2009 Week 7 College Football Ratings/Rankings and Summary

Ratings through week 7 (games played thru 17-Oct).  These are also at http://sites.google.com/site/computerratings/ncaa-football/ncaa-ratings.

After several upsets, closer than expected games, and a few blowouts, there are a bunch of changes in this week rankings.

Boise State started things with a much closer than expected win (7 points) over then #65 Tulsa that dropped them over 3 ratings points and 2 spots to #4.  That left room for a new #2 and even before the weekend games Cincinnati made their case with a solid 17 point win on the road against formerly unbeaten and then #10 South Florida.  Florida had a golden opportunity to get to #2 by playing a second tough opponent after an easy schedule, but nearly lost and did not impress winning by only 3 at home and only moved to #3 while actually losing ratings points.

Texas also had a good opportunity to move up and coupled with some of the other surprises and and their 3 point win over OU did gain over 5 ratings points and move up to #5 but did luck out with Bradford getting hurt and missed out on moving up more due to such a narrow win over a now 3 loss team.  Helping Texas improve that many points was prior opponent Texas Tech taking it to Nebraska which jumped them an unbelievable 11 points and 36 spots.  They were helped by Texas winning and prior opponent Houston also winning big.

Iowa continues to win as underdogs and is up to #12 while fellow unbeaten and former #12 TCU moves up 3 to #9.  The aforementioned South Florida dropped over 4 points and 9 spots to #19 while another upset loss, this one by Ohio State, dropped them all the way to #24.  Ohio State's loss was also the big reason why USC did not improve and in fact dropped them 2 spots to #8.  Of course, USC didn't help themselves by letting Notre Dame come back from 20 down in the 4th to nearly take it to overtime.

With 7 undefeated teams left, it appears Florida and Alabama may be destined to face each other that way and Cincinnati, Texas, and Boise State are poised to make their case, but others vying to stay that way continues to include TCU and Iowa as Kansas, never really ranked highly by the computer, finally fell this week.  Check out the season projections to see how the teams should end up.

With all the upsets and strange games with teams not playing as expected, my computer did not have a good week predicting games.  The record against the spread was 26-27-2 and it went 40-15 picking winners.  Best picks were a disappointing 4-5 and even the locks missed a game for the first time going 2-2.  Full prediction performance here and if you are interested in a newsletter with additional details on picks against the spread, best picks, and locks, see this.

Rank	Team	Rating	Record	Schedule	Change
Rank	Team	Rating	Record	Schedule	Change
1	Alabama	95.193	7-0	73.413	+0, -1.095
2	Cincinnati	90.509	6-0	68.377	+5, +3.669
3	Florida	90.304	6-0	70.365	+1, -2.241
4	Boise St	89.476	6-0	70.245	-2, -3.365
5	Texas	89.463	6-0	71.034	+6, +5.200
6	Texas Tech	87.946	5-2	67.052	+35,+11.068
7	Virginia Tech	87.155	5-2	79.065	-2, -3.906
8	Southern Cal	86.448	5-1	73.770	-2, -0.764
9	TCU	85.959	6-0	69.379	+3, +2.130
10	Oregon	85.908	5-1	74.613	-2, -0.901
11	Oklahoma	84.862	3-3	70.091	-2, -0.460
12	Iowa	83.819	7-0	74.474	+3, +0.857
13	Nebraska	83.172	4-2	72.409	-10, -9.454
14	Miami FL	82.828	5-1	74.940	+0, -0.323
15	Georgia Tech	82.583	6-1	76.227	+4, +2.273
16	LSU	81.114	5-1	72.744	+0, +0.342
17	Arizona	80.820	4-2	76.177	+1, +0.478
18	Arkansas	80.678	3-3	77.076	+4, +1.491
19	South Florida	80.241	5-1	61.069	-9, -4.355
20	Clemson	80.184	3-3	72.945	+8, +2.057
21	California	80.041	4-2	73.825	+9, +2.142
22	Utah	79.624	5-1	69.503	-5, -0.856
23	Penn State	79.200	6-1	63.667	+9, +1.711
24	Ohio State	79.096	5-2	70.065	-11, -4.383
25	Oregon St	79.045	4-2	71.186	+8, +1.685
26	Fresno St	78.738	3-3	73.576	-2, -0.190
27	Stanford	78.301	4-3	72.742	-4, -0.739
28	Notre Dame	77.933	4-2	74.350	-1, -0.353
29	Tennessee	77.909	3-3	71.009	-4, -0.490
30	Mississippi	77.809	4-2	67.992	+9, +0.916
31	South Carolina	77.767	5-2	73.307	+12, +2.453
32	Kentucky	77.649	3-3	76.703	+19, +4.016
33	Brigham Young	77.583	6-1	68.211	-4, -0.379
34	Auburn	77.548	5-2	70.678	-8, -0.765
35	Pittsburgh	77.534	6-1	67.438	-4, -0.278
36	West Virginia	76.643	5-1	66.295	+11, +2.043
37	Florida St	76.636	2-4	76.635	+1, -0.267
38	Kansas	76.577	5-1	64.526	-18, -3.665
39	Houston	76.547	5-1	67.351	+24, +4.671
40	Oklahoma St	76.465	5-1	63.479	+12, +3.163
41	Michigan	76.419	5-2	65.470	-4, -0.498
42	Central Michigan	76.208	6-1	63.575	+2, +0.932
43	Connecticut	75.978	4-2	67.006	-3, -0.901
44	Washington	75.977	3-4	78.295	-8, -0.980
45	Boston College	75.338	5-2	67.766	+4, +1.251
46	Georgia	75.309	4-3	76.097	+9, +2.214
47	Arizona St	75.102	4-2	64.217	+17, +3.885
48	UCLA	74.988	3-3	75.267	-13, -2.131
49	Wisconsin	74.917	5-2	71.259	-15, -2.319
50	Michigan St	74.615	4-3	69.611	-8, -1.157
51	Virginia	74.226	3-3	71.405	+6, +1.633
52	Northern Illinois	73.283	3-3	67.024	-6, -1.337
53	Iowa St	73.207	4-3	66.691	+13, +2.272
54	Missouri	73.089	4-2	68.291	-33, -6.364
55	Navy	72.600	5-2	65.583	-5, -1.121
56	Minnesota	72.514	4-3	72.058	-2, -0.638
57	Air Force	72.443	4-3	63.471	-4, -0.764
58	Duke	71.900	3-3	66.288	-10, -2.620
59	Tulsa	71.892	4-2	63.677	+6, +0.942
60	Wake Forest	71.793	4-3	69.548	-15, -3.351
61	Ohio U.	71.454	5-2	65.798	-2, -1.055
62	Mississippi St	71.407	3-4	68.264	+15, +3.854
63	Idaho	71.350	6-1	66.635	-3, -1.004
64	North Carolina	71.148	4-2	65.523	+3, +0.351
65	Rutgers	70.945	4-2	58.661	-4, -1.373
66	Kansas St	70.777	4-3	68.755	+16, +4.956
67	Purdue	70.733	2-5	72.930	+9, +3.029
68	Colorado St	70.453	3-4	73.893	+0, -0.292
69	Troy	70.159	4-2	69.705	-13, -2.699
70	Baylor	69.301	3-3	68.057	-8, -2.912
71	Bowling Green	68.849	3-4	70.371	-2, -1.583
72	Marshall	68.439	4-3	69.776	-1, -0.912
73	Colorado	68.175	2-4	73.620	+19, +4.069
74	Nevada	67.911	3-3	68.178	-2, -1.189
75	Southern Miss	67.698	4-3	63.934	+4, +1.215
76	Temple	67.600	4-2	62.518	+2, +0.533
77	Arkansas St	67.272	1-4	67.192	-19, -5.295
78	Western Michigan	67.234	3-4	67.679	-3, -0.495
79	North Carolina St	66.842	3-4	65.479	-9, -2.870
80	Louisiana-Monroe	66.679	4-2	65.604	+21, +4.902
81	Indiana	66.395	4-3	66.424	+0, +0.358
82	SMU	66.347	3-3	68.359	+2, +0.834
83	Louisville	66.342	2-4	67.356	+0, +0.696
84	Texas A&M	66.165	3-3	65.271	-11, -2.833
85	Northwestern	65.680	4-3	62.058	+3, +0.640
86	East Carolina	65.619	4-3	64.946	+1, +0.493
87	Toledo	65.097	4-3	67.598	+12, +2.269
88	San Jose St	64.757	1-5	75.970	-8, -1.531
89	San Diego St	64.667	2-4	68.091	-4, -0.811
90	Syracuse	64.394	2-4	69.394	+0, +0.057
91	Middle Tennessee St	64.390	3-3	68.212	-17, -3.599
92	Central Florida	64.348	3-3	63.596	-3, -0.108
93	UTEP	64.052	2-4	69.326	-2, -0.075
94	Louisiana Tech	64.037	3-3	62.054	+3, +0.854
95	Maryland	63.607	2-5	70.785	-9, -1.644
96	Wyoming	63.489	4-3	66.021	+9, +2.557
97	Louisiana-Lafayette	62.824	4-2	64.560	-1, -0.466
98	Buffalo	62.521	3-4	64.830	+0, -0.646
99	Utah St	62.420	1-5	68.244	-6, -1.247
100	Alabama-Birmingham	62.174	2-4	66.059	-6, -1.468
101	Florida Atlantic	62.167	1-4	70.739	+9, +4.496
102	Hawaii	61.778	2-4	66.219	+0, +0.406
103	Kent St	61.563	3-4	62.562	+3, +1.219
104	UNLV	61.139	2-5	67.072	+0, +0.011
105	Vanderbilt	60.641	2-5	64.334	-2, -0.565
106	Illinois	60.267	1-5	69.937	-11, -3.226
107	North Texas	59.590	1-5	68.575	-7, -2.735
108	Florida Int'l	59.243	1-5	70.421	+6, +2.588
109	Washington St	58.816	1-5	75.147	+2, +1.167
110	Memphis	58.323	2-5	64.006	-3, -1.543
111	Army	58.322	3-4	61.931	-3, -1.330
112	Akron	57.836	1-5	66.394	+0, +0.615
113	Tulane	57.299	2-4	66.322	-4, -0.486
114	Miami OH	56.065	0-7	76.509	+1, +0.948
115	Ball St	55.421	0-7	66.127	-2, -1.284
116	New Mexico St	52.528	3-4	60.488	+0, -2.019
117	Eastern Michigan	50.451	0-6	67.632	+0, -3.157
118	Western Kentucky	48.451	0-6	67.842	+2, +4.239
119	Rice	48.174	0-7	71.477	-1, -1.983
120	New Mexico	48.110	0-6	69.077	-1, -1.251

Thoughts on today's games

Here are some thoughts on some of today's games using my computer's ratings as a reference point for the result.

In the early big game, Texas held on to beat Oklahoma 16-13 so my computer missed the upset pick but got the pick against the spread having picked OU by 1.1.  But one could argue that with Bradford playing more than 8 plays, OU could have won and that without him the result is as expected.  In fact, if Bradford hadn't played against Baylor and that resulted in the game being 13 points closer, the computer would have picked Texas by 1.

Ohio State was upset by Purdue, and while the computer certainly didn't pick the upset, it has thought Ohio State was overrated all year never being in the top-10 and being #13 coming into this game.  Due to that, it had Purdue plus the 14 points and got that one right.

Iowa is another team that hasn't gotten their due from the voters and Vegas having been underdogs in 3 games this season, 2 of those after the upset of Penn State put them on the map, at home by 8 to Michigan and by 3 on the road to Wisconsin today.  The computer picked them to win and they came through again to get the pick right straight up and against the spread.

The computer liked Florida by 16.4 so with the spread Florida by 24.5 the clear pick was Arkansas.  While I wouldn't say the computer hasn't liked Florida, it did have them #4 and not the consensus #1 like the polls.  And the computer has liked Arkansas a bit more than the polls having them #22 while they were around #30 in the polls.  The result?  Arkansas missed 2 field goals and Tebow got them in position for the 3 point win to eke it out and the computer got both the win and spread right.

USC visited Notre Dame with the spread by 11 and my computer liked them by only 5.1 due in large part to them being up and down and having the terrible loss to Washington.  If the "real" USC shows up, i.e. we omit the loss to Washington, USC is the computer's pick by 14.  Well, the "real" USC showed up for awhile taking a 20 point lead into the 4th but they nearly let the game get away but won by 7 just to make my computer's pick look more accurate and get the win and spread correct.

The computer has really liked Nebraska this year and picked them by 18.7 over what must be a disappointed Texas Tech, but Nebraska laid an egg and lost by a ton meaning the computer missed on both the win and spread.  Nebraska, and as a consequence Virginia Tech and Alabama may be dropping in the rankings this week.

That's it for now, perhaps more later.

2009 Week 7 College Football Preview

Week 7 has already started and while the computer has stunk it up already going 0-3 against the spread, I'm hopefully that is just an aberration and the trend of the past few weeks (32-26-1, 29-19-1, 31-22-2 and 4-0 over the 3 weeks on locks) will continue.

While the overall picks did well and the 1 lock of the week came through, the games I highlighted weren't so good going 1-4.  I'm not sure that's an omen though as the highlighted  picks were nearly perfect the week before.  In any case, here are some of the interesting games of the weekend.

The computer has really liked Nebraska this year due to their 1 loss being on the road by 1 point to a team with 1 loss that being to the far and away #1 team.  But they will perhaps be tested this week with Texas Tech coming to town.  Vegas picks Nebraska by 6.5 but the computer likes them by 18.7 which seems like a lot.  Either they prove the computer right, or their ranking could drop a bit next week even with a win.

Iowa takes their high wire act to Wisconsin this week where they are 3 point underdogs but the computer likes them by 2.7.  After getting as high as #5 in the rankings 2 weeks ago they've dropped to #15 after close wins over not so great teams, and while Wisconsin may not be great they are 5-1 and the game is in Madison.

In a game between my computer's #5 and #19, with both being 5-1, and with "Tech" in both teams names how could a techie like me not look at it, Virginia Tech visits Georgia Tech.  Vegas as Virginia Tech by 3 but the computer likes them by 7.8.  Given the linkages, the outcome of this game and the Nebraska game could have a big impact on next weeks rankings.

In another battle of 1-loss teams, USC visits Notre Dame and is favored by 11 but the computer likes them by only 5.9.  This game will ultimately depend on which USC team shows up as they are one of the more up and down teams this year.

And in one of the more anticipated games of the weekend, even if Oklahoma has 2 losses, they face Texas in Dallas.  Vegas likes Texas by 4, but the computer is not impressed with their schedule to date and picks Oklahoma by 1.1.  One could argue that had Oklahoma had Bradford they wouldn't have 2 losses so there is a good chance this one is closer than the spread.


If you want more of the details behind the picks and which are the best picks and locks, take a look at my newsletter that is sent out weekly.

Enjoy the games.

Degrees of Separation

Ratings systems are complicated beasts, but one thing that is certain is that for a system to be accurate in rating the teams and predicting games, there has to be enough data.  But more importantly, when comparing two specific teams with a pool of teams as large as the NCAA has, you need to have a linkage between the teams and ideally there are enough linkages that are short enough for the comparison to be meaningful.

Something I track to help with this is the degrees of separation between teams playing each week.  The term "degrees of separation" was popularized a number of years ago with the Six Degrees of Kevin Bacon trivia game.  And what I do with predictions is basically the same thing.  Let's take a look at an example.

In tonight's game between Cincinnati and South Florida the computer is picking South Florida by 0.8 while the line is Cincinnati by 3.5.  That much of a difference seems pretty significant, but how sure should we be of this pick?  How many degrees of separation are there between the teams?

A quick check reveals that the shortest path is 4 and there are three ways to accomplish it:

• South_Florida-->Wofford-->Wisconsin-->Fresno_St-->Cincinnati
• South_Florida-->Western_Kentucky-->Florida_Int'l-->Rutgers-->Cincinnati
• South_Florida-->Syracuse-->Northwestern-->Miami_OH-->Cincinnati

There are also 9 paths with 5 degrees.

What does all this mean?  Well, 4 degrees of separation is still quite a bit and having 4 paths that are pretty distinct helps but ideally there are fewer degrees of separation so my computer would put a moderate probability on the pick being correct.

The degrees of separation, along with other factors including the predicted margin, the Vegas line, and how predictable the teams are, all enter in to the best picks and locks I publish each week in my newsletter.  Take a look and subscribe if you are interested.

Boise State Wins, But ..., and Cincinnati vs South Florida

Boise State won last night, beating Tulsa 28-21, and while a win is a win in many voters eyes (at least when rating a highly ranked team from a top conference), the computer doesn't agree and a 7 point win on the road against the #65 team in the rankings is likely going to drop Boise State from their current #2 spot in the next ratings to #5 or #6.

How much?  Right now with only the Tuesday night game the other game played, Boise State would lose just over 3 ratings points and drop to #5.  Florida would be the beneficiary moving to #2 and Virginia Tech would move to #4.

There is another big game tonight that should help sort things out for those vying to get into the BCS with Cincinnati visiting South Florida.  With Vegas favoring Cincinnati by 3.5 or so, the computer is picking the upset with South Florida by 0.8.  Hopefully it will be a competitive game.  A win by Cincinnati could move them into the top-5 and a win by South Florida could get them to #6 or #7.

The South Florida pick is not one of the computer's best picks or locks this week, but if you are interested in seeing which are, take a look at the weekly newsletter subscription that is available.

Comparing ESPN's power rankings with my computer

ESPN publishes their NFL Power Rankings each week and I thought it would be interesting to compare them with my computers rankings.

Team	ESPN	Schmidt Computer
Indianapolis	1	2
NY Giants	2	11
Minnesota	3	4
New Orleans	4	1
Denver	5	3
Cincinnati	6	6
Philadelphia	7	14
Atlanta	8	7
NY Jets	9	8
New England	10	5
Baltimore	11	10
Chicago	12	9
Pittsburgh	13	13
Green Bay	14	15
San Francisco	15	12
San Diego	16	20
Dallas	17	18
Miami	18	16
Arizona	19	17
Seattle	20	19
Jacksonville	21	21
Houston	22	22
Carolina	23	27
Washington	24	28
Buffalo	25	24
Detroit	26	25
Tennessee	27	26
Cleveland	28	23
Oakland	29	29
Kansas City	30	30
Tampa Bay	31	32
St. Louis	32	31

All in all, things aren't terribly out of whack, but a few things stand out.

ESPN likes the Giants and Philly a lot more than my computer ranking them 2 and 7 versus 11 and 14 for the computer.  Perhaps the Giants are indeed a top team, but it is difficult to rank them there when they've played two winless teams, one 1 win team (and that win over a winless team), and only one team with a winning record and that teams wins are over teams with a combined 1-13 record.  Philadelphia also suffers from playing a similar weak schedule.

On the other side of the coin, ESPN only ranks New England 10 but the computer has them 5.  Yes, they have two losses but two of their wins are over 3 win teams.

So, which is more accurate?  Just for fun, I went back through each week to see how well the ESPN rankings did in predicting winners and for the year they are 51-25.  For comparison, my computer is 53-23 and the last 2 weeks when both ESPN and my computer should have better data, my computer is 21-7 versus ESPNs 18-10.  So I'm liking that my rankings are at this point more accurate than the ESPN analysts!

I'll continue to monitor this through the year and report back.

2009 Week 5 NFL Projected Records

Below are the projected end of regular season records after week 5 of the NFL season.  These can also be found regularly at http://sites.google.com/site/computerratings/ncaa-football/nfl-projections.

New Orleans was hurt a bit in the ratings by 3 previous opponents losing, particularly the Jets, and that causes them to now be projected to have 1 loss.  They do still have a 32.6% chance of going undefeated though.  The other undefeated teams have an 8.7% (Indianapolis), 5.0% (Denver), 1.8% (Minnesota), and virtually 0% (NY Giants) chance of finishing that way.

If the projections are correct, the AFC playoff teams would be Indianapolis, Denver, Cincinnati, NY Jets, New England, and Pittsburgh.  The NFC would be New Orleans, Minnesota, NY Giants, San Francisco, Atlanta, and one of Chicago or Philadelphia.

Format below is team, most likely record, chance of that record, chance of 1 game better, and chance of 1 game worse.

Team	Record	% chance	% +1	% -1
Team	Record	% chance	% +1	% -1
New Orleans	15-1	41.5	32.6	20.3
Indianapolis	14-2	31.2	25.1	22.0
Denver	14-2	35.3	23.0	25.3
Minnesota	13-3	27.6	21.5	22.3
Cincinnati	12-4	31.0	21.9	24.3
NY Giants	11-5	27.5	21.5	22.0
NY Jets	11-5	28.6	17.8	26.2
Atlanta	11-5	25.7	16.7	25.2
New England	10-6	28.2	27.6	17.3
Chicago	10-6	25.4	21.2	20.5
Philadelphia	10-6	24.9	20.2	21.0
Pittsburgh	10-6	26.9	17.4	25.5
San Francisco	10-6	24.9	17.0	24.0
Arizona	9-7	26.0	21.3	20.9
Green Bay	9-7	25.5	20.8	21.0
Baltimore	9-7	25.4	20.1	21.6
Dallas	9-7	28.8	19.9	24.8
San Diego	8-8	26.0	25.3	16.8
Seattle	8-8	25.2	21.7	19.8
Miami	8-8	25.9	20.2	22.0
Jacksonville	8-8	24.2	16.4	24.0
Houston	6-10	27.2	19.2	24.3
Cleveland	5-11	25.9	22.8	19.3
Buffalo	5-11	25.8	18.5	23.8
Oakland	4-12	28.8	18.4	27.0
Carolina	4-12	31.9	18.1	29.1
Washington	4-12	38.4	17.9	32.7
Detroit	4-12	29.6	16.8	29.6
Tennessee	3-13	26.5	26.1	16.3
St Louis	2-14	31.0	21.4	25.7
Kansas City	2-14	32.8	20.4	28.0
Tampa Bay	1-15	41.0	25.8	23.3