ranking on the Massey Composite that has Penn as #5.

Glad to see you're staying on message, at least, no matter how misguided it is. xlolx

Penn has one loss - to the No. 1 team in the country at the time. Explain to me how having them ranked highly is incorrect. If anything, all the other polls (human, computer and indexes, including the GPI) have massively underrated them. While I don't have Penn as No. 5, they have been in my Top 25 every week.

Guess it's as "messed up" as having one computer poll have Robert Morris at #24. Wow, how "messed up"! xlolx

danefan
October 14th, 2010, 09:21 AM
Glad to see you're staying on message, at least, no matter how misguided it is. xlolx

Penn has one loss - to the No. 1 team in the country at the time. Explain to me how having them ranked highly is incorrect. If anything, all the other polls (human, computer and indexes, including the GPI) have massively underrated them. While I don't have Penn as No. 5, they have been in my Top 25 every week.

Guess it's as "messed up" as having one computer poll have Robert Morris at #24. Wow, how "messed up"! xlolx

If you can't see how Penn at #5 is messed up then you have absolutely no clue about college football. Can you honestly say that Penn is in any stretch of the imagination a top 5 team in the country?

RMU at 24 sounds pretty accurate to me actually. Anywhere from 35-20 would be accurate for RMU at this point. Same for Penn in my mind.

NHwildEcat
October 14th, 2010, 10:27 AM
A win this weekend would move UNH back into the top 25...which the more I read on this weekend's game seems more and more posisble with every passing minute.

UNH Fanboi
October 14th, 2010, 11:18 AM
A win this weekend would move UNH back into the top 25...which the more I read on this weekend's game seems more and more posisble with every passing minute.

Why is that? I've been cautiously optimistic about this game because of JMU's offensive struggles, but I haven't seen anything to make me more optimistic. We've still got some injury issues and it doesn't look like Jeannot will be playing.

bluehenbillk
October 14th, 2010, 12:34 PM
Hold on...I am in the bathroom. Anyone have TP available? Can I borrow the paper the GPI is printed on?

Ditto, my sentiments exactly, and my team is ranked #1.

WMTribe90
October 14th, 2010, 01:02 PM
But is it really comparing the GPI considering the AGS poll is a component of the GPI.

I guess you can look at it and say the GPI does nothing but make the AGS poll worse (or better if your a Western Illinois fan I guess).

Take a look at Massey's Composite ratings. http://www.masseyratings.com/cf/compare1aa.htm

Some of the same computer rankings are used in the GPI. There are some real messed up results. For example, there is a computer ranking on the Massey Composite that has Penn as #5.


Correct me if I'm wrong, but doesn't the GPI look at 7 computer rankings and toss out the high and low ranking for each team and averages the ranking from the remaing 5 polls. So, in your example, the #5 ranking of Penn in Massey was tossed as an outlier. Penn does not appear in the top 25 of the GPI (maybe they should) and looking at the GPI top 25 I don't see anything blatantly out of whack.

For political junkies, its no different than averaging poll results for any given election contest to more accurately predict the outcome. No new data was added, but again, a system of averaging the results and throwing out the outliers often leads to a more representative (true) result.

danefan
October 14th, 2010, 01:11 PM
Correct me if I'm wrong, but doesn't the GPI look at 7 computer rankings and toss out the high and low ranking for each team and averages the ranking from the remaing 5 polls. So, in your example, the #5 ranking of Penn in Massey was tossed as an outlier. Penn does not appear in the top 25 of the GPI (maybe they should) and looking at the GPI top 25 I don't see anything blatantly out of whack.

For political junkies, its no different than averaging poll results for any given election contest to more accurately predict the outcome. No new data was added, but again, a system of averaging the results and throwing out the outliers often leads to a more representative (true) result.

I think it does throw out the highs and lows for that particular team, which in theory sounds great. In practice however, if a ranking spits out Penn at #5, the entire foundation for that ranking is flawed. So while the GPI may throw out the highs and lows from certain polls for certain teams, the rankings for the rest of the teams derived from that inherently flawed system are used.

Plus I just don't think a computer can accurately value a team's strength. I'd much prefer any human poll that is correctly audited (read anyone but the TSN poll) before I'd give any computer any credance.

appfan2008
October 14th, 2010, 01:19 PM
Sagarin's (which is component of the GPI) uses the components of your opponent's opponents as a weight factor also. SDSU is so high because in addition to Delaware (who beat a FCS school that beat a top 25 BCS team), we also have Nebraska, and UNI (SFA/NDSU/IA St) as big counters.

Cal Poly's ranking should go up playing two FBS killers (USD & UC-D). Not certain if Montana can increase due competitors' competitors from here on out, where as Cal Poly's future games include several 2 FBS playing FCS teams.

using that logic... asu will only be helped by playing the top few teams in our conference and then florida between now and the playoffs...

NHwildEcat
October 14th, 2010, 01:58 PM
Why is that? I've been cautiously optimistic about this game because of JMU's offensive struggles, but I haven't seen anything to make me more optimistic. We've still got some injury issues and it doesn't look like Jeannot will be playing.

I know our offense has been struggling...but I just think it will come down to who can do the most with the least (offensively) and I am confident that our offense can grind it out with dink and dunk passes and beat these guys. Obviously it can go either way, but I just have this feeling. I don't think it is all doom and gloom for us...not yet at least.

bluehenbillk
October 14th, 2010, 02:33 PM
There are only 34 FBS schools better than Delaware. If that isn't enough for you, Delaware is better than Ohio State, take that YSU fans!!
...
...
Well, of course my source is 2 of the computer rankings that comprise the BCS. Even if you wanna throw one of those out for being wacky, you have to keep the other one. Coming up with a stupid-*** system is stupid. Copying a stupid-*** system well, what the hell do you call that????

Lehigh Football Nation
October 14th, 2010, 03:33 PM
I think it does throw out the highs and lows for that particular team, which in theory sounds great. In practice however, if a ranking spits out Penn at #5, the entire foundation for that ranking is flawed.

But is a ranking flawed when it spits out Robert Morris at #24, when almost all of the rest of them put them in the #44-#54 range? After all, the NEC champ is no different than the Ivy League champ in this case. By your definition, Sag should be thrown out... xconfusedx

Go...gate
October 14th, 2010, 03:37 PM
However they came up with the numbers, i am afraid they are correct with regard to the Patriot League.

danefan
October 14th, 2010, 04:38 PM
But is a ranking flawed when it spits out Robert Morris at #24, when almost all of the rest of them put them in the #44-#54 range? After all, the NEC champ is no different than the Ivy League champ in this case. By your definition, Sag should be thrown out... xconfusedx

I think you are trying to make some kind of point, but I'm not really getting what you are saying.

Take a look at Massey's comparison site, there is a ranking (Nolan Power Index) in there with RMU at 6. That computer system is clearly flawed. Look at it.
http://warrennolan.com/football1aa/2010/npi

There's a big difference between a 10 spot spread and a 30-50 point spread btw.

Keep reaching LFN and you're going to pull a hammy.

Tod
October 14th, 2010, 07:00 PM
I really have to questioned how Montana is ranked higher than Cal Poly. Cal Poly beat them in head to head and both have 4-2 records. Montana does not have a win against a team with a winning record. Sac State is their best win and they are 3-3. Montana wins: 0-6 DII Western State, Sac State, Idaho State and Northern Colorado. Cal Poly's wins: 5-1 DII Humbolt State, Montana, McNeese State and Old Dominion. Cal Poly's losses: Texas State and Fresno State. Montana's losses: Cal Poly and Eastern Washington. It does not add up to me.

I agree 100%, slo.

Nebuta
October 15th, 2010, 11:30 AM
It will all work out eventually by the end of the season. With more games and a larger sample size this GPI will look more reasonable. Right now its pretty pointless. (You could say the same thing about the BCS formula.) No need to sweat the small stuff. )