Championship Subdivision Football | FCS Football | Stadiums | Blogs | Forums

Top Rankings for Computer and Human Poll Biases - Week after 10/31 Games*


Top 10 Teams Whose Rankings are Biased Towards the Computer's Favor

1 Northern Arizona (5-3) 0.419 (Last Week's Ranking #1)
Middle 5 Computer Rankings: #13, #13, #13, #15, #19
Human Rankings: #26, #27, #28-29 (Tied)

2 Delaware (5-3) 0.405 (Last Week's Ranking #2)
Middle 5 Computer Rankings: #11, #11, #12, #13, #14
Human Rankings: #22, #23, #25

3 Massachusetts (4-4) 0.323 (Last Week's Ranking #3)
Middle 5 Computer Rankings: #16, #17, #17, #17, #20
Human Rankings: #29, #32, #33

4 James Madison (3-5) 0.312 (Last Week's Ranking #17)
Middle 5 Computer Rankings: #15, #17, #18, #19, #21
Human Rankings: #37-38(Tied), #38-39(Tied), Unranked

5 Southern Utah (4-4) 0.309 (Last Week's Ranking #8)
Middle 5 Computer Rankings: #15, #15, #16, #20, #24
Human Rankings: #33-34(Tied), #34, #37-38(Tied)

6 Eastern Washington (6-3) 0.307 (Last Week's Ranking #5)
Middle 5 Computer Rankings: #10, #10, #11, #13, #14
Human Rankings: #16, #18, #21

7 Jacksonville State (5-3) 0.183 (Last Week's Ranking #6)
Middle 5 Computer Rankings: #10, #11, #11, #14, #18
Human Rankings: #15, #17, #17

8 Northern Iowa (5-3) 0.150 (Last Week's Ranking #7)
Middle 5 Computer Rankings: #7, #7, #9, #9, #10
Human Rankings: #11, #12, #12

9 Cal Poly SLO (4-4) 0.148 (Last Week's Ranking #13)
Middle 5 Computer Rankings: #19, #20, #20, #21, #27
Human Rankings: #26, #26, #27

10 South Dakota State (7-1) 0.133 (Last Week's Ranking #9)
Middle 5 Computer Rankings: #5, #5, #6, #6, #6
Human Rankings: #9, #9, #9


Dropped Out of Top 10
UC Davis
Montana


Top 10 Teams Whose Rankings Are Biased Towards the Human Poll's Favor

1 McNeese State (6-2) -0.573 (Last Week's Ranking #1)
Middle 5 Computer Rankings: #23, #25, #26, #29, #29
Human Rankings: #11, #11, #12

2 Holy Cross (7-1) -0.427 (Last Week's Ranking #4)
Middle 5 Computer Rankings: #26, #29, #30, #31, #31
Human Rankings: #13, #13, #14

3 Appalachian State (6-2) -0.333 (Last Week's Ranking #2)
Middle 5 Computer Rankings: #11, #12, #16, #19, #21
Human Rankings: #7, #7, #8

4 Liberty (6-2) -0.260 (Last Week's Ranking #7)
Middle 5 Computer Rankings: #24, #24, #25, #25, #25
Human Rankings: #15, #16, #17

5 Colgate (8-1) -0.242 (Last Week's Ranking #8)
Middle 5 Computer Rankings: #38, #39, #40, #42, #48
Human Rankings: #18, #20, #21

6 Central Arkansas (5-3) -0.239 (Last Week's Ranking #3)
Middle 5 Computer Rankings: #31, #34, #36, #41, #41
Human Rankings: #19, #19, #20

7 Eastern Illinois (7-2) -0.151 (Last Week's Ranking Unranked)
Middle 5 Computer Rankings: #16, #18, #21, #21, #22
Human Rankings: #13, #14, #14

8 Florida A&M (6-2) -0.123 (Last Week's Ranking #13)
Middle 5 Computer Rankings: #27, #30, #35, #36, #38
Human Rankings: #23, #24, #24

9 Prairie View (5-1) -0.117 (Last Week's Ranking #12)
Middle 5 Computer Rankings: #36, #36, #40, #41, #43
Human Rankings: #22, #22, #29

10 South Carolina State (7-1) -0.095 (Last Week's Ranking #5)
Middle 5 Computer Rankings: #12, #12, #12, #13, #14
Human Rankings: #10, #10, #10


Dropped Out of Top 10
Elon
Southern Illinois
Weber State


*Computer Rankings: (% of points of Massey, Sagarin, Laz Index, Keeper, Self, Ashburn, and Sauceda (-high and low rankings))/5
Human Poll: (% of points AGS + % of points SN + % of points coaches poll)/3
Bias = Computer Rankings - Human Poll



Bonus Ranking

Top 2 Teams Whose Rankings are Biased Between Human Polls

1 Prairie View 0.159** % of points of TSN poll - % of points of AGS poll
2 Eastern Washington 0.142 *** % of points of TSN poll - % of points of FCS poll

** Prairie View received 612 points out of 142 voters in the The Sports Network poll and only 29 points out of 90 voters in the Any Given Saturday poll.
*** Eastern Washington received 225 points out of 28 voters in the FCS Coaches poll and only 634 points out of 142 voters in The Sports Network poll.

Link to the Gridiron Power Index (GPI) 11/02/2009
http://www.collegesportingnews.com/stat ... 02gpi.html

Good stuff.

Your use of color for emphasis is excellent. However, I have no fucking idea what I just read!

CatMom wrote:Your use of color for emphasis is excellent. However, I have no fucking idea what I just read!

CatMom wrote:Your use of color for emphasis is excellent. However, I have no fucking idea what I just read!

I just used my decoder ring:

Computers might actually have this crap worked out this year.
Humans completed f'ed up the AGS, TSN and Coaches Poll this week.

CatMom wrote:Your use of color for emphasis is excellent. However, I have no fucking idea what I just read!

Find a man to explain it to you.



















j/k

...or maybe a grad from a CAA school.

BlueHen86 wrote:

CatMom wrote:Your use of color for emphasis is excellent. However, I have no fucking idea what I just read!

Find a man to explain it to you.


j/k

...or maybe a grad from a CAA school.

I know more about football than the men in my house.......volunteers?

CatMom wrote:

BlueHen86 wrote:
Find a man to explain it to you.


j/k

...or maybe a grad from a CAA school.

I know more about football than the men in my house.......volunteers?

If you look at the computer polls...and the human polls...there are some big discrepancies...

The original posts show the top 10 widest discrepancies where the computer favors a team versus a human poll...and the top 10 widest discrepancies where the human polls favor a certain team...

It's a very interesting comparison...some folks love the computer polls and some love the human polls...

I also think it suggests that the computers give more weight to the power conferences.

Don't like polls. I don't need a computer to figure out the 8 auto-bid conference winners. Does anyone else? After that, with any form of intelligence, I could probably pick the next 8 best teams. The committee does it. I know it's not always the next best 8 teams but, you get the idea...right?

Last year we made the field of 16 for the playoffs. Yeah, we lost but, to me that puts those 16 teams as 1-16. We ended up 22nd, I think. Had a better record than McNeese and beat them and never made a poll till the post season field was announced (basically, when we won conference) and didn't skip over them until after the playoffs. Polls suck. This is my opinion folks, so trying to change it is futile.

I'm sure it (the above poll scenario) has happened before and will continue as long as polls are around and that will likely be forever.

CatMom wrote:Don't like polls. I don't need a computer to figure out the 8 auto-bid conference winners. Does anyone else? After that, with any form of intelligence, I could probably pick the next 8 best teams. The committee does it. I know it's not always the next best 8 teams but, you get the idea...right?

Last year we made the field of 16 for the playoffs. Yeah, we lost but, to me that puts those 16 teams as 1-16. We ended up 22nd, I think. Had a better record than McNeese and beat them and never made a poll till the post season field was announced (basically, when we won conference) and didn't skip over them until after the playoffs. Polls suck. This is my opinion folks, so trying to change it is futile.

I'm sure it (the above poll scenario) has happened before and will continue as long as polls are around and that will likely be forever.

That's a great point and helps explain why if a 9-2 team from a non-power conference doesn't get a bid it's ok because nobody thought they'd make it out of the first round anyway, whereas if an 8-3 team that's hot at the end of the season from a power conference gets snubbed, they at least could have upset the bracket or stayed hot and won it all.

kalm wrote:

CatMom wrote:Don't like polls. I don't need a computer to figure out the 8 auto-bid conference winners. Does anyone else? After that, with any form of intelligence, I could probably pick the next 8 best teams. The committee does it. I know it's not always the next best 8 teams but, you get the idea...right?

Last year we made the field of 16 for the playoffs. Yeah, we lost but, to me that puts those 16 teams as 1-16. We ended up 22nd, I think. Had a better record than McNeese and beat them and never made a poll till the post season field was announced (basically, when we won conference) and didn't skip over them until after the playoffs. Polls suck. This is my opinion folks, so trying to change it is futile.

I'm sure it (the above poll scenario) has happened before and will continue as long as polls are around and that will likely be forever.

That's a great point and helps explain why if a 9-2 team from a non-power conference doesn't get a bid it's ok because nobody thought they'd make it out of the first round anyway, whereas if an 8-3 team that's hot at the end of the season from a power conference gets snubbed, they at least could have upset the bracket or stayed hot and won it all.

Not sure of your tone but I can state a case in point that almost everyone else in the FCS world chimes in about all the time.
Last season UCA was 9-2. Should they be in the polls? yes. Top 10 - can't say but as they were ineligible they should be outside the final top 16. Why? They didn't go to the playoffs. They will when eligible. Were they a better team than TXST, yes but we were the NCAA declared/recognized champion and we went. Should McNeese have been ranked higher than TXST? No. Why? after week 2 of conference we beat them. We both lost to UCA. Old news anyway.
If UCA had been eligible would I have wanted TXST to go to the playoffs? No, not as an 8-4 team. Being the conference champions, regardless of the record, gives that auto-bid, plain and simple. Afterwards you really have to have been following the teams all year (SoS, etc) and know what has been happening to make those informed decisions. People will always cry foul (believe me, I know)

Look out cause if things roll a particular way the SLC could have a non 7 DI win champion this year. Then what do you do?

"It's all a riddle, wrapped in a mystery inside an enigma."

(and now I'll shut the fuck up because I don't know what I'm talking about)

CatMom wrote:Don't like polls. I don't need a computer to figure out the 8 auto-bid conference winners. Does anyone else? After that, with any form of intelligence, I could probably pick the next 8 best teams. The committee does it. I know it's not always the next best 8 teams but, you get the idea...right?

Last year we made the field of 16 for the playoffs. Yeah, we lost but, to me that puts those 16 teams as 1-16. We ended up 22nd, I think. Had a better record than McNeese and beat them and never made a poll till the post season field was announced (basically, when we won conference) and didn't skip over them until after the playoffs. Polls suck. This is my opinion folks, so trying to change it is futile.

I'm sure it (the above poll scenario) has happened before and will continue as long as polls are around and that will likely be forever.

Polls don't suck (except for the FBS polls). They exist to generate interest and discussion of the sport and there is nothing wrong with that.

The FBS polls suck because they are used to determine a champion, which is dumb, championships should be decided on the field, not on paper.