Since 1936, the Associated Press (AP) has released a weekly college football ranking. Why 1936? Because of a man named Alan J. Gould.

Gould was the one who first organized a collective of coaches and critics to name the “All-America” team after the original creator, Walter Camp, had, for years, used the collective of he, himself, and he to create the team. Alan Gould was the sports editor for the Associated Press from 1922-1938.

On November 15, 1934, the first AP ranking was released as a Top 10 with Minnesota topping the poll with 50 first place votes. The following year, Gould would release his end of season rankings and declared a three-way tie for first place: Minnesota, Princeton, and Southern Methodist University.

This didn’t sit well with many fans (because who wants to tie for first?) and so, the following year, 1936, Gould arranged for a weekly ranking with input from 100 coaches and critics from across the sport. Ever since, we have had a Top 20 (then a Top 10 in 1961, then a Top 20 in 1968, and now, a Top 25 since 1989) ranking every week of the college football season.

89 Years Later

We have the AP Poll, the Coaches Poll, and, late in the season, the only poll that matters with the CFP Rankings. For the AP Poll, there are 66 voters. The final rankings are based on a point value system for each team’s individual ranking on all 66 ballots (25 for first, 24 for second, and so on).

Now, I’m gonna take a look at the ballots cast for this week and show you the range for each team selected to the Top 25.

Week 8		Highest	Lowest
1	Ohio State	1	4
2	Miami	1	3
3	Indiana	1	5 (Ralph Russo)
4	Texas A&M	3	6
5	Ole Miss	4	12 (Koki Riley)
6	Alabama	3	9
7	Texas Tech	4	11 (Spencer Ripchik)
8	Oregon	6	12 (Chad Bishop)
9	Georgia	5 (Pete Yanity)	12
10	LSU	6	18 (Sean Reider)
11	Tennessee	8	16
12	Georgia Tech	6 (Stephen Means)	25 (Koki)
13	Notre Dame	7 (Kirk Kenney)	21 (Randy Johnson)
14	Oklahoma	9	23
15	BYU	10 (Means)	NR (Koki)
16	Missouri	8 (Spencer)	23
17	Vanderbilt	13	24
18	Virginia	12	NR (Koki)
19	South Florida	13	NR
20	USC	13 (Koki)	NR (Keith Farmer)
21	Texas	12	NR
22	Memphis	12	NR
23	Utah	13	NR
24	Cincinnati	17	NR
25	Nebraska	16	NR
	Illini	17	NR
	Michigan	17	NR
	Washington	18	NR

The names in parenthesis are the singular voters putting a team so far away from their mean that it sticks out. Notice anything? Maybe one name in particular? Yeah, Koki Riley.

Here’s Koki’s entire ballot:

Week 8
1	Miami
2	Indiana
3	Texas A&M
4	Ohio State
5	Texas Tech
6	Alabama
7	Georgia
8	Notre Dame
9	Oregon
10	Oklahoma
11	Tennessee
12	Ole Miss
13	USC
14	Missouri
15	LSU
16	South Florida
17	Texas
18	Mississippi State
19	Utah
20	Illinois
21	Michigan
22	Nebraska
23	Cincinnati
24	Vanderbilt
25	Georgia Tech
NR	BYU
NR	Virginia
NR	Memphis

I think Ole Miss at 12 is low, Texas and Miss State in the ranking at all is high, Michigan should be booted, and Georgia Tech has played much better than a 25th ranking.

But, overall, its one ballot out of 66.

Extreme Voting

So, if Koki is averaging 10 “extreme” picks (5+ spots from actual as defined by collegepolltracker), how can he continue to be allowed to be a picker? Well, it’s actually up to Louisiana. You see, each state gets to pick their AP pollsters via approximately 1 AP pollster per 3 FBS schools in the state. So, blame LA’s media for such a varied ranking.

Out of all 66 ballots, only two have more than 7 extreme picks: Koki and Stephen Means. The rest average 2.7 extreme picks. But is that what we really want? Because that means one of two things. Either, the outliers with 5+ extremes are throwing off the rest of the country’s consensus. Or, the 10 ballots with 0 extreme picks just understand how their peers think and we’re getting in the groupthink territory.

I lean towards the latter. Otherwise, how do you reconcile 10 SEC teams in the top 25? How do you reconcile leaving out 5-1 Washington in favor of 4-2 Texas? Where’s the love for 6-0 Navy? If Ole Miss is really the 5th best team in the country, wouldn’t a road loss to the Rebs mean Tulane deserves to be in discussion for top 25 territory?

Give me more Koki and Stephen ballots that throw the book out the window. (Means’ ballot only had 8 SEC teams on it) Give me a compilation of overall team efficiency, football power index (unfairly biased based on preseason rankings), strength of record, and game control. Well, okay. Here ya go.

Presenting the first F2P Compilation Rankings!

Week 8
1	OSU (2.25)
2	Indiana (2.5)
3	Texas Tech (4.5)
4	Miami (4.5)
5	Alabama (6)
6	Oregon (7)
7	Texas A&M (8.25)
8	Georgia (10.5)
9	Notre Dame (12)
10	Vanderbilt (12.75)
11	USC (14)
12	BYU (14.75)
13	Tennessee (14.75)
14	Utah (15.25)
15	Oklahoma (15.5)
16	LSU (16.25)
17	Ole Miss (16.5)
18	Texas (19.5)
19	Illinois (20.5)
20	Cincinnati (21.5)
21	Michigan (21.5)
22	Missouri (23.5)
23	Washington (25.5)
24	South Florida (25.75)
25	Auburn (26.75)
NR	Georgia Tech (27.5)
NR	Virginia (28)
NR	Nebraska (30)
	Determined by Eff+FPI+SOR+GC/4

If you drop the bias of FPI, the top 5 stay the same but Georgia Tech and Virginia trade out with Washington and Auburn. And, of note, Cincinnati jumps 7 spots to 13. And why exactly should we drop FPI? Just one data point proves the preseason bias: 3-3 Penn State is still ranked #21.

So, does the AP Poll matter? You decide. But for me, I’ll continue using my own methods, like Mr. Gould of ole.

1. https://historicimages.com/products/rsa65205?srsltid=AfmBOorBrqaYjgF5DyDL8X6LFuolboKrfvC9–oxNao5IPMBGDwcjQe- ↩︎