As of the end of Nevada - Hawaii, these are the top 10 and bottom 10, along with conference ratings (straight average of the scores for each team in the confernece).
Remember that this is a predictive model, designed to pick games and show how good a team actually is. Its results can be very different
from what you'll see elsewhere. The workings of the model are confidential (it is, after all, designed to make winning picks),
but I'm happy to answer questions about the models' results.
| Rank || Team || League || Score || Schedule Rank || BCS Rank |
| 1 || Boise State || WAC || 0.89 || 27 || 3 |
| 2 || Texas Christian || Mountain West || 0.86 || 34 || 5 |
| 3 || Oregon || Pac-10 || 0.73 || 47 || 2 |
| 4 || Stanford || Pac-10 || 0.70 || 5 || 12 |
| 5 || Oklahoma || Big 12 || 0.70 || 16 || 1 |
| 6 || Nebraska || Big 12 || 0.67 || 41 || 16 |
| 7 || Missouri || Big 12 || 0.64 || 48 || 11 |
| 8 || Virginia Tech || ACC || 0.63 || 35 || 25 |
| 9 || Auburn || SEC || 0.60 || 25 || 4 |
| 10 || Alabama || SEC || 0.58 || 44 || 8 |
| 111 || Ball State || MAC || -0.62 || 113 || |
| 112 || Middle Tennessee State || Sun Belt || -0.62 || 116 || |
| 113 || Kent || MAC || -0.64 || 114 || |
| 114 || Memphis || C-USA || -0.65 || 78 || |
| 115 || Western Kentucky || Sun Belt || -0.68 || 95 || |
| 116 || North Texas || Sun Belt || -0.71 || 119 || |
| 117 || Eastern Michigan || MAC || -0.74 || 108 || |
| 118 || New Mexico State || WAC || -0.80 || 100 || |
| 119 || New Mexico || Mountain West || -0.82 || 87 || |
| 120 || Akron || MAC || -0.91 || 107 || |
| League || Rating |
| Pac-10 || 0.35 |
| SEC || 0.28 |
| Big 12 || 0.27 |
| Big Ten || 0.12 |
| ACC || 0.10 |
| Indep || 0.06 |
| Mountain West || 0.00 |
| Big East || -0.02 |
| WAC || -0.03 |
| C-USA || -0.23 |
| MAC || -0.41 |
| Sun Belt || -0.49 |
Some thoughts on the list:
1) As usual, I'm posting the compu-picks top and bottom ten, and will slowly expand the list as the season goes on.
The reason I do this is that the teams at the very top and very bottom have largely separated themselves by now,
while the teams on the next tier can largely be jumbled together.
For instance, Ohio St has beaten Miami, blown out Indiana,
and won at Illnois, but just lost at Wisconsin; meanwhile, Michigan St barely held off Notre Dame at home, struggled against FAU,
beat Wisconsin at home, beat Illinois at home,
and has only had one road game so far, at Michigan. You can make a reasonable argument for rating either one above the other, and would
have a difficult time arguing that either has been substantially better than the other.
You'd have a much tougher time, however, arguing either one above Oklahoma, who (by the model's ratings) has had a tougher schedule than either
one of them, is undefeated, and has won a couple games in truly dominant fashion.
2) Virginia Tech at #8 is a bad result. IF you pretend that the James Madison loss didn't happen,
it's a reasonable rating, but it did happen, and I'm not going to insult your intelligence by pretending
the rating is correct or saying "oh, it'll all even out eventually". It won't (though with more games the resume
will get more defined, and that single result will become a less substantial part of their overall resume, so it will
at least become a smaller problem;
moreover, it's at least slightly possible they'll eventually deserve a lofty rating, though I'm a skeptic).
Unfortunately, Virginia Tech is a team whose ranking here is skewed by the fact that the model doesn't count AA games.
Unfortunately, it's a choice between counting AA games (which creates a substantial amount of error for a large number of teams)
or not counting them (which can lead to results like these). There are no easy answers on this issue; the best I can do is
point out what's happening, admit I don't agree with it, and move on to the results that are more worth discussing.
3) Boise St is rated #1, partially because they've been generally dominant,
and partially because they've played a pretty good schedule to date (though Virginia Tech's rating is currently
inflated and likely to fall, affecting Boise's schedule strength a bit). Obviously,
their schedule strength will start dropping as the year goes on (though they're already
played two of their worst opponents, NM St and SJ St), and moreover, there's a very reasonable
chance that one or more of the AQ teams will get on a hot streak and simply pass them.
However, provided that they don't suffer stumbles like their surprisingly poor 28-21
result at Tulsa last year, I would expect their rating to remain very high. Every indication so far
is that they really are an excellent football team. Of course, they're going to get hurt by the BCS's horrifically awful decision
to force all of the computer rankings to ignore margin; if it turns into a close race at the end of the year, that could
very well turn into the difference between whether or not Boise gets a national title bid.
TCU is largely the same story, rated highly for much the same reasons (generally dominant against a tough
schedule so far). They'll likely take less of a schedule beating the rest of the way, but like Boise could
possibly get passed or drop if they have an off week or two, or if one of the teams below them really steps
up the rest of the way.
4) An especially noticeable team near the top of the list is Stanford, the highest-rated one-loss team
(though Nebraska isn't very far behind). Essentially, they've played an extremly difficult schedule
(as noted elsewhere, the AA game counts as a bye and doesn't affect the numbers), and done very well
against it. Losing by 21 points to Oregon hurt quite a bit, but they've also had some extremely well-rated
performances, especially the 35-0 rout at UCLA (which only looked better after Texas, who UCLA beat easily,
won at Nebraska this weekend). I think the model is over-rating them, but I do understand why it has them there.
Based on what we've seen from them so far, it's at least possible that they really are a top 5 team.
And I do think that the BCS is underrating them at 12. A fair rating is probably 8th or 9th, midway between the Compu-Picks
and BCS pegs. I'm OK with that level of model noise at this stage of the season.
5) Another noticeable team near the top of the list Nebraska, the next highest-rated one-loss team.
Unlike Stanford, they aren't getting a substantial boost for schedule played. What's really keeping them so
highly rated is how utterly dominant they've been in their wins. 56-21 at Washington, and 48-13 at Kansas St
are very impressive performances by any measurement (right now, 48-13 at Kansas St is rated one of the best performances
by ANY team in ANY game so far this year). And even 49-10 against WKU and 38-17 against Idaho were sufficiently
dominant that the games didn't kill their rating too badly (they did hurt, though). I'd put them lower than the model has them,
but the BCS is underrating them at 16. A "fair" rating is probably right around the mid-way point between the two, although you can easily
argue that they realy do deserve to be called a top ten team, and that the model is closer to right than the BCS.
I'm OK with that level of model noise at this stage of the season.
6) It's interesting to see some of the SEC teams clustering together (Auburn-Alabama are 9-10, LSU is barely behind at 11th, and there are
a couple others not very far behind). Since the seeming "big three" of the league (Auburn-Alabama-LSU) have all yet
to play each other, there really hasn't been much of a chance for separation yet. All three have good points, and all three
have warts. Alabama, despite their loss, has dominated in a number of their wins. Auburn has had a very tough slate,
but has had a disturbing number of close calls (3-point wins at both Miss St and Kentucky, a 3-point home overtime win over Clemson;
even the South Carolina win was only by 8). LSU's resume is much like Auburn's, with a tough schedule but disturbing number of
close calls (only won by 6 against both UNC and West Virginia, only won by 4 against Florida, and only won by 2 against Tennesse).
It's entirely possible one of these three emerges as a true power, but so far Alabama has been tagged and both Auburn and LSU look
like they're ready to get tagged. "Clutch", "the will to win", etc. are all nice, but at some point if you're competing for a national
title you can't keep letting teams hang around, especially if you're doing that before even hitting the toughest part of the schedule.
You can feel free to disagree with the model's ratings of these three, but I stand behind them as being perfectly reasonable based on
what each has achieved to date. I will almost certainly revisit the SEC West in next week's commentary, since LSU and Auburn are playing
each other in one of the week's biggest games.
7) Oklahoma is an interesting case as well. If their whole resume was the ass-kickings of Florida St, Iowa St and the Texas win, they'd be #1.
If their whole resume was the Utah St, Air Force and Cincy squeakers, they be somewhere between 20th and 40th. Instead, both sets of results happened,
and they're rated #5. I'd consider that reasonable for a team with the extreme sorts of swings they've had. It'll be really interesting
to watch them over the next few weeks to so if they're really the #1 team they've looked like at times, the "meh" team they've looked like at times,
of if they're going to continue the huge up and down trend they've shown so far. They certainly could end up earning their #1 BCS ranking...
but I agree with the model that it's too early to annoint them at that spot just yet.
8) Oregon basically gets a "TBD" for their rating. They've done quite well so far, but barring a huge upset by UCLA or Washington, their resume
will be defined by how they do against the set of: @ USC, @ Cal, vs Arizona, @ Oregon St. They'll definitely need to get better with their road
performances, but if they do, jumping to #1 in this system is very much on the table. For a team that hasn't really hit the toughest part of the schedule,
I'd consider #3 to be a totally reasonable rating.
9) Michigan St is ranked 7th by the BCS. That's too much. They needed a fake field goal in overtime to beat Notre Dame at home, they only beat
FAU by 13 at home (no that's NOT a "road" game, it was in Detroit against a team from Florida; that's a home game), they've only had one actual road game, against what looks
like a pretty mediocre Michigan team. Yes, the Wisconsin and Illinois wins were nice, but overall that's just not a top 10 team. Beat Iowa (which I doubt)
and we'll talk. I have absolutely zero problem with the model not having them in the top ten.
10) I'll expand my discussion of the league ratings as the year goes on and each league's resume becomes clearer. For now I'll note that it correlates very well
with the Sagarin league numbers ( link here ). While that certainly doesn't
"prove" anything, I take it as a sign that it's pretty reasonable. If there's interest I can go into the league numbers in some detail next week.
11) 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 31, UCLA 22
UCLA 34, @ Texas 12
@ Hawaii 27, Nevada 21
@ Nevada 52, Cal 31
@ Cal 52, Colorado 7
@ Colorado 31, Hawaii 13
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 (UCLA had both games on the road, they get a bonus;
Nevada's loss was the only close result from their list, 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).
Questions, comments or suggestions? Email me at firstname.lastname@example.org