2010 Compu-Picks Ratings - College

The Compu-Picks model rates the top and bottom teams in college football after the end of the regular season

One of the things that I like to do with this model is come up with interesting bits of information about the season. Rankings, schedule strength, which teams got better and which got worse, which teams had an especially strong home-field edge and which did relatively better on the road (yes there WERE a few, as there usually are), and more. I hope you enjoy the list and commentary, and if you have some ideas for things you'd like to see in future versions of this, please let me know. I'm always on the lookout for good new ideas, either to incorporate into the model itself or into the exhibits I provide based on the model.

Please remember that these ratings are used for the sole purpose of predicting future results. There are no bonus points for being undefeated, or being from a top conference, or playing X number of top 25 teams, or beating a team with a similar ranking head to head, or any single other sort of adjustment that many people use when making their rankings. Schedule strength counts a lot, and margin counts a lot. A close loss against a great team gives a better rating than a close win against a lousy team. Many people object to this, but in terms of predictive value, I consider it the only way to go. These rankings look different than most human polls, and for that matter they look different from most computer polls. That's OK. It's supposed to look different, both because its goals are different, and because it operates differently from other models. The only standard I use to judge it is how accurate its picks are, and so far this year it's been pretty solid, going 111-93 (54.4%) ATS.

Team Rankings - 1 to 120

1 Oregon	31 Iowa	61 Brigham Young	91 Rutgers
2 Stanford	32 Florida	62 Fresno State	92 Texas-El Paso
3 Texas Christian	33 Hawaii	63 Mississippi	93 Vanderbilt
4 Boise State	34 Air Force	64 Colorado	94 Alabama-Birmingham
5 Auburn	35 Georgia	65 South Florida	95 Utah State
6 Virginia Tech	36 Clemson	66 Iowa State	96 Arkansas State
7 Alabama	37 Maryland	67 Southern Methodist	97 Kansas
8 Ohio State	38 California	68 Connecticut	98 Wyoming
9 Arkansas	39 Pittsburgh	69 Syracuse	99 Marshall
10 Oklahoma	40 North Carolina	70 Houston	100 Central Michigan
11 Missouri	41 San Diego State	71 Toledo	101 Louisiana-Monroe
12 Nebraska	42 Illinois	72 Temple	102 Colorado State
13 Oklahoma State	43 Navy	73 East Carolina	103 Nevada-Las Vegas
14 South Carolina	44 Kansas State	74 Northwestern	104 Kent
15 Wisconsin	45 Central Florida	75 Army	105 North Texas
16 Louisiana State	46 Baylor	76 Louisiana Tech	106 Rice
17 Nevada	47 Washington	77 Idaho	107 Tulane
18 Florida State	48 Texas Tech	78 Cincinnati	108 Middle Tennessee State
19 Arizona	49 Michigan	79 Troy State	109 San Jose State
20 Texas A&M	50 Penn State	80 Washington State	110 Florida Atlantic
21 Southern California	51 Northern Illinois	81 Minnesota	111 Louisiana-Lafayette
22 North Carolina State	52 Texas	82 Ohio	112 Bowling Green State
23 Utah	53 Tennessee	83 Western Michigan	113 Western Kentucky
24 Arizona State	54 Boston College	84 Purdue	114 Ball State
25 Michigan State	55 Southern Mississippi	85 Wake Forest	115 New Mexico
26 West Virginia	56 Tulsa	86 Virginia	116 New Mexico State
27 Miami (Florida)	57 UCLA	87 Duke	117 Memphis
28 Notre Dame	58 Georgia Tech	88 Florida International	118 Eastern Michigan
29 Mississippi State	59 Louisville	89 Miami (Ohio)	119 Buffalo
30 Oregon State	60 Kentucky	90 Indiana	120 Akron

Detailed Top 30 Team Ratings

Rank	BCS Rank	Team	League	Score	Schedule Rank
1	2	Oregon	Pac-10	0.93	6
2	4	Stanford	Pac-10	0.88	3
3	3	Texas Christian	Mountain West	0.83	42
4	10	Boise State	WAC	0.81	38
5	1	Auburn	SEC	0.75	8
6	13	Virginia Tech	ACC	0.69	34
7	16	Alabama	SEC	0.67	14
8	6	Ohio State	Big Ten	0.65	57
9	8	Arkansas	SEC	0.59	15
10	7	Oklahoma	Big 12	0.59	12
11	12	Missouri	Big 12	0.56	23
12	18	Nebraska	Big 12	0.56	27
13	14	Oklahoma State	Big 12	0.53	39
14	20	South Carolina	SEC	0.52	10
15	5	Wisconsin	Big Ten	0.51	51
16	11	Louisiana State	SEC	0.50	17
17	15	Nevada	WAC	0.50	66
18	23	Florida State	ACC	0.50	16
19	NR	Arizona	Pac-10	0.47	5
20	17	Texas A&M	Big 12	0.44	20
21	NR	Southern California	Pac-10	0.42	11
22	NR	North Carolina State	ACC	0.38	25
23	19	Utah	Mountain West	0.38	62
24	NR	Arizona State	Pac-10	0.37	2
25	9	Michigan State	Big Ten	0.35	54
26	22	West Virginia	Big East	0.35	67
27	NR	Miami (Florida)	ACC	0.34	19
28	NR	Notre Dame	Indep	0.33	30
29	21	Mississippi State	SEC	0.32	21
30	NR	Oregon State	Pac-10	0.31	1

Schedule Strength - Top and Bottom 30

Team	Sagarin SOS Ranking	NCAA SOS Ranking	FEI SOS Ranking		Team
1 Oregon State	1	12	12		91 Army
2 Arizona State	6	50	21		92 Tulsa
3 Stanford	10	85	62		93 Marshall
4 UCLA	4	74	42		94 Tulane
5 Arizona	12	57	28		95 Central Michigan
6 Oregon	19	150	73		96 Toledo
7 Washington	2	37	24		97 Louisiana-Monroe
8 Auburn	15	3	13		98 Alabama-Birmingham
9 California	5	35	41		99 Central Florida
10 South Carolina	14	3	1		100 Rice
11 Southern California	7	71	40		101 Florida Atlantic
12 Oklahoma	8	10	25		102 Texas-El Paso
13 Washington State	3	28	18		103 Troy State
14 Alabama	17	20	8		104 Louisiana-Lafayette
15 Arkansas	21	9	2		105 Temple
16 Florida State	25	33	11		106 Eastern Michigan
17 Louisiana State	24	6	4		107 Miami (Ohio)
18 Florida	16	5	23		108 Arkansas State
19 Miami (Florida)	32	15	14		109 Bowling Green State
20 Texas A&M	11	2	34		110 Florida International
21 Mississippi State	29		6		111 Buffalo
22 Wake Forest	31		5		112 Northern Illinois
23 Missouri	26		58		113 Kent
24 Clemson	37		7		114 Akron
25 North Carolina State	45		39		115 Western Michigan
26 Iowa State	13		46		116 North Texas
27 Nebraska	28		48		117 Western Kentucky
28 North Carolina	34		17		118 Ohio
29 Georgia	35		10		119 Ball State
30 Notre Dame	23		47		120 Middle Tennessee State

League Ratings

League	Rating	OOC Schedule Rating	Home/Away/Neutral Splits	OOC vs Top 10	OOC vs 11-20	OOC vs 21-40	OOC vs 41-60	OOC vs 61-80	OOC vs 81-100	OOC vs Bottom 20
Pac-10	0.35	0.16	10 / 14 / 0	0 - 2	0 - 4	3 - 1	3 - 1	4 - 2	3 - 0	1 - 0
SEC	0.30	-0.16	26 / 9 / 2	0 - 1	1 - 1	4 - 0	4 - 0	4 - 3	6 - 1	12 - 0
Big 12	0.22	-0.07	26 / 11 / 3	0 - 2	1 - 0	3 - 3	8 - 2	7 - 0	4 - 0	10 - 0
ACC	0.16	0.07	18 / 15 / 2	0 - 6	0 - 2	2 - 4	3 - 1	5 - 3	4 - 1	4 - 0
Indep	0.12	-0.06	15 / 12 / 1	0 - 1	0 - 0	3 - 5	1 - 2	3 - 1	6 - 2	4 - 0
Big Ten	0.12	-0.14	23 / 10 / 1	0 - 1	0 - 2	4 - 2	1 - 1	3 - 2	6 - 0	12 - 0
Big East	-0.01	-0.08	16 / 15 / 0	0 - 1	0 - 1	2 - 7	0 - 4	1 - 2	6 - 0	6 - 1
WAC	-0.04	0.02	14 / 18 / 0	1 - 2	0 - 3	2 - 2	1 - 4	5 - 2	2 - 2	5 - 1
Mountain West	-0.04	0.09	14 / 18 / 1	0 - 3	0 - 6	2 - 3	3 - 2	5 - 2	1 - 3	2 - 1
C-USA	-0.24	0.00	20 / 22 / 0	0 - 4	0 - 2	2 - 6	0 - 11	3 - 3	3 - 0	6 - 2
MAC	-0.43	0.00	11 / 31 / 0	0 - 4	0 - 2	0 - 4	0 - 10	2 - 9	3 - 6	2 - 0
Sun Belt	-0.47	0.04	8 / 26 / 0	0 - 3	0 - 6	0 - 5	0 - 6	0 - 4	1 - 6	1 - 2
League	Rating	OOC Schedule Rating	OOC vs AQ's	OOC vs Non-AQ's	OOC vs Pac-10 / SEC	OOC vs Big 12 / ACC / Big Ten	OOC vs Big East	OOC vs MWC / WAC	OOC vs CUSA	OOC vs MAC / Sun Belt
Pac-10	0.35	0.16	10 - 5	4 - 5	1 - 0	6 - 4	2 - 0	2 - 4	1 - 1	1 - 0
SEC	0.30	-0.16	10 - 6	21 - 0	0 - 1	7 - 4	3 - 1	2 - 0	8 - 0	11 - 0
Big 12	0.22	-0.07	8 - 4	25 - 3	4 - 3	3 - 1	1 - 0	10 - 2	6 - 1	9 - 0
ACC	0.16	0.07	6 - 13	12 - 4	2 - 7	0 - 3	4 - 2	1 - 1	3 - 1	6 - 0
Indep	0.12	-0.06	6 - 6	11 - 5	1 - 1	4 - 4	1 - 1	2 - 3	3 - 1	6 - 1
Big Ten	0.12	-0.14	7 - 5	19 - 3	2 - 3	2 - 1	1 - 0	2 - 1	2 - 0	15 - 2
Big East	-0.01	-0.08	3 - 12	12 - 4	1 - 5	2 - 6	0 - 0	1 - 2	2 - 1	8 - 1
WAC	-0.04	0.02	5 - 9	11 - 7	2 - 3	2 - 6	1 - 0	8 - 4	0 - 2	2 - 0
Mountain West	-0.04	0.09	5 - 10	8 - 10	2 - 1	2 - 7	1 - 1	4 - 8	1 - 1	1 - 1
C-USA	-0.24	0.00	5 - 22	9 - 6	1 - 9	2 - 11	1 - 2	3 - 1	0 - 0	6 - 2
MAC	-0.43	0.00	3 - 25	4 - 10	0 - 4	2 - 17	1 - 3	1 - 3	1 - 3	1 - 1
Sun Belt	-0.47	0.04	0 - 26	2 - 6	0 - 8	0 - 13	0 - 5	0 - 0	1 - 3	1 - 1

Top and Bottom 25 Home-Road Splits

This describes how much better a team played at home vs on the road. It's well worth noting that it's neither good nor bad to be high or low on this list; it's good to have an especially strong home-field edge, with a loud crowd and possibly a weather edge, but it's also good to be able to play well on the road. Consequently, you will tend to see most really good teams around the middle of the list, because they do have a strong home-field edge but also are very capable of playing well on the road.

1 Connecticut	96 Miami (Florida)
2 Iowa	97 Temple
3 California	98 Nebraska
4 Nevada-Las Vegas	99 North Texas
5 Colorado	100 Arizona
6 Northern Illinois	101 Arkansas
7 Rice	102 Vanderbilt
8 Colorado State	103 Toledo
9 Marshall	104 Utah State
10 Georgia	105 Stanford
11 Arizona State	106 Louisiana-Lafayette
12 Kentucky	107 Southern California
13 Louisiana-Monroe	108 Tulane
14 Arkansas State	109 Utah
15 Wyoming	110 Troy State
16 Kent	111 Buffalo
17 North Carolina State	112 Virginia Tech
18 Middle Tennessee State	113 Washington State
19 Brigham Young	114 Florida International
20 Southern Methodist	115 Ball State
21 Iowa State	116 Central Michigan
22 New Mexico State	117 Western Kentucky
23 East Carolina	118 Texas
24 UCLA	119 Navy
25 Oklahoma	120 Syracuse

Top and Bottom 25 Most / Least Improved

This describes how much better a team played at the end of the year compared to how they played early on. Since the system is rating how good a team is now, it's better to be higher than lower on this list, though of course the best result is to play at a really high level the whole year long.

1 Connecticut	96 Utah State
2 West Virginia	97 Michigan
3 Washington State	98 Southern Methodist
4 Auburn	99 Illinois
5 Brigham Young	100 California
6 Louisville	101 East Carolina
7 Boston College	102 Nebraska
8 Washington	103 Florida
9 Notre Dame	104 Kansas
10 Maryland	105 Central Michigan
11 South Florida	106 Missouri
12 Duke	107 Eastern Michigan
13 Syracuse	108 Ball State
14 Pittsburgh	109 Bowling Green State
15 Stanford	110 Northwestern
16 Arkansas	111 Temple
17 Georgia	112 Colorado
18 Navy	113 Baylor
19 Hawaii	114 Texas
20 Oregon	115 Kansas State
21 Texas-El Paso	116 Florida Atlantic
22 Southern Mississippi	117 Iowa
23 Virginia Tech	118 Iowa State
24 Wisconsin	119 Vanderbilt
25 New Mexico	120 Buffalo

Top Forty wins of the Year

Game Rank	Team	Opponent	Location	Score
1	Oregon	Stanford	HOME	52 - 31
2	Auburn	South Carolina	NEUTRAL	56 - 17
3	Texas Christian	Utah	AWAY	47 - 7
4	Stanford	California	AWAY	48 - 14
5	Arkansas	South Carolina	AWAY	41 - 20
6	Missouri	Texas A&M	AWAY	30 - 9
7	Stanford	Oregon State	HOME	38 - 0
8	Auburn	Arkansas	HOME	65 - 43
9	Stanford	Washington	AWAY	41 - 0
10	Oklahoma	Florida State	HOME	47 - 17
11	Florida State	Miami (Florida)	AWAY	45 - 17
12	Nebraska	Kansas State	AWAY	48 - 13
13	Boise State	Hawaii	HOME	42 - 7
14	Boise State	Fresno State	HOME	51 - 0
15	Oregon	Southern California	AWAY	53 - 32
16	Stanford	UCLA	AWAY	35 - 0
17	Stanford	Arizona	HOME	42 - 17
18	Utah	Iowa State	AWAY	68 - 27
19	Stanford	Notre Dame	AWAY	37 - 14
20	Texas Christian	Air Force	HOME	38 - 7
21	Miami (Florida)	Pittsburgh	AWAY	31 - 3
22	South Carolina	Clemson	AWAY	29 - 7
23	Oregon State	Southern California	HOME	36 - 7
24	South Carolina	Florida	AWAY	36 - 14
25	California	Arizona State	HOME	50 - 17
26	Oregon	Oregon State	AWAY	37 - 20
27	Alabama	Tennessee	AWAY	41 - 10
28	Oregon	Arizona	HOME	48 - 29
29	Boise State	Toledo	HOME	57 - 14
30	Notre Dame	Utah	HOME	28 - 3
31	Oregon	UCLA	HOME	60 - 13
32	Wisconsin	Ohio State	HOME	31 - 18
33	Oklahoma	Iowa State	HOME	52 - 0
34	Texas Christian	Baylor	HOME	45 - 10
35	Oregon	Washington	HOME	53 - 16
36	Oregon	Arizona State	AWAY	42 - 31
37	Alabama	Florida	HOME	31 - 6
38	Iowa	Michigan State	HOME	37 - 6
39	West Virginia	Pittsburgh	AWAY	35 - 10
40	South Carolina	Alabama	HOME	35 - 21

Bottom Forty losses of the Year

Game Rank	Team	Opponent	Location	Score
1327	Akron	Ohio	AWAY	10 - 38
1328	Miami (Ohio)	Ohio	HOME	13 - 34
1329	Buffalo	Miami (Ohio)	HOME	9 - 21
1330	Western Kentucky	Louisiana-Monroe	HOME	30 - 35
1331	Ball State	Kent	AWAY	14 - 33
1332	Vanderbilt	Wake Forest	HOME	13 - 34
1333	Middle Tennessee State	Memphis	AWAY	17 - 24
1334	Louisiana-Lafayette	Florida International	HOME	17 - 38
1335	Middle Tennessee State	Troy State	HOME	13 - 42
1336	Alabama-Birmingham	Marshall	HOME	17 - 31
1337	Wyoming	Nevada-Las Vegas	AWAY	16 - 42
1338	North Texas	Florida International	HOME	10 - 34
1339	Central Michigan	Ball State	HOME	17 - 31
1340	Kent	Western Michigan	AWAY	3 - 38
1341	New Mexico State	Texas-El Paso	AWAY	10 - 42
1342	Florida Atlantic	Middle Tennessee State	AWAY	14 - 38
1343	Buffalo	Eastern Michigan	HOME	17 - 21
1344	Louisiana-Monroe	Middle Tennessee State	AWAY	10 - 38
1345	New Mexico State	Kansas	AWAY	16 - 42
1346	Buffalo	Akron	AWAY	14 - 22
1347	Memphis	Houston	HOME	17 - 56
1348	New Mexico	Texas-El Paso	HOME	20 - 38
1349	Nevada-Las Vegas	Colorado State	AWAY	10 - 43
1350	Marshall	Bowling Green State	AWAY	28 - 44
1351	Eastern Michigan	Northern Illinois	HOME	3 - 71
1352	New Mexico	Colorado State	AWAY	14 - 38
1353	Eastern Michigan	Toledo	HOME	7 - 42
1354	Florida Atlantic	Troy State	HOME	7 - 44
1355	Akron	Temple	AWAY	0 - 30
1356	Louisiana-Lafayette	Middle Tennessee State	HOME	14 - 34
1357	New Mexico	Nevada-Las Vegas	AWAY	10 - 45
1358	Bowling Green State	Western Michigan	HOME	7 - 41
1359	Colorado State	Wyoming	AWAY	0 - 44
1360	Eastern Michigan	Central Michigan	HOME	14 - 52
1361	Akron	Western Michigan	HOME	10 - 56
1362	Buffalo	Temple	HOME	0 - 42
1363	Eastern Michigan	Vanderbilt	AWAY	6 - 52
1364	Bowling Green State	Kent	HOME	6 - 30
1365	Buffalo	Ball State	HOME	3 - 20
1366	Western Kentucky	North Texas	HOME	6 - 33

25 Biggest Head-Scratchers

These were the games that, after all of the games have been played, stick out as especially weird. For instance, if a 20-point underdog wins by 10 early in the year, but then they go 8-4 and the team they upset goes 4-8, well, that's not really so weird. But if a 14-point favorite wins by 35, then ends up 4-8 while the team they beat goes 8-4, then that would stick out as a weird result despite it making sense at the time.

Game Rank	Team	Opponent	Location	Score
1	Washington State	Oregon State	AWAY	31 - 14
2	Vanderbilt	Mississippi	AWAY	28 - 14
3	Texas	Nebraska	AWAY	20 - 13
4	Houston	Southern Methodist	AWAY	45 - 20
5	Syracuse	Cincinnati	AWAY	31 - 7
6	Western Kentucky	Louisiana-Lafayette	AWAY	54 - 21
7	Ohio	Miami (Ohio)	AWAY	34 - 13
8	Miami (Florida)	Pittsburgh	AWAY	31 - 3
9	Navy	Notre Dame	NEUTRAL	35 - 17
10	Kent	Bowling Green State	AWAY	30 - 6
11	Missouri	Texas A&M	AWAY	30 - 9
12	North Texas	Western Kentucky	AWAY	33 - 6
13	Utah	Iowa State	AWAY	68 - 27
14	Utah State	Brigham Young	HOME	31 - 16
15	Arkansas	South Carolina	AWAY	41 - 20
16	Syracuse	West Virginia	AWAY	19 - 14
17	Tulane	Texas-El Paso	AWAY	34 - 24
18	West Virginia	Pittsburgh	AWAY	35 - 10
19	Rice	East Carolina	HOME	62 - 38
20	Colorado	Hawaii	HOME	31 - 13
21	Marshall	Alabama-Birmingham	AWAY	31 - 17
22	Florida State	Miami (Florida)	AWAY	45 - 17
23	Kent	Ohio	HOME	28 - 6
24	California	Arizona State	HOME	50 - 17
25	Miami (Ohio)	Temple	HOME	23 - 3

Some thoughts on the results:

1) As part of my regular season analysis, I'm providing some schedule strength numbers reflecting the top and bottom 30 schedules according to compu-picks, as well as comparisons for the top 30 schedules to some other sources of schedule ranking: Sagarin, NCAA, and FEI (from football outsiders). The numbers from these sites (except Sagarin) are all from before the Army-Navy game, so it is likely that they will be updated by the time you look at them, though I would be surprised to see material differences. One thing about the compu-picks schedule numbers that is important to note is that they do not include 1-AA games (these games are counted as byes, and do not affect schedule, margin, etc; it's as if they never happened). So in that regard, the Sagarin numbers (which do include those games) are probably a bit more accurate in terms of rating how difficult a team's entire year-long schedule actually was. Of course, the Sagarin numbers also correlate pretty closely to the compu-picks numbers, so I'm not really sweating the difference. If you could somehow back out the AA games, my guess is that his numbers would correlate extremely closely to these.

The most obvious thing you can see from looking at the table is how wildly different compu-picks' numbers are from both the NCAA and FEI numbers (especially the NCAA numbers). The NCAA numbers are calculating using simple W/L records as a proxy for schedule strength. This approach is, of course, ridiculous. Anyone who disagrees is simply wrong. There's really no argument here. If you believe the NCAA numbers, then you believe (for starters - there's plenty more dumb numbers here):

Villanova had a tougher schedule than Oregon St, Alabama, and was light-years tougher than UCLA and Tennessee
Appalaichan St had a tougher schedule than UCLA, Tennessee, and was light-years tougher than Virginia, Vandy and Oregon
UNLV had a tougher schedule than Washington, Iowa, Mississippi St, Texas Tech, Clemson, and Colorado

That's completely absurd even without bothering to factor in home-field advantage (which, of course, the NCAA numbers don't bother to do). I refuse to take seriously anyone who cites these numbers are being reasonable, much less correct, and you should too. That's why I didn't bother to fill in the full list; just looking at 20 of them it's obvious enough that the numbers are garbage, so I didn't feel like bothering to do 30 (plus the .pdf format the NCAA puts the numbers in makes pulling the results an annoying manual process, whereas it was much easier for the other two sources).

The most significant difference between the FEI schedule numbers and the compu-picks numbers is that FEI is defining schedule strength not as the average difficulty of a game, but the odds that an elite team would go undefeated against the schedule. The effect here is that higher difficulty games are strongly weighted, while there really isn't much of a difference between, say, playing the 80th rated team and the 120th rated team. That's not an entirely unreasoanble approach, but it's definitely measuring something much different than what compu-picks measures. Since he doesn't publish "average difficulty" numbers, it's impossible for me to say how much of the differences are due to his measuring something different and how much is due to differences in team ratings. My guess is that the bulk majority is due to what he's measuring... but that's just a guess.

Ultimately, I think the compu-picks schedule ratings are very good, for what they're trying to measure. I certainly accept the fact that, by throwing out AA games, it's telling an incomplete story. However, even accounting for that I believe that its numbers are generally accurate, and compare very well to just about everything out there (though again, I'm inclined to think Sagarin's schedule numbers are a bit better, mainly because he does account for the AA games).

2) Speaking of schedule strength, it's worth asking whether it really makes sense to adjust for home-field advantage. Does it really make a difference?

The answer is yes, as shown in the table below. With only a 53% winning percentage overall, it's pretty clear that home-field isn't much of a predictor in aggregate; however, when you throw out the mismatches (loosely defined here as games where one team ended up with 2 or more league wins than the other), the story is much different. When the teams ended up with the same number of league wins, the home team won slightly more than two-thirds of the time (it's a small sample size, of course, but it's still a very compelling number). When you expand the list to minor mismatches (teams ending up within one win of each other), the win rate drops, but is still close to 60%. So in answer to the question, yes I do think that it's a relevant factor that makes a lot of sense to include in this rating system. In fact, I would go so far as to say that any system which fails to account for this is flawed; certainly that includes the NCAA's ridiculous schedule numbers, but it also includes most or all of the BCS computers as well.

League	Home Team W/L (%) - overall	Home Team W/L (%) - same # wins	Home Team W/L (%) - within 1 win
Pac-10	23 - 22 (51%)	1 - 3 (25%)	11 - 5 (69%)
SEC	27 - 20 (57%)	3 - 2 (60%)	10 - 7 (59%)
Big 12	23 - 21 (52%)	4 - 2 (67%)	11 - 6 (65%)
ACC	26 - 22 (54%)	7 - 1 (88%)	12 - 8 (60%)
Indep	0 - 1 (0%)	0 - 0 (%)	0 - 1 (0%)
Big Ten	22 - 21 (51%)	4 - 1 (80%)	8 - 7 (53%)
Big East	12 - 16 (43%)	2 - 2 (50%)	3 - 9 (25%)
WAC	20 - 16 (56%)	3 - 0 (100%)	6 - 2 (75%)
Mountain West	23 - 13 (64%)	5 - 0 (100%)	9 - 1 (90%)
C-USA	29 - 19 (60%)	4 - 1 (80%)	13 - 4 (76%)
MAC	25 - 27 (48%)	1 - 2 (33%)	4 - 9 (31%)
Sun Belt	17 - 19 (47%)	2 - 3 (40%)	8 - 10 (44%)
Total	247 - 217 (53%)	36 - 17 (68%)	95 - 69 (58%)

3) After five out of the six weeks I've posted comments, at least one teams the model thought that the BCS overrated got exposed (I'm ignoring this past week's results, since there was only one game played):

After week 8: Auburn, Wisconsin, Michigan St - Michigan St loses 37-6 at Iowa
After week 9: Auburn, Utah, Oklahoma, Wisconsin, Michigan St - Utah loses 47-7 at TCU and Oklahoma loses 33-19 at Texas A&M
After week 10: Auburn, LSU, Wisconsin, Michigan St, Mississippi St - Mississippi St loses 30-10 at Alabama (Bama and MSU were only seven spots apart in the BCS before the game)
After week 11: Auburn, LSU, Wisconsin, Michigan St, Mississippi St - no teams got exposed, though LSU and Michigan St came pretty close
After week 12: Auburn, LSU, Wisconsin, Michigan St - LSU loses 31-23 at an Arkansas team rated seven spots below them (and Michigan St nearly got knocked off at a Penn St team that wasn't even close to being ranked)
After week 13: Auburn, Wisconsin, LSU, Michigan St, Mississippi St, Northern Illinois - NIU loses 26-21 to a pretty bad Miami(OH) team (though Auburn came up with a fantastic win)

Of course, there have been ones going the other way, most notably Oregon St's stunning implosion against Washington St (followed by a great win against USC). But it certainly seems like there have been more noteworthy bad showings by teams the model thought overrated than those it thought underrated.

4) League ratings have been a fairly popular source of comments, so let's talk about them again. The popular perception (as shown by a CBS columnist here), is that the Pac-10 is somehow "down". The basic reasoning is that the overall records aren't outstanding (the league's OOC was close to .500), and that the league is VERY top heavy (Oregon and Stanford have basically dominated the league). It's also been said here that the Pac-10's problem was that the league "fail(ed) to hold any momentum it might have had from the non-conference schedule"... basically, that evne though the non-conference went great, somehow the intra-league games somehow caused the league's rating to decline. This is, of course, silly.

As shown in the above tables, the Pac-10 has the BEST record nationwide against other AQ leagues. Moreover, of all the AQ leagues, they're the one which has played by far the toughest slate. They're the only AQ league which has played a majority of its 1-A games on the road; they've played the fewest by far games againt the bottom 40 teams (as rated by compu-picks). They've played precisely ONE game against the Mac and Sun Belt combined; as a contrast, the Big Ten has lost more games to those two leagues (both double-digit HOME losses to the MAC) than the Pac-10 has games against them. In fact, if you break down the OOC records by groupings of 20 (as in the above table), the only leagues that show comparable results to the Pac-10 are the SEC and Big 12, and that's before you factor in home-field advantage. It's also worth noting that the Pac-10 is one of the only leagues without a AA loss; while the model doesn't factor in those games, as humans we can do so, and it's another point in the league's favor.

Focusing on the bottom 40 for a bit, the Pac-10 doesn't have any bottom 40 OOC losses (only the Big 12 and Big Ten can say the same), and had only one game with less than a 7 point win against that group (USC's win over Virginia), as opposed to the SEC's two close calls against UAB; the Big Ten's close calls against Ark St, CMU, and Vandy; the Big 12's close call against Troy; the ACC's close call against Rutgers; and the Big East's close calls against Marshall and FIU.

Basically, the unfortunate truth is that the BCS rewards easy schedules and punishes tough ones. When a league as a whole "gimmicks up" its record (overwhelming number of home games, majority of OOC games against bad competition), it gets rewarded. When the Big Ten schedules a ridiculous 17 games against the MAC/Sun Belt (not to mention the AA games), and actually LOSES two of them, it gets rewarded for it. When the Pac-10 has nine league games, a brutal OOC slate, has an outstanding 10-5 record against other AQ's, and has a near-total lack of "bad losses" (the worst was Wazzu at SMU, hardly a MAC-level loss), it gets punished for it with the silly perception that it's "down". The inescapable conclusion is that there is NOTHING that the league could have done to be perceived as excellent this year given the schedules it had to face.

Besides the bias and laziness of most analysts that attempt to evaluate leagues, what does that mean going forward? Mainly, it means that the Pac-10 needs to gimmick up its schedules too. That means no more paycheck games on the road (see: Colorado at Ohio St coming up, as well as recent games such as Oregon St at TCU, Wazzu at Notre Dame, Auburn, Wisconsin [technically a 2:1 but I've got a bridge to sell you if you think it's likely the Badgers trek to Pullman as scheduled], etc.). That means fewer games against other AQ's (15 of the league's 28 OOC slots were against other AQ's, plus there were Oregon St's games against Boise and TCU). That means getting paycheck, 2:1 or 3:1 deals with the mid-majors (as opposed to ASU's upcoming home and home with New Mexico and insane home and home deal with UTSA, Wazzu's home and home with SMU, Washington's home and home with BYU, Oregon and Oregon St's home and homes with Boise, Arizona's home and home with New Mexico and apparent upcoming home and home with Nevada, Stanford's home and home with Navy and upcoming home and home with Army, etc.). It has to be a priority to get more home games. Home games provide a meaningful edge, and other leagues are taking advantage, while the Pac-10 clearly isn't.

Quite frankly, if Oregon St had cancelled one of the Boise/TCU road games and scheduled a home game creampuff win, they'd have been at six wins with a bowl game, instead of five wins and no bowl game. Would that have made them a better team in any way? Of course not. But in the half-assed world of league ratings analysis and human polls rating teams, it would have made them look like a better team, would have put them in the postseason (which would have helped the Pac-10's bowl partners) and would have given a nice boost to Oregon and Stanford (who came fairly close to not having a top 4 BCS ranking, which probably would have meant the Alamo Bowl for them). That's tangible value to the rest of the league, even if it would have been around a net wash or loss for Oregon St's bottom line.

I'm not sure how the league can move its schedules in this direction: incentives for extra home games, changing the bowl payout for a team from an even 1/12 to something like 2/13 (the bowl team gets a double share, everyone else gets an equal cut), or actual league mandates, but there's no question that it's imperative that the league do this. The incentive structure for gimmicking up schedules couldn't be clearer, and the if Pac-12 refuses to join the party, they're only going to hurt themselves in the process.

5) More on league ratings: I've gotten some comments from SEC people who believe that the model has them too low (although a 0.30 rating is still VERY good). The two biggest things holding back that league's rating are the % of home games and the very large number of games against 1-A cupcakes.

It's also worth noting that the SEC has the worst single OOC game rating of any AQ league by a substantial margin, Vandy's horrific 21-point home loss to a pretty bad Wake Forest team; in fact, that was the only performance by any AQ team to rate in the bottom 40 losses nationwide. It's a somewhat minor point (since it's just one game out of 37), but by itself the game probably dropped the league's rating by somewhere between 0.02 and 0.05, which turned out to be a pretty big chunk of the difference between the SEC and the Pac-10.

Overall, while I certainly believe that the model's outputs are reasonable and defensible, it may be true that it's over-penalizing the SEC for the cupcake games. That's going to be high on the list of things to dig into further over the off-season. My suspicion is that it won't result in a material change (and perhaps not any change at all), but I could very well be wrong.

I've also gotten some comments from Big Ten people who believe the model has them too low. I don't buy it. Multiple MAC losses, few quality wins (the best ones were against Miami, ASU and Notre Dame [twice] ), a gimmick schedule with half of the OOC games against bottom 40 teams and over 2/3 of the games at home, only a 7-5 record vs AQ's... it just wasn't a good year for the league. It's reasonable to argue them over the ACC (again, it's possible that the system is over-penalizing for the cupcake games), but that's it. They were a clear few steps below the top three leagues (Pac-10, SEC, Big 12) in 2010. It's certainly possible that they have a good bowl run, which would change the numbers, but for now, the resume just isn't there.

6) The following teams are ranked materially higher by the model than the BCS: Oregon, Stanford, Boise St, Virginia Tech, Alabama, Nebraska, South Carolina, Florida St, Arizona, USC, NC St, Arizona St.

Oregon has the #1 win, a number of other quality performances, and is undefeated (and generally dominant) against a very tough schedule.

Stanford has a boatload of high-quality performances, consistently blowing teams out even against an extremely tough schedule. They have consistently played at an extremely high level this year, and the model rewards them for that feat.

A few weeks ago, Boise got crushed in the BCS rankings for losing a very close game on the road to a top 20 team. That was simply dumb. They've been a dominant team all year long against a schedule that, while not fantastic, compares reasonably well to at least a few other members of the top ten, including Ohio St (who themselves played plenty of "Sisters of the Poor" opponents while playing only ONE quality team - Wisconsin, who beat them).

Virginia Tech is an interesting case. Part of this difference is that the model doesn't count AA games (which means that the JMU loss doesn't count against them here), but just as important is the fact that they've gone on a tear through the ACC, 9-0 with a boatload of utter ass-kickings. #6 may be a spot or two high, but they're clearly a top 10 team, especially if you weigh recent games more strongly than early-season games (which many people do when thinking about teams). The BCS has them way too low.

Alabama is a great example of how in the BCS, hard schedules get punished, and easy schedules get rewarded. Few teams have had tougher schedules than the Tide, and few have had worse luck in close games (two of their losses were by 3 or less). It hasn't been their year, but that doesn't mean they're not an excellent team; Michigan St got hosed by having to face the Tide in their bowl game.

Nebraska has three losses, which is why they're rated poorly by the BCS. However, other than the awful Texas loss, the other two were all quality performances. 3-point loss in title game to a very good Oklahoma team, 3-point loss at A&M are two losses that are absolutely nothing to be ashamed of, which basically means (once again) the BCS is punishing a team for its schedule. Moreover, Nebraska has had a number of very good showings, from the dominant wins at Washington and Kansas St to the double-digit quality wins over Mizzou and Oklahoma St.

South Carolina has four losses... but two are to the BCS #1 team (though the latter was an ugly ass-kicking), one is to a very good Arkansas team (though it was an ugly loss), and the other was an admittedly embarrassing defeat to Kentucky. Of course, they also had a great win over Alabama, dominating wins over Florida and Clemson, and have clearly put together a strong record against an extremely tough schedule.

Florida St beat the crap out of Florida, then lost to an excellent Virginia Tech team and the net effect was to drop a spot in the BCS, which is silly, yet another instance of the BCS actively punishing tough schedules. The 'Noles have had a legitimately tough schedule, they've had close losses (except at Oklahoma), and generally dominant wins (though BC and Clemson were close). They're underrated in the BCS.

Arizona is the same type of story as most other teams that compu-picks thinks is underrated: very tough schedule, tendency towards blowout wins and close losses (except at Stanford and Oregon, compu-picks' #'s 1 and 2). They're better than their record, and despite what the BCS thinks, there simply aren't 25 teams out there better than the Arizona Wildcats.

USC isn't eligible to be ranked by the BCS. If they were eligible, they'd very possibly be ranked. Five losses is never fun, but they've had a tough schedule, they've beaten a top 25 Arizona team and a nearly top 25 Hawaii team (they're top 25 in the BCS though), both on the road, and two of their losses have come to elite Oregon and Stanford teams (the Stanford loss was a nail-biter as well).

I don't really have much to say about NC St. I think the BCS is punishing them a bit because it's generally underrating the ACC; other than that they have the usual formula for an underrated team, tough schedule, generally close losses, a number of dominant wins (though some were pretty close).

Arizona St is the poster child for a team that's been punished for a really tough schedule. They're 4-6 in 1-A games... but they've played six road games, they've played all 10 games against AQ opponents, and two of their home games were against top five teams (and one was a very close loss). They also played a very good Wisconsin team (rated #5 in BCS) on the road and only lost by a single point. They had an admittedly atrocious showing at Cal, but other than that it's been a boatload of quality performances, just against a schedule that was ridiculously hard. It's extremely unfortunate that this team isn't allowed to make a bowl game, especially since SJ St bailed on them late. That said, those are the rules, and they should have found a replacement. In week one, Baylor, Kansas, Miami, Virginia (who ALSO had 2 AA games on the schedule), West Virginia, Rutgers, Air Force, SD St, Nevada, Louisiana Tech, and probably a bunch of others were all playing AA teams; if ASU tried harder, they probably could have gotten one of those teams on the schedule. Presuming a win in such a game (fair assumption against at least half the list, though teams like Nevada, Miami, WV etc. would have been tough), failing to do so cost them a bowl game. In 2010, ASU was a better team than around half of the teams who are going to bowls, which says it all about both how good ASU was and the administrative screw-ups that cost them a bowl game.

7) The following teams are ranked materially lower than the model than the BCS: Auburn, Wisconsin, LSU, Michigan St, Mississippi St, Hawaii, UCF.

Auburn is playing fantastically right now, but they have had a number of early struggles (most notably against Clemson), which are affecting their numbers. I'm inclined to think that compu-picks is underrating them... but I can see the argument.

Wisconsin is playing very well right now, but, even more than Auburn, had a number of early struggles (the 1-pt ASU win, the 13-pt SJ St win, and worst of all, the 10-point loss to a pretty mediocre Michigan St team). I would say that compu-picks probably underrates them a bit... but unless you choose to totally ignore their early-season issues, the BCS is overrating them.

LSU's schedule may not be at the very top of the list, but it's been strong and absolutely in line with their neighbors in the rankings. Where they're getting hurt is, unsurprisingly, their large number of close to very close wins. 6 points (UNC), 6 points (WV), 2 points (Tenn), 4 points (Florida), 3 points (Bama), and 7 points (Ole Miss) has pretty much defined their resume. They dominated Miss St, which was a very good showing, but other than that it's been the Auburn loss, the Arkansas loss, a bunch of close to very close wins, and comfortable wins against Vandy and ULM, neither of which are going to move the dial much.

In the BCS, Michigan St recently rose two points after barely beating a mediocre Penn St team, one week after they rose two spots after barely beating a bad Purdue team at home. Once again, the BCS punishes tough schedules and rewards easy ones. Moreover, in terms of season-long resumes, Michigan St has not been especially dominant (especially after they got waxed at Iowa), and has not had much of a schedule to date (Notre Dame was the only decent non-conference opponent, and they missed Ohio St, and they had a ridiculous eight home games [one was AA]). Even if you think compu-picks is too low on them, putting them into the top 10 is flat-out ridiculous. They're basically a homeless man's LSU, with the same tendency towards way too close wins, a lack of good performances against teams not named Wisconsin, a crummy schedule, and the massive beating Iowa laid on them.

Mississippi St played four top 25 teams and lost to them all, though two of the four were close losses. They barely held on against an awful UAB team at home, and struggled against mediocre Ole Miss and Kentucky teams. Top 25 is a bit of a reach for them, though to be fair it's not like compu-picks thinks there's a huge difference between them and #25 on the list.

Hawaii is 10-3... but with a really weak schedule, three double-digit losses (including a 35-point bludgeoning by Boise, a 18-point loss at a poor Colorado team, and a 13-point home loss to USC). They also had a couple squeaker wins (Army by just 3, Nevada by 6). They're a good team... but top 25 is too much.

One week after the ridiculous ranking of NIU, they gag in a bad loss and get replaced by another joke, Central Florida. UCF probably doesn't even belong in the top 40, much less the top 25. Their schedule was atrocious, they've played precisely one team that's even arguably top 25 (NC St... though the BCS doesn't have them ranked), and they have three losses anyway. The best two teams they beat were probably Houston and SMU. UCF shouldn't even be within shouting distance of the top 25. Utterly ridiculous.

8) This isn't directly to do with the list, but here's a couple fun lists of results:

Texas 20, @ Nebraska 13
Nebraska 48, @ Kansas St 13
Kansas St 39, @ Texas 14

@ Michigan St 34, Wisconsin 24
Wisconsin 31, @ Iowa 30
@ Iowa 37, Michigan St 6

Alabama 24, @ Arkansas 20
Arkansas 41, @ South Carolina 20
@ South Carolina 35, Alabama 21

USC 24, @ Arizona 21
@ Arizona 44, Washington 14
Washington 32, @ USC 31

@ Air Force 35, BYU 14
@ BYU 24, San Diego St 21
@ San Diego St 27, Air Force 25

If you try to apply "head to head is the only thing that matters" logic to this list, your head will explode. You can tease out certain information from these lists (Wisconsin, Alabama and Washington had both of their games on the road, they get a bonus; Iowa, Nebraska, Arkansas, Arizona and Air Force had losses much closer than their wins, therefore they get a bonus; etc.), but what it really does is highlight that each of these results was JUST ONE GAME. To properly evaluate a team, you need to evaluate the whole resume, not pretend that a single result means everything and the rest almost nothing just because of head to head "logic". That's why Compu-Picks doesn't give ANY special consideration to head to head results. You are what your resume says you are. Period.

Technical notes about the lists:

1) Conference ratings are straight averages of all of the teams in the league. There is no "central averaging" (like Sagarin does), or over-weighting the top teams, or anything like that. Such approaches would yield different numbers, and could potentially change the order of some of the leagues.

2) Games against AA teams are not counted. There are many good arguments both for and against counting such games (see this link for an interesting analysis of the issue). I have elected not to count these results in the Compu-Picks model. As is the case almost every year, this means that one or two especially surprising AA upsets don't make it into the numbers, skewing the results to a fair degree for a couple of teams. I believe that this is a more than acceptable tradeoff given the substantial issues that counting AA games would create, but you are certainly welcome to disagree with my decision on this matter.

3) As mentioned here, the purpose of this system is to make picks, not to create a list used for rankings. As such, I evaluate the system solely on the basis of how good a job it does making picks. I do not evaluate the system on the basis of whether or not it agreed with AP polls, BCS rankings, the BCS computers, or any other such list out there. In fact, the system has a long and established history of being substantially different than those sources. I am fine with these differences. To be honest, I publish these lists because I find them interesting and thought-provoking, and because I believe it is a good thing to introduce an approach that doesn't simply regurgitate the same avenues of thinking as you can find in most places.

4) The system is noisy, especially earlier in the year. This is why I start with only a top 10 / bottom 10 list, and slowly expand it. While I believe that the numbers are reasonable, I certainly accept that they're not perfect. If you believe that a specific team is over- or under-ranked, you may well be right. I bring this up because if you're going to criticize the system for being wrong about a team, I'd appreciate it if you explain why you think the system is substantially wrong, rather than just marginally so (if it's just one or two slots off, especially well before the end of the year, I'd consider that well within a reasonable error range).

5) Some people have expressed curiosity about compu-picks' schedule ratings. Essentially, it's the average difficulty of all 1-A games that a team has played to date, which basically means the average rating of all of a team's 1-A opponents, then adjusted for things like home-field advantage. This, of course, varies enormously from the NCAA's official schedule ratings, which simply look at the opponents' winning percentages and that's it. Needless to say, the NCAA's approach is silly and deeply flawed; it's better than picking numbers out of a hat, but no serious analyst should rely on it at all when trying to evaluate how difficult a team's schedule actually was.

2010 Compu-Picks Blog

Questions, comments or suggestions? Email me at [e-mail]