2013 Gridiron Power Index (GPI) Opens With NDSU On Top Of The FCS

[CSN-info]

2013 Gridiron Power Index (GPI) Opens With NDSU On Top Of The FCS (http://www.collegesportingnews.com/content.php?965-2013-Gridiron-Power-Index-%28GPI%29-Opens-With-NDSU-at-1)

http://www.collegesportingnews.com/attachment.php?attachmentid=1117&cid=18&thumb=1&stc=1 (http://www.collegesportingnews.com/content.php?965-2013-Gridiron-Power-Index-%28GPI%29-Opens-With-NDSU-at-1)
COLLEGE SPORTING NEWS

North Dakota State enters the 2013 campaign on top of the Gridiron Power Index (GPI), the index ranking for the NCAA Division I Football Championship Subdivision and a top indicator of at-large playoff selection. The Bison rule the range again after leading the entire 2012 regular season and winning its second straight title. Conference mate Northern Iowa is solid in second.

10/1/2013 GPI TOP 25

Rank School (No. of #1 Ratings) (Rating)
1. N Dakota St (6) (1.00)
2. Northern Iowa (1) (2.57)
3. Towson (3.29)
4. E Illinois (4.00)
5. Sam Houston St (5.57)
6. Coastal Carolina (9.71)
7. E Washington (10.71)
8. Maine (12.57)
9T. Bethune-Cookman (13.86)
9T. Fordham (13.86)
9T. McNeese St (13.86)
12. Villanova (14.00)
13. S Dakota St (16.00)
14. Northern Arizona (16.71)
15. Southern Utah (17.71)
16. TN Martin (18.14)
17. Cent Arkansas (18.71)
18. William & Mary (19.29)
19. Ga Southern (20.14)
20. Youngstown St (20.29)
21. Tennessee St (22.29)
22. Old Dominion (23.57)
23. SE Louisiana (24.14)
24. Montana St (24.57)
25T. Cal Poly (25.43)
25T. Wofford (25.43)

CONFERENCE RANKING:

Rank, League, Total Average
1. Missouri Valley Football Conference (29.16)
2. Southland Conference (31.52)
3. Colonial Athletic Association (34.30)
4. Ohio Valley Conference (36.19)
5. Big Sky Conference (39.60)
6. Southern Conference (45.63)
7. Big South Conference (47.86)
8. Ivy League (54.38)
9. Independents (60.69)
10. Mid-Eastern Athletic Conference (61.30)
11. Patriot League (63.53)
12. Northeast Conference (70.29)
13. Southwestern Athletic Conference (78.73)
14. Pioneer Football League (83.92)



Read More (http://www.collegesportingnews.com/content.php?965-2013-Gridiron-Power-Index-%28GPI%29-Opens-With-NDSU-at-1)

[dystopiamembrane]

Looks good. Included are Georgia Southern and Old Dominion. Removing them brings Harvard in at #25.

[HensRock]

This looks pretty similar to my AGS Top 25 ballot.
Notice everyone:

No Gardner-Webb
No Charleston Southern
No Lehigh
No Delaware
No JMU
No UNH

[URMite]

Isn't there usually a link to the detail? Showing each computer rating?

And I still hope to see the SRS ratings in the next week...

[dystopiamembrane]

This looks pretty similar to my AGS Top 25 ballot.
Notice everyone:

No Gardner-Webb
No Charleston Southern
No Lehigh
No Delaware
No JMU
No UNH
And, No Griz

[dystopiamembrane]

Isn't there usually a link to the detail? Showing each computer rating?
The link is at the top of the post.

[Grizalltheway]

Complete joke, as expected.

[ccd494]

Isn't ODU above 63 equivalencies? Why are they included?

[HensRock]

And, No Griz

Yeah, Well the Griz' did make mine; albeit at #24.

[HensRock]

Apparently, the Hens have now joined the MEAC:


10. Mid-Eastern Athletic Conference (61.30)
9T. Bethune-Cookman (13.86)
36. S Carolina St (32.71)
52. NC Central (44.29)
53. NC A&T (45.57)
82. Delaware (66.71)
83. Florida A&M (66.86)
92. Norfolk St (71.86)
103. Howard (79.43)
106. Hampton (80.71)
107. Savannah St (83.43)
117. Morgan St (88.86)

[tingly]

Complete joke, as expected. It's too early for GPI or any rating/poll to be accurate. They get better as the season goes along.

[Lehigh Football Nation]

Is it responsible to release an index without any link to the data that makes up the spreadsheet? I sure don't think so.

[tingly]

GPI is a snap to derive if you're up for it.

EDIT: What I meant to say was http://www.collegesportingnews.com/stats/writer/gpi/20130930GPI.htm

[lionsrking2]

Finally, rankings that make some sense!

[Lehigh Football Nation]

Computer rankings:
MAS = Massey (http://www.masseyratings.com/rate.php?lg=cf), ASH/ABC = Ashburn (http://www.atomicfootball.com/), SAG = Sagarin (http://usatoday30.usatoday.com/sports/sagarin/fbt12.htm), LAZ = Laz Index (http://www.lazindex.com/), KEE = Keeper (http://home.comcast.net/~keepersfootball/site/), SEL = Self (https://github.com/jeffself/collegefootballrankings), BRN = Born (http://www.bornpowerindex.com/).
Polls:
FCP = FCS Coaches (http://www.collegesportingnews.com/content.php?938-2013-FCS-Coaches-Poll), SNW = Sports Network (http://www.collegesportingnews.com/content.php?940-2013-Sports-Network-Poll)

So, 5 computers and 2 human polls. No use of the SRS that the NCAA is having computed for them. No AGS poll. And human polls don't even comprise 1/3 of the total inputs.

Carry on.

[WestCoastAggie]

Computer rankings:
MAS = Massey (http://www.masseyratings.com/rate.php?lg=cf), ASH/ABC = Ashburn (http://www.atomicfootball.com/), SAG = Sagarin (http://usatoday30.usatoday.com/sports/sagarin/fbt12.htm), LAZ = Laz Index (http://www.lazindex.com/), KEE = Keeper (http://home.comcast.net/~keepersfootball/site/), SEL = Self (https://github.com/jeffself/collegefootballrankings), BRN = Born (http://www.bornpowerindex.com/).
Polls:
FCP = FCS Coaches (http://www.collegesportingnews.com/content.php?938-2013-FCS-Coaches-Poll), SNW = Sports Network (http://www.collegesportingnews.com/content.php?940-2013-Sports-Network-Poll)

So, 5 computers and 2 human polls. No use of the SRS that the NCAA is having computed for them. No AGS poll. And human polls don't even comprise 1/3 of the total inputs.

Carry on.


Any access to the SRS?

[Wallace]

Any access to the SRS?

Of course not, and the Lehigh fanatic can't even get % of total inputs right. No wonder he is aghast that Lehigh was unqualified last year for the playoffs. Some people are just not football folk.

Here's the reason he's all snitty:

11. Patriot League (63.53)
9T. Fordham (13.86)
41. Lehigh (36.00)
74. Holy Cross (60.29)
95T. Lafayette (74.43)
105. Bucknell (80.43)
112. Colgate (87.00)
121. Georgetown (92.71)

[Lehigh Football Nation]

Of course not, and the Lehigh fanatic can't even get % of total inputs right. No wonder he is aghast that Lehigh was unqualified last year for the playoffs. Some people are just not football folk.

Here's the reason he's all snitty:

11. Patriot League (63.53)
9T. Fordham (13.86)
41. Lehigh (36.00)
74. Holy Cross (60.29)
95T. Lafayette (74.43)
105. Bucknell (80.43)
112. Colgate (87.00)
121. Georgetown (92.71)

Potshot much? xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx xlolx

Total = (ARC + 2 polls used)/7
ARC (Adjusted Rankings of the Computers) = Sum of 5 computer ranking numbers (removing the max and min number from the 7 used)

5 computers, 2 polls, "highest" and "lowest" computer rankings removed. Still more than 2/3 weighted towards computer output.

Remember, kids, this is from the actual spreadsheet that lists all the teams:

http://www.collegesportingnews.com/stats/writer/gpi/20130930GPI.htm

Just reporting the facts.

[Wallace]

Potshot much?... Just reporting the facts.

Repost the facts sir, in their entirety without your clueless remarks.xrulesx
The ARC and the 2 polls are equal - just letting you know. 1 ARC, 2 polls. 3 factors. The GPI has done this for almost 15 years. Carry on.

[Lehigh Football Nation]

Repost the facts sir, in their entirety without your clueless remarks.xrulesx
The ARC and the 2 polls are equal - just letting you know. 1 ARC, 2 polls. 3 factors. The GPI has done this for almost 15 years. Carry on.

Methodology has changed drastically over the last 5 years, something you curiously omit.

The ARC and the 2 polls are also not equal, unless 5 magically equals 2.

[tingly]

As someone who did GPI independently last year from not having the patience to wait for CSN's posts, 3 equal factors - 2 polls and ARC. The ARC factor has 7 components, 2 tossed. Also, its Totals column shouldn't be named Totals. Try Average or GPI.

[MplsBison]

Computer rankings:
MAS = Massey (http://www.masseyratings.com/rate.php?lg=cf), ASH/ABC = Ashburn (http://www.atomicfootball.com/), SAG = Sagarin (http://usatoday30.usatoday.com/sports/sagarin/fbt12.htm), LAZ = Laz Index (http://www.lazindex.com/), KEE = Keeper (http://home.comcast.net/~keepersfootball/site/), SEL = Self (https://github.com/jeffself/collegefootballrankings), BRN = Born (http://www.bornpowerindex.com/).
Polls:
FCP = FCS Coaches (http://www.collegesportingnews.com/content.php?938-2013-FCS-Coaches-Poll), SNW = Sports Network (http://www.collegesportingnews.com/content.php?940-2013-Sports-Network-Poll)

So, 5 computers and 2 human polls. No use of the SRS that the NCAA is having computed for them. No AGS poll. And human polls don't even comprise 1/3 of the total inputs.

Carry on.

And it's completely laughable that college football still allows coaches to have a say, period. It should be a vestigial tradition, at most.

Would you allow inmates to vote on granting parole?

[Wallace]

Methodology has changed drastically over the last 5 years, something you curiously omit. The ARC and the 2 polls are also not equal, unless 5 magically equals 2.

Obviously you choose to not understand it and characterize it falsely for your own reasons. If you are a fan of the FCS then just carry on.

As someone who has watched it over the years, it was a lot better than the national polls and maybe its influence has made the polls better. The GPI is a mix of computer rankings and national polls. The advisor is Kenneth Massey. It was created the same year the FBS (then I-A) came up with the BCS system. Obviously the BCS chose teams for their mystical title but the FCS version just indicates which teams should get the at-large bids. It is a good indicator. Little Joe Blow with his crayon may do better but the GPI has been doing it week-in and week-out for 15 years.

[MplsBison]

Obviously you choose to not understand it and characterize it falsely for your own reasons. If you are a fan of the FCS then just carry on.

As someone who has watched it over the years, it was a lot better than the national polls and maybe its influence has made the polls better. The GPI is a mix of computer rankings and national polls. The advisor is Kenneth Massey. It was created the same year the FBS (then I-A) came up with the BCS system. Obviously the BCS chose teams for their mystical title but the FCS version just indicates which teams should get the at-large bids. It is a good indicator. Little Joe Blow with his crayon may do better but the GPI has been doing it week-in and week-out for 15 years.

Any index, formula, hand-crank or other method of ranking college football teams algorithmically which includes a coaches poll is absolutely flawed. That includes the BCS formula as well.

[Professor Chaos]

Obviously you choose to not understand it and characterize it falsely for your own reasons. If you are a fan of the FCS then just carry on.

As someone who has watched it over the years, it was a lot better than the national polls and maybe its influence has made the polls better. The GPI is a mix of computer rankings and national polls. The advisor is Kenneth Massey. It was created the same year the FBS (then I-A) came up with the BCS system. Obviously the BCS chose teams for their mystical title but the FCS version just indicates which teams should get the at-large bids. It is a good indicator. Little Joe Blow with his crayon may do better but the GPI has been doing it week-in and week-out for 15 years.
No it's not. This was proven last year (http://www.anygivensaturday.com/showthread.php?124455-Which-poll-is-the-best-playoff-predictor) and I'd bet it will be proven again this year.

[Lehigh Football Nation]

Obviously you choose to not understand it and characterize it falsely for your own reasons. If you are a fan of the FCS then just carry on.

As someone who has watched it over the years, it was a lot better than the national polls and maybe its influence has made the polls better. The GPI is a mix of computer rankings and national polls. The advisor is Kenneth Massey. It was created the same year the FBS (then I-A) came up with the BCS system. Obviously the BCS chose teams for their mystical title but the FCS version just indicates which teams should get the at-large bids. It is a good indicator. Little Joe Blow with his crayon may do better but the GPI has been doing it week-in and week-out for 15 years.

In your world, 5 equals 2. Got it. xthumbsupx

[citdog]

SUNDAY! SUNDAY! SUNDAY!

CHUCK VS RALPH IN A LUMBERJACK MATCH!

http://mimg.ugo.com/200809/28797/lumberjack-match.jpg

[tingly]

I forgot, there are 2 GPI's. One does it like Wallace says. The other does it like Lehigh says. The GPI in this thread does Lehigh's version. The fractions in post #1 are sevenths, not thirds.

I don't like polls being involved in any postseason process. I don't imagine that they play much of a role, though. Strength of schedule, who you beat, record against the top X teams are usually the reasons cited by NCAA committees. I've never heard them say "Jones Univ. is out cuz GPI had them 12th on the at-large list."

[dbackjon]

ODU and GSU in there? Next.


Worthless

[heath]

I forgot, there are 2 GPI's. One does it like Wallace says. The other does it like Lehigh says. The GPI in this thread does Lehigh's version. The fractions in post #1 are sevenths, not thirds.

I don't like polls being involved in any postseason process. I don't imagine that they play much of a role, though. Strength of schedule, who you beat, record against the top X teams are usually the reasons cited by NCAA committees. I've never heard them say "Jones Univ. is out cuz GPI had them 12th on the at-large list."
Agree with what you say here. I think when seeding the playoff teams they do look at the polls then. When you have so many AQs, knowing some may not belong, the polls really don't mean much.

[dystopiamembrane]

ODU and GSU in there? Next.
Worthless
c'mon, seriously?

[bluehenbillk]

Why you guys even bother commenting on this is just as bad as the index itself. The AGS Poll has outperformed this 4 years running......

[dystopiamembrane]

The AGS Poll has outperformed this 4 years running......
What do you mean by 'outperformed'?

[citdog]

What do you mean by 'outperformed'?

think Cal Poly last Thursday night.

THAT!

[dystopiamembrane]

think Cal Poly last Thursday night.
I understand what the word 'outperformed' means, but I honestly would like to know how one ranking system outperformed another. What was the criteria used to determine performance?

[citdog]

I understand what the word 'outperformed' means, but I honestly would like to know how one ranking system outperformed another. What was the criteria used to determine performance?


See Post #25 in this thread.

[Wallace]

I understand what the word 'outperformed' means, but I honestly would like to know how one ranking system outperformed another. What was the criteria used to determine performance?

AGS does a poll, I know because I started it, ran it and publicized it for years. Over the 15 years the GPI has been around, no poll has done better. Like all systems used in the GPI, it is public information who is involved. Not in the AGS poll now. It is just a fun thing.

Hey looky there! LehighFootballNation's business partner gave me another negative rep!

Negative Reputation Received 10-02-2013 06:25 AM Mr. C

Kudos to you Mr. C! Backing up your business interests even when you fail to comment in the thread!

anyway...

Alright! #1 vs. #2 and three other ranked battles this Saturday while almost half the top 10 are idle! Scoreboard watching fun!!

Saturday, October 5, 2013

TOP 25 GPI

#2 Northern Iowa @ #1 NORTH DAKOTA ST
New Hampshire @ #3 TOWSON
#4 Eastern Illinois IDLE
#5 Sam Houston State IDLE
#6 Coastal Carolina IDLE
Weber State @ #7 EASTERN WASH
Delaware @ #8 MAINE
#9T Bethune-Cookman IDLE
Lehigh @ #9T FORDHAM
#9T McNeese State @ #17 CENTRAL ARKANSAS
#18 William & Mary @ #12 VILLANOVA
Southern Ill @ #13 SOUTH DAKOTA ST
#14 Northern Ariz @ #24 MONTANA STATE
UC Davis @ #15 SOUTHERN UTAH
Jacksonville St @ #16 TENN-MARTIN
#19 Ga Southern @ SAMFORD
#20 Youngstown St @ INDIANA STATE
SE Missouri St @ #21 TENNESSEE ST
Liberty @ #22 OLD DOMINION
Incarnate Word @ #23 SE LOUISIANA
Yale @ #25T CAL POLY
Presbyterian @ #25T WOFFORD

[dystopiamembrane]

See Post #25 in this thread.
I read that post. My argument would be this. The playoffs are determined by a panel of fleshy-brained machines. Only a few fleshy-brained machines in history have had the ability take into account all of the data that is a college football season and offer unbiased output. My guess is that none of those fleshy-brained machines were on this panel. To argue that the results that came from this panel validate or invalidate a given ranking system is to err in judgement.

[citdog]

I read that post. My argument would be this. The playoffs are determined by a panel of fleshy-brained machines. Only a few fleshy-brained machines in history have had the ability take into account all of the data that is a college football season and offer unbiased output. My guess is that none of those fleshy-brained machines were on this panel. To argue that the results that came from this panel validate or invalidate a given ranking system is to err in judgement.

Results matter. The AGS Poll is THE MOST ACCURATE PREDICTOR OF AT LARGE BERTHS TO THE FCS PLAYOFFS. PERIOD!

[superman7515]

Results matter. The AGS Poll is THE MOST ACCURATE PREDICTOR OF AT LARGE BERTHS TO THE FCS PLAYOFFS. PERIOD!

If you paid attention to his ballot, you would know that results do not matter in Portland.

[dystopiamembrane]

Results matter. The AGS Poll is THE MOST ACCURATE PREDICTOR OF AT LARGE BERTHS TO THE FCS PLAYOFFS. PERIOD!
I don't think you understood what I wrote.

[dystopiamembrane]

If you paid attention to his ballot, you would know that results do not matter in Portland.
are your opinions more valid than mine, superman?

[superman7515]

Yes

[dystopiamembrane]

Yes
then, I accept your ridicule.

[URMite]

Of course not, and the Lehigh fanatic can't even get % of total inputs right. No wonder he is aghast that Lehigh was unqualified last year for the playoffs. Some people are just not football folk.

Here's the reason he's all snitty:

11. Patriot League (63.53)
9T. Fordham (13.86)
41. Lehigh (36.00)
74. Holy Cross (60.29)
95T. Lafayette (74.43)
105. Bucknell (80.43)
112. Colgate (87.00)
121. Georgetown (92.71)

Why "of course not"? I thought it had been stated that they would publish it?

[URMite]

The link is at the top of the post.

I was looking for a link to this

http://www.collegesportingnews.com/stats/writer/gpi/20130930GPI.htm

here

http://www.collegesportingnews.com/content.php?965-2013-Gridiron-Power-Index-%28GPI%29-Opens-With-NDSU-at-1

[tingly]

All I know is everything that is not the FCS selection committee is inaccurate.

[ursus arctos horribilis]

Yes


then, I accept your ridicule.

xlolx to both you funny mofo's. That was excellent. xlolx

[URMite]

This was on TSN on 9/11

"The SRS will factor in all of a team's games against FCS, FBS and Division II competition. In past years, a game against a Division II opponent did not count toward a team's resume.

But the new system will incorporate them into the equation and rightfully differentiate the levels of wins, including both home and away, which will provide a true strength of schedule tool. It will be recalculated on a weekly basis through the final regular-season games on Nov. 23 and the announcement of the playoff field the next morning.

The selection committee also will consider potential at-large teams with six wins against Division I competition, when in the past it basically considered seven to be the benchmark."

Are they recalculating it each week but keeping it to themselves? Or will it be released?

[FUBeAR]

I know we have #1 vs. #2 this week, but I'm also looking forward to seeing how the GPI 'does' in this matchup:



93
Mercer
PFL
4-0
72.57












94
San Diego
PFL
2-2
73.14



Pretty hard to believe a 1st year program can even hope to stay on the field with the 2-time defending league champion and pre-season favorite on their home field, especially after the Toreros just put a 59-0 beatdown on another 1st year program as the visiting team after a cross-country trip. Stetson, the 59-0 victim, is almost identical to Mercer in terms of the academic requirements, relative strength of HS football in the area, Head Coach experience in FCS, and timeline for their start-up. Guess those factors aren't considered by the non-fleshy brains responsible for the GPI.

[Wallace]

... Guess those factors aren't considered by the non-fleshy brains responsible for the GPI.Unfortunately the sports information directors that in fact vote for both the "media" and coach's polls are fleshy-brained, like all poll voters.

[bluehenbillk]

Results matter. The AGS Poll is THE MOST ACCURATE PREDICTOR OF AT LARGE BERTHS TO THE FCS PLAYOFFS. PERIOD!

This!!

[Wallace]

saw this:
http://www.collegesportingnews.com/stats/writer/gpi/20130930compresult.jpg
the polls are very subjective at this time of year :)
the GPI is very rarely not in first place

[dbackjon]

c'mon, seriously?

Yes, seriously. They are not FCS teams. They have no business being in the GPI.

They have moved on from FCS to "greener" pastures. I have done the same. As should the rest of us.

[Lehigh Football Nation]

There are 2 GPIs.

[tingly]

Massey's average is 1st. GPI is 2nd. But Massey stresses that being leftmost doesn't mean "best." Even the rightmost one has a tiny chance of being best. Frankly, crowing about GPI is a bit "standing on the shoulders of giants." Sagarin, Massey, Ashburn,... did 99% of the work. Any middle schooler could take an average of the better ratings and do as well if not better.

Sagarin had UC Davis in his D1 ratings when they were D2. They played in a I-AA conference. I don't mind transition teams being included in FCS ratings/polls, but it is a little confusing for the teams behind them.

[MplsBison]

All I know is everything that is not the FCS selection committee is inaccurate.

Correct.

And that's the way it should be, if an all encompassing, single-elimination tournament (that means including every single team in the sub-division) can't be held to determine the champion.


In other words, if you can't do a Minnesota state high school style "everyone gets one game in the playoffs" tournament, then you're going to have to pick teams to play in the tournament.

No algorithm will ever be perfect or perfectly unbiased (in both cases, they are at the whims of their programming humans). Therefore, a panel of humans who subjectively pick teams for the tournament is the only possible correct way to select those teams.


End

[MplsBison]

I know we have #1 vs. #2 this week, but I'm also looking forward to seeing how the GPI 'does' in this matchup:



93
Mercer
PFL
4-0
72.57












94
San Diego
PFL
2-2
73.14



Pretty hard to believe a 1st year program can even hope to stay on the field with the 2-time defending league champion and pre-season favorite on their home field, especially after the Toreros just put a 59-0 beatdown on another 1st year program as the visiting team after a cross-country trip. Stetson, the 59-0 victim, is almost identical to Mercer in terms of the academic requirements, relative strength of HS football in the area, Head Coach experience in FCS, and timeline for their start-up. Guess those factors aren't considered by the non-fleshy brains responsible for the GPI.

Well said.

At some point, only actual humans can 'get it'. Computers are too rigid and always will be bounded by the parameters of the programming.

[MplsBison]

Unfortunately the sports information directors that in fact vote for both the "media" and coach's polls are fleshy-brained, like all poll voters.

Why are you saying that SID's vote for media polls?

[MplsBison]

Yes, seriously. They are not FCS teams. They have no business being in the GPI.

They have moved on from FCS to "greener" pastures. I have done the same. As should the rest of us.

You're just being obtuse.

They're not officially FBS this year, they're playing mostly FCS teams = they should be included in the FCS polls. End

[superman7515]

You're just being obtuse.

They're not officially FBS this year, they're playing mostly FCS teams = they should be included in the FCS polls. End

The NCAA says they're not. His mama call him Clay, Imma call him Clay.

http://i.imgur.com/kT8FU.jpg

[MplsBison]

The NCAA says they're not. His mama call him Clay, Imma call him Clay.



But the NCAA also says they're not FBS and they are playing mostly FCS teams. Therefore, they should be included in FCS polls.

[dbackjon]

You're just being obtuse.

They're not officially FBS this year, they're playing mostly FCS teams = they should be included in the FCS polls. End

Schedule doesn't matter. They give out more than 63 schollies = NOT FCS.

End discussion. I win. you lose. And, your mother dresses you funny.

[clenz]

Schedule doesn't matter. They give out more than 63 schollies = NOT FCS.

End discussion. I win. you lose. And, your mother dresses you funny.
If Ohio State/Michigan played mostly FCS schools, would they be FCS schools as well?


Here is the guildine for being an FCS school

Ready?


May only give the equivalent of 63 scholarships to no more than 85 players.

ODU, GSU, and ASU fans have all admitted (as have their athletic departments) that they are all somewhere in the 70s.


How are people too ****ing stupid to understand that?

[ursus arctos horribilis]

If Ohio State/Michigan played mostly FCS schools, would they be FCS schools as well?


Here is the guildine for being an FCS school

Ready?


May only give the equivalent of 63 scholarships to no more than 85 players.

ODU, GSU, and ASU fans have all admitted (as have their athletic departments) that they are all somewhere in the 70s.


How are people too ****ing stupid to understand that?

Most are not. Some just want the argument and don't mind actin' the fool to do so. It's too bad cuz he was actually having some good discussions for a while but can't resist going back to his silliness. That's OK, it gives me good reason to start f'n with him so it's a win for me.

[dbackjon]

But the NCAA also says they're not FBS and they are playing mostly FCS teams. Therefore, they should be included in FCS polls.


Then have a transitional poll

[ursus arctos horribilis]

Then have a transitional poll

http://www.metaspring.com/blog/wp-content/uploads/2010/06/0000007501_20060920143802.jpg

[Lehigh'98]

This looks pretty similar to my AGS Top 25 ballot.
Notice everyone:


No Gardner-Webb
No Charleston Southern
No Lehigh
No Delaware
No JMU
No UNH

Why no GW or UD?? They both have impressive wins(Wofford & JMU)

[Bisonwinagn]

Why no GW or UD?? They both have impressive wins(Wofford & JMU)

I think we need to still wait a few more weeks before we can determine impressive wins.

[dystopiamembrane]

My final two cents on the subject -
I have been following college football intently for 24 years, first as a mathematical exercise, but quickly as a true fan of the sport, its variegation across divisions and regions, the ardent supporters groups, the beauty of the season and the game, NJCAA, FBS and everything in between, becoming a fan of teams I've never come close to seeing live, talking about periods of the year based on how close they are to the season, spending between 10 and 13 hours every season Saturday in the greatest sports bar on earth, the mighty SkyBox (http://www.skyboxpub.com) in Portland, Oregon, to swallow as much as possible, finding every scrap of info on the internet, in numerous sports pages, etc., yes, I call it an addiction, I assume many of you have the same disease.
That said, I, myself, and this is just a declaration for me, personally, I am not able to take all that data, thousands of games per year, and use my brain to come up with an accurate ranking system. Sure, I can make some really good guesses and come up with an acceptable AGS submission, but always, in the back of my mind, I know I'm short-changing some teams and giving others more credit than they deserve. I truly respect all of you on this board and the many of you who do create your AGS submission this way. I just can't do it that way. I know it seems to many counter to real-life, but I would prefer that all rankings and the selection of playoff participants to be chosen by a single accepted algorithm. Just me.
And, I am done. You'll see no more rants from me on the subject.
Thank you all for being fans of the fine game and respecting the "small schools" as you do. I hope all your favorite teams give you the joy and heartache we all crave from this sport.

[clenz]

The problem is computers leave so much data out. No matter how much time is put into setting the algorithm it's still greatly flawed. It also has whatever bias the programmer wants to build into it.

Great, the computer sees the scores, and the scores of the opponents....it can't take into play a million other variables.

It can't take into account weather - a downpour greatly affecting a passing team, a field so sloppy that neither team can actually accomplish anything, a snow storm causing mayhem.
It can't take into account elevation - a team from less than 1000 feet playing at more than 2000 feet is going to be at a great disadvantage no matter how you want to spin it.
It can't take into account blown calls from a ref - a fumble that wasn't a fumble, a catch that wasn't a catch, a penalty that wasn't a penalty, etc... Those things can and do affect a game every single week.
It can't take into account injuries - Montana State without DM is a very different team than with DM...not just offensively but as a complete unit. They did't look the same without him.
It can't take into account an anomaly of a game to the full extent - it can try to reason it by shifting teams why up or way down, ignoring the result, etc... but it can't see the real result. Take the Montana/NAU game. I highly doubt Montana is that bad, but all the computer will see is the score (not that NAU is a bad opponent). That not only greatly affects Montana, but by proxy the entire Big Sky and most of the country.
It can't take into account anything other than scores and whatever variables you want to program from "strength".


I can't watch every game, I can't read about every game, I can't look at the box score for every game, I can't do any of that any more than a computer can with looking at each score and assigning a strength rating on it.

I am able to take input from the games, other than score, to see what may have played into it. Computers can't.



AGS is a collection of about 100 people from all across the country. None of them can watch every game, read about every game, look at the box score for every game, any of that any more than a computer can with looking at each score and assigning a strength rating on it. However, adding the human element of over 100 people across the nation gets a much more accurate rating than any computer models could ever put together.

[MplsBison]

Schedule doesn't matter. They give out more than 63 schollies = NOT FCS.

End discussion. I win. you lose. And, your mother dresses you funny.

No, because you make the same mistake that superman7515 does in trying to argue that ASU & GSU are not officially designated as FCS teams.

I wasn't arguing that they should be officially designated as FCS teams.

[MplsBison]

If Ohio State/Michigan played mostly FCS schools, would they be FCS schools as well?


Here is the guildine for being an FCS school

Ready?


May only give the equivalent of 63 scholarships to no more than 85 players.

ODU, GSU, and ASU fans have all admitted (as have their athletic departments) that they are all somewhere in the 70s.


How are people too ****ing stupid to understand that?

I am not arguing that ASU & GSU should be designated as FCS teams.

Furthermore, FBS teams are required by the rules of that sub-division to play a minimum number of games against other FBS teams. That precludes their schedules from ever being against mostly FCS teams. Therefore, your hypothetical is no more valid than proposing a piece of broccoli playing mostly FCS teams.

[MplsBison]

Then have a transitional poll

No need. They play mostly FCS teams, therefore they should be included in the FCS poll.

[MplsBison]

The problem is computers leave so much data out. No matter how much time is put into setting the algorithm it's still greatly flawed. It also has whatever bias the programmer wants to build into it.

Great, the computer sees the scores, and the scores of the opponents....it can't take into play a million other variables.

It can't take into account weather - a downpour greatly affecting a passing team, a field so sloppy that neither team can actually accomplish anything, a snow storm causing mayhem.
It can't take into account elevation - a team from less than 1000 feet playing at more than 2000 feet is going to be at a great disadvantage no matter how you want to spin it.
It can't take into account blown calls from a ref - a fumble that wasn't a fumble, a catch that wasn't a catch, a penalty that wasn't a penalty, etc... Those things can and do affect a game every single week.
It can't take into account injuries - Montana State without DM is a very different team than with DM...not just offensively but as a complete unit. They did't look the same without him.
It can't take into account an anomaly of a game to the full extent - it can try to reason it by shifting teams why up or way down, ignoring the result, etc... but it can't see the real result. Take the Montana/NAU game. I highly doubt Montana is that bad, but all the computer will see is the score (not that NAU is a bad opponent). That not only greatly affects Montana, but by proxy the entire Big Sky and most of the country.
It can't take into account anything other than scores and whatever variables you want to program from "strength".


I can't watch every game, I can't read about every game, I can't look at the box score for every game, I can't do any of that any more than a computer can with looking at each score and assigning a strength rating on it.

I am able to take input from the games, other than score, to see what may have played into it. Computers can't.



AGS is a collection of about 100 people from all across the country. None of them can watch every game, read about every game, look at the box score for every game, any of that any more than a computer can with looking at each score and assigning a strength rating on it. However, adding the human element of over 100 people across the nation gets a much more accurate rating than any computer models could ever put together.

Well said!!

Computer algorithms that rank college football teams are nothing more than for entertainment. As in "oh funny, look at what Massey ranks [team A] at vs. what Sagarin ranks them. Hehehe."

[ursus arctos horribilis]

No need. They play mostly FCS teams, therefore they should be included in the FCS poll.

Nope.

[MplsBison]

Nope.

I'm right. You're wrong.

[robsnotes4u]

There are things a computer does better, and things people may do better. One thing that really sticks out when people pick, is the preconceived thoughts people have about teams. For instance, a Griz fan that hates the Cats or the other way around, will place them lower in a pool even though all of the facts say they are better. It is just like an reading an email your perception is dependent upon what you think of the person and your mood, because there isn't anything else to draw from. You can't get away from it your perceptions are built on your experiences, and we will make excuses for teams to prove our perceptions correct. Just ask yourself this question, why is it so hard for a team that has never been in the top 25 to break into it, and easy to keep a perennial top 25 team in there that may have a 1-2 record?

[robsnotes4u]

My opinion on polls and transitional teams. I think it is simple, if you are not eligible to be in the playoffs and play in the National Championship there is no reason to be in the polls. Right or wrong.

[ursus arctos horribilis]

I'm right. You're wrong.

I was sort of delighted to see you go back to your idiotic childish the other day MPLS cuz I was tossing around an idea that I was setting aside due to your not being such an asshole and trying to have productive conversations for a short time. Get ready there buttercup, I got a surprise for you as it pertains to Jay's fund drive.:D

[clenz]

There are things a computer does better, and things people may do better. One thing that really sticks out when people pick, is the preconceived thoughts people have about teams. For instance, a Griz fan that hates the Cats or the other way around, will place them lower in a pool even though all of the facts say they are better. It is just like an reading an email your perception is dependent upon what you think of the person and your mood, because there isn't anything else to draw from. You can't get away from it your perceptions are built on your experiences, and we will make excuses for teams to prove our perceptions correct. Just ask yourself this question, why is it so hard for a team that has never been in the top 25 to break into it, and easy to keep a perennial top 25 team in there that may have a 1-2 record?
I don't disagree with that either. I certainly believe there is *some* merit to computer rankings. When I was doing weekly voting I used computers are a resource.

I was one of the people involved in starting the CS.com poll...ursus and I have had many lengthy discussions about voting/polls/computers in the past couple years. I always felt as though the poll I submitted was extremely high quality - and was usually very very close to the AGS/CS polls, at time landing 10+ teams in identical spots. Sadly, I don't have the amount of time to properly do that anymore, which is why I don't vote.

Bias plays a part of it in humans - which is why the best run polls have algorithms in place to take team/conference/regional bias into account. I know the CS poll has that, and I think the AGS poll does to an extent as well. It's unfortunate those things happen, but they do...don't kid yourself though they happen in computers just as easily. Computer algorythems are subject to the same human biases of those that write them. I don't have the knowledge to know how to write one, but if I did I could write one with such an algorithm that no matter what the MVFC teams get major boosts for every W and almost no hit for an L and make it so a Patriot team gets dinged for almost everything. It's the fact of the situation.

As for the "why is it so hard to new teams to break into the top 25" thing - you see it in computers as well. Humans are slow to react to an unknown (or may overreact). Especially if the team has a history of being no good, or the conference has a history of being no good. I don't remember which thread I said it on but there is a reason CCU, CSU, and EIU are slow getting "proper" respect in some eyes. It's the fool me once thing. How many times have we seen an OVC/Big South team look like great during the season but come playoffs it's a first round exit by a large margin. The OVC is something like 0-21 since about 1999 in the playoffs. Is it fair to this years OVC to hold that against them? Maybe, maybe not. However, there is really no history to prove more respect should be given. We see it year after year with the MEAC schools like SC State and Bethune Cookman as well. That kind of history is something a computer can't take into account.


There's no fool proof method, but to trust any computer system over a well run, and diverse, poll like AGS is asinine, IMO.


FWIW, guess where almost every computer had 5-6 UNI last season?


I'll wait while you guess...



Okay...ready for this?

Top 10-20....nearly every single one that is respected.

I believe they finished about 20-25 in most human polls. Why is that? Based on name, or the fact they were actually pretty damn good but just had some real bad luck last year?

[Walkon79]

But the NCAA also says they're not FBS and they are playing mostly FCS teams. Therefore, they should be included in FCS polls.


Why? No one seems to have a good answer for that. These teams are in "transition". Out of the FCS and into FBS. They belong to nobody right now, right?

[clenz]

Why? No one seems to have a good answer for that. These teams are in "transition". Out of the FCS and into FBS. They belong to nobody right now, right?
Basically....They aren't FCS, they aren't FBS.

They are...nothing

[Walkon79]

What's the point? If one of the transition teams ended up high in the poll, it still means nothing.

[clenz]

What's the point? If one of the transition teams ended up high in the poll, it still means nothing.
For the FCS all polls are meaningless...so to take your point further: Why have them at all?

They aren't playing by FCS rules...they aren't eligible for the FCS playoffs...they aren't eligible for their conference title (the are in their conference INO)...They aren't FCS.

[Lehigh Football Nation]

Today computer polls have built-in inaccuracy because there is simply an inadequate amount of data points to effectively judge FCS "wins" or "losses" vs. FBS teams. FBS teams almost always only have one data point against an FCS school, and it's (most often) a W or (occasionally) a L.

If Michigan beats Ohio State and Ohio State goes 7-1 we can to some degree effectively judge Michigan's strength against Ohio State and other FBS schools because not only does Ohio State have many common opponents with Michigan, the teams they play might have a paper trail to common opponents to Michigan, adding extra data.

When Appalachian State beat Michigan, though, how to effectively judge which team is better together? No other SoCon team played Michigan. No other App State opponents played Michigan. Georgia Southern beat App State who beat Michigan, which gave the Eagles and the rest of the SoCon huge jumps in plenty of computer polls like Sag and the like.

What ultimately ends up happening is that you end up judging App's win over Michigan, and the wins Michigan has in FBS. By definition, when you beat an FCS team, you also get a boost from who they played - at most, one FBS team. But if you beat an FBS team, you also have a chance for collateral wins against 11 other FBS opponents.

Last year, Eastern Michigan was horrible, and would have had a tough time staying over .500 in most FCS leagues. Illinois State beat them. Given 11 other chances to win FBS games, EMU won twice against other horrible FBS teams - Army and Western Michigan. Illinois State got a "boost" in the polls from those collateral wins, no matter how crappy those teams actually are.

How do you reconcile this? You can't. There simply isn't enough data for a computer to consider to do so. Somewhere, along the line, even the absolute worst FBS team will better than more than half of the teams in FCS. And that's why the data they spew out seems to look like such garbage. It's a system based on a closed, fixed number of opponents that needs a certain minimum amount of data to become "right". Once you open systems, it breaks down.

[Bisonator]

They are in transitional purgatory!:D

[MplsBison]

For the FCS all polls are meaningless...so to take your point further: Why have them at all?

They aren't playing by FCS rules...they aren't eligible for the FCS playoffs...they aren't eligible for their conference title (the are in their conference INO)...They aren't FCS.

Which doesn't mean they should be excluded from an FCS poll. After all, they play mostly FCS teams.

[Walkon79]

Which doesn't mean they should be excluded from an FCS poll. After all, they play mostly FCS teams.

That's your ONLY argument?

Since you say it's so, it must be I guess. :)

[clenz]

Which doesn't mean they should be excluded from an FCS poll. After all, they play mostly FCS teams.

so again, if Michigan or Ohio State played mostly fcs schools, should they be eligible

Sent from my SCH-R530U using Tapatalk 2

[Bisonator]

Which doesn't mean they should be excluded from an FCS poll. After all, they play mostly FCS teams.

xrotatehx

They have more than 63 schollies. End of story!xbangx

[ElCid]

Today computer polls have built-in inaccuracy because there is simply an inadequate amount of data points to effectively judge FCS "wins" or "losses" vs. FBS teams. FBS teams almost always only have one data point against an FCS school, and it's (most often) a W or (occasionally) a L.

If Michigan beats Ohio State and Ohio State goes 7-1 we can to some degree effectively judge Michigan's strength against Ohio State and other FBS schools because not only does Ohio State have many common opponents with Michigan, the teams they play might have a paper trail to common opponents to Michigan, adding extra data.

When Appalachian State beat Michigan, though, how to effectively judge which team is better together? No other SoCon team played Michigan. No other App State opponents played Michigan. Georgia Southern beat App State who beat Michigan, which gave the Eagles and the rest of the SoCon huge jumps in plenty of computer polls like Sag and the like.

What ultimately ends up happening is that you end up judging App's win over Michigan, and the wins Michigan has in FBS. By definition, when you beat an FCS team, you also get a boost from who they played - at most, one FBS team. But if you beat an FBS team, you also have a chance for collateral wins against 11 other FBS opponents.

Last year, Eastern Michigan was horrible, and would have had a tough time staying over .500 in most FCS leagues. Illinois State beat them. Given 11 other chances to win FBS games, EMU won twice against other horrible FBS teams - Army and Western Michigan. Illinois State got a "boost" in the polls from those collateral wins, no matter how crappy those teams actually are.

How do you reconcile this? You can't. There simply isn't enough data for a computer to consider to do so. Somewhere, along the line, even the absolute worst FBS team will better than more than half of the teams in FCS. And that's why the data they spew out seems to look like such garbage. It's a system based on a closed, fixed number of opponents that needs a certain minimum amount of data to become "right". Once you open systems, it breaks down.

This is a pretty good explanation of the shortcomings. I suppose it could also explain how Harvard ends up so high each year in the computer ratings becaue the Ivy has such a closed pool of teams that it plays. You only need to win one or two OOC games against medium opponents and if those opponents have a descent schedule and do well otherwise it boosts you nicely. Not to knock Harvard too bad, I know they have a pretty good team, but really? They are in the top 20 on the computer lately and I just do not think they are really that good. Sagarin has them ahead of Cal-Poly, ODU, Portland, Wofford, Stoney Brook, and a host of other teams that would win and possibly smoke em.

[MplsBison]

That's your ONLY argument?

Since you say it's so, it must be I guess. :)

Why do I need more than one correct argument?

[MplsBison]

so again, if Michigan or Ohio State played mostly fcs schools, should they be eligible

Repeating an invalid hypothetical does not help your argument. I've explained why Michigan and Ohio State can't play mostly FCS teams.

Next year, ASU & GSU also won't be able to play mostly FCS teams by the same rule. This year, however, they aren't bound by such a rule and in fact do play mostly FCS teams. Therefore, they should be included in an FCS poll/ranking.

[MplsBison]

xrotatehx

They have more than 63 schollies. End of story!xbangx

I'm not arguing for them to be designated as FCS teams.

[clenz]

Hypotheticals are just that.

I know you won't answer the question, because that's who you are...

Sent from my SCH-R530U using Tapatalk 2

[MplsBison]

Hypotheticals are just that.

I know you won't answer the question, because that's who you are...


??? Your statement makes no sense.

If you were going to propose an absurd hypothetical, why didn't you propose an NFL team playing mostly FCS teams?

[clenz]

Sure...

If they are playing mostly fcs schools shouldn't they be able to be ranked

Sent from my SCH-R530U using Tapatalk 2

[MplsBison]

Sure...

If they are playing mostly fcs schools shouldn't they be able to be ranked

I don't entertain absurdities. The world is vexing enough as it is.

ASU & GSU play mostly FCS teams, so they should be included in any ranking or poll of FCS teams.


Neither youself nor anyone else has even attempted to explain why not. That they aren't classified as FCS teams and that FBS or NFL teams aren't included in FCS rankings & polls are not valid counter arguments.

[clenz]

Does anyone else see the irony in mpls talking about entertaining absurdity?

So the fact that they aren't FCS isn't reason enough for you them to be exluded from the FCS polls?


Huh....I'm done here.

[citdog]

I don't entertain absurdities. The world is vexing enough as it is.

ASU & GSU play mostly FCS teams, so they should be included in any ranking or poll of FCS teams.


Neither youself nor anyone else has even attempted to explain why not. That they aren't classified as FCS teams and that FBS or NFL teams aren't included in FCS rankings & polls are not valid counter arguments.


ok mpls I'll play for a second. Appy and Pig's Ass have more than 63 schollys and that makes it unfair to rank them. clear enough?

[Southern Bison]

Citdog...I'll fill in for MPLS for a second. "But...but...but..."

[bisonboone11]

I don't entertain absurdities. The world is vexing enough as it is.

ASU & GSU play mostly FCS teams, so they should be included in any ranking or poll of FCS teams.


Neither youself nor anyone else has even attempted to explain why not. That they aren't classified as FCS teams and that FBS or NFL teams aren't included in FCS rankings & polls are not valid counter aguments
Let me start off with this: I couldn't care less if they are included in fcs polls or not.

Now, you simply stating "that they aren't classified as FCS teams and that FBS or NFL teams aren't included in FCS rankings & polls are not valid counter arguments" does not in fact make those counter arguments invalid. They are just as valid as (and likely more valid than) your argument that they should be included because they play mostly FCS teams. But since you have deemed their arguments invalid because they contradict your argument, how about this one. The NCAA has determined that teams with greater than 63 scholarships have an unfair advantage over FCS teams. For that reason, those teams should not be included in FCS polls and rankings.

Or how about this counter argument. They should not be included in FCS polls and rankings because they are not FCS teams. When I look at NFL rankings, I expect not to see college teams. There is no reason to have non-NFL teams in NFL polls and rankings. Similarly, there is no reason to have non-FCS teams in FCS polls.

[bisonboone11]

Do any of the polls state that "the criteria for determining which teams shall be included in FCS polls is whether or not they play mostly FCS teams"? I honestly don't know. I haven't looked carefully enough at them to see that. I have, however, noticed that they do indicate they are FCS polls, which to me would mean they would not include non-FCS teams.

[Wallace]

The GPI isn't a poll and it states that transitional teams are included. That's it.

Does the NCAA rule that "teams with greater than 63 scholarships have an unfair advantage over FCS teams" and "for that reason, those teams should not be included in FCS polls and rankings"?

ASU, GSU, ODU, etc. are all transition teams. They can't win any FCS title etc. because they are outside FCS rules. But they do mostly play FCS schedules this year so the FCS is the closest group they fit in. No big deal in indicating teams in a quest for the title, just leave them out as the Ivy and some SWAC are left out.