2011 BCS Rankings: A Comparison of Major Ranking Services vs. SportsMeasures

Once again, the annual argument about the validity of the Bowl Championship Series (BCS) rankings has begun. I am happy to lead the charge against this assault on the intelligence of the American sporting public.

My public rantings have been raged in this space many times before and will continue. The BCS is as unscientific, invalid, and therefore, inherently unfair to everyone involved, as anything foisted upon the sporting world.

It combines subjective and biased sportswriters’ and coaches’ opinions, flawed statistical analysis and deception to create the illusion of accurate rankings. It is anything but that.

OK, so that’s out of the way. Let’s compare the AP Poll (no longer part of the BCS equation), the USA Today Poll, the BCS rankings to the only objective measurement methodology available anywhere.

A short refresher for those of you unfamiliar with our methodology. We use only games results data to create our measures. That means there is no subjectivity or bias of any kind in our measures.

We make no assumptions about the distribution of the data like the Bayesian method does. We let the actual data do the talking and that is what you see reflected in the table below.

SportsMeasures rankings and measures are a pure reflection of the wins and losses of all the teams at all levels of college football. If you disagree with anything in this comparison chart, then you’re disagreeing with the actual results of the games on the field. That statement reflects the absolute independent nature of our methodology.

SM ranking	Team name	measure	error	AP Top 25	USA Today	BCS	wins	losses
1	Alabama	97.30	6.10	2	3	2	7	0
2	LSU	95.94	6.24	1	2	1	7	0
3	Boise St	95.00	6.50	5	7	5	6	0
4	Oklahoma	89.19	6.19	3	1	3	6	0
5	Oklahoma St	89.03	5.26	6	6	4	6	0
6	Wisconsin	88.81	7.13	4	4	6	6	0
7	Stanford	86.62	8.79	7	5	8	6	0
8	SMU	84.90	6.09	29	29	NR	5	1
9	Clemson	84.72	5.42	8	8	7	7	0
10	Oregon	84.69	6.41	9	8	10	5	1
11	Virginia Tech	82.91	5.77	16	14	12	6	1
12	Houston	81.45	8.55	21	20	19	6	0
13	Kansas St	80.96	6.39	12	16	11	6	0
14	Michigan	80.63	6.14	18	17	18	6	1
15	West Virginia	80.31	6.05	11	14	15	5	1
16	Texas A&M	80.21	5.62	17	18	17	4	2
17	Arkansas	80.14	5.36	10	10	9	5	1
18	Arizona St	79.20	5.12	24	25	NR	5	2
19	Nebraska	78.96	6.11	13	11	13	5	1
20	Penn State	78.76	5.97	26	22	21	6	1
21	Michigan St	78.75	5.67	15	13	16	5	1
22	Temple	78.67	5.61	NR	35	NR	5	2
23	TCU	78.57	5.38	NR	37	NR	4	2
24	South Carolina	78.38	5.72	14	12	14	6	1
25	Toledo	78.16	5.17				4	3
26	Notre Dame	77.28	5.81	27	27	NR	4	2
27	Georgia	76.30	5.51	24	26	NR	5	2
28	Rutgers	75.99	6.21	35	28	NR	5	1
29	Washington	75.37	6.76	22	24	25	5	1
30	Texas Tech	75.20	5.68				4	2
31	Miami FL	74.82	5.83				3	3
32	Illinois	74.47	5.89	23	21	23	6	1
33	Georgia Tech	73.92	5.60	20	19	22	6	1
34	Southern Miss	73.76	5.82	32	31	NR	5	1
35	North Carolina	73.51	5.53	NR	33	NR	5	2
36	Utah	72.86	6.58				3	3
37	Missouri	72.66	6.86				3	3
38	Baylor	72.11	6.55	30	34	NR	4	2
39	Texas	71.92	6.57	31	30	24	4	2
40	Florida St	71.61	6.43				3	3
41	Southern Cal	71.60	6.79	28	NR	NR	5	1
42	South Florida	71.47	5.43				4	2
43	Auburn	70.90	6.25	19	23	20	5	2
44	Ohio State	70.01	5.94				4	3
45	Iowa	69.13	7.19				4	2
46	Louisiana-Lafayette	68.23	10.12				6	1
47	Cincinnati	67.89	6.93	33	32		5	1
48	San Diego St	67.75	7.05				4	2
49	Arkansas St	67.73	8.52				4	2
50	Florida	66.04	8.68				4	3
51	Nevada	65.83	6.56				3	3
52	Tennessee	64.43	6.92				3	3
53	Western Michigan	64.34	6.12				4	3
54	Virginia	62.16	6.48				4	2
55	Mississippi St	62.12	10.71				3	4
56	UTEP	62.03	8.49				3	3
57	Hawai`i	62.03	5.68				3	3
58	Vanderbilt	61.51	6.61				3	3
59	Tulsa	61.13	13.70				3	3
60	Wake Forest	60.92	7.45	34			4	2
61	Central Florida	60.84	6.52				3	3
62	North Carolina St	59.82	7.05				3	3
63	Eastern Michigan	59.73	6.47				4	3
64	Washington St	59.54	7.01				3	3
65	California	59.43	7.86				3	3
66	Pittsburgh	59.05	6.84				3	4
67	Utah St	59.02	6.03				2	4
68	Northern Illinois	58.84	6.62				4	3
69	Florida Intl	58.12	7.43				4	2
70	Ohio U	58.06	5.68				4	3
71	Air Force	56.83	8.01				3	3
72	Syracuse	56.08	8.03				4	2
73	Connecticut	56.03	6.15				3	4
74	Navy	55.58	8.10				2	4
75	UCLA	55.53	7.73				3	3
76	Brigham Young	54.10	7.68				5	2
77	Maryland	54.01	7.55				2	4
78	Bowling Green	53.09	6.85				3	4
79	Wyoming	52.59	7.83				4	2
80	Iowa St	51.87	6.00				3	3
81	Louisville	51.72	7.42				2	4
82	Northwestern	51.33	7.09				2	4
83	Ball St	50.88	6.52				4	3
84	Marshall	50.56	6.30				3	4
85	Louisiana-Monroe	50.55	8.28				2	4
86	Louisiana Tech	49.80	7.90				2	4
87	San Jose St	48.81	7.95				3	4
88	Purdue	47.02	8.83				3	3
89	Rice	45.32	9.12				2	4
90	Fresno St	44.56	6.72				3	4
91	Duke	44.29	7.04				3	3
92	New Mexico St	43.96	9.20				3	3
93	Buffalo	43.85	6.09				2	5
94	Central Michigan	42.35	5.61				2	5
95	Kansas	41.26	6.79				2	4
96	Colorado	40.51	5.87				1	6
97	East Carolina	40.24	11.33				2	4
98	North Texas	39.64	7.80				2	5
99	Mississippi	39.47	5.95				2	4
100	Army	38.90	7.09				2	4
101	Colorado St	38.12	6.18				3	3
102	Boston College	37.84	6.67				1	5
103	Kentucky	37.50	6.66				2	4
104	Miami OH	36.57	6.51				2	4
105	Indiana	36.13	6.85				1	6
106	Troy	35.27	8.38				2	4
107	Idaho	34.36	6.13				1	6
108	Arizona	33.09	8.16				1	5
109	Oregon St	32.19	7.72				1	5
110	Kent St	31.54	6.25				1	6
111	UNLV	30.79	6.27				1	5
112	Minnesota	30.77	6.06				1	5
113	Western Kentucky	30.03	6.58				2	4
114	Akron	28.30	13.41				1	5
115	Tulane	26.03	8.19				2	5
116	Middle Tennessee St	25.99	6.46				1	4
117	Florida Atlantic	22.03	8.85				0	6
118	Alabama-Birmingham	17.83	8.08				0	6
119	Memphis	16.70	6.11				1	6
120	New Mexico	13.41	10.55				0	6

Our method takes into account the whole data matrix—who played who played who played who, etc. That is virtually impossible for any voting sportswriter or coach to do. These measures also allow us to create realistic strength of schedule calculations as well as fair and objective conference strength measures. See our other article on that subject—search by author name (Patrick B. Fisher).

The most glaring omission in the major rankings is Southern Methodist University. SMU has a 5-1 record, with their lone loss coming at (#16) Texas A&M, which has the toughest schedule (SoS) so far.  Our SoS calculations are based on games played so far.

Their victories have been over UTEP (No. 56), Northwestern St (No. 44 in FCS; would be No. 84 in FBS), at Memphis (No. 119), at TCU (No. 23) and Central Florida (No. 61). Each win has come by a sizable margin and the TCU game was on the road. Houston (No. 12), 6-0 so far, is ranked 21st in the AP Poll, No. 19 in the BCS.

Their SoS is 111th of 120 and are getting more respect and attention than SMU. Houston’s weak schedule is reflected partly in their high error of measurement (8.55). Compare that to Oklahoma St. with a 6-0 record and an error of 5.26 (No. 20, SoS). 

Boise State is once again not getting all the respect they deserve. They’ve responded to the critics by adding a road game at Georgia to their schedule, which they promptly went out and won. The BCS has them at No. 5, but SportsMeasures has them fighting for a spot in the BCS Championship. 

Other large discrepancies include No. 17 Arkansas (No. 9, BCS), No. 18 Arizona St. (unranked BCS) and, No. 24 South Carolina (No. 14, BCS). These examples highlight the problems with the BCS method. Another commonly argued (and valid) point is the syndrome of beginning the season highly ranked and staying there because you won, not because your schedule strength or performance warranted staying there.

Since SportsMeasures makes no assumptions, every team starts at the same place at the beginning of the season and their performance dictates their ranking as the season goes. 

The other big issue with the BCS is how they hope and expect there to be very few (read: 2) undefeated teams so that the national championship game teams are easily paired.

Currently, there are 10 undefeated teams in the FBS, which opens the floodgates for the other undefeated teams to argue their case for being in the BCS championship game. Because our method is subjected to scientific rigor, it will stand up to any scrutiny by anyone. 

I will be writing shorter follow-ups for the duration of the season focusing on the Top 25.