2010 Compu-Picks Ratings - College

Mr Pac Ten
Posted Oct 27, 2010

The Compu-Picks model rates the top and bottom teams in college football after week eight

As of the end of LA Tech - Boise, these are the top 15 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 Texas Christian Mountain West 0.86 31 4
2 Boise State WAC 0.85 37 3
3 Oregon Pac-10 0.77 47 2
4 Nebraska Big 12 0.70 24 14
5 Missouri Big 12 0.68 34 6
6 Alabama SEC 0.64 39 7
7 Stanford Pac-10 0.60 16 13
8 Arizona Pac-10 0.60 23 15
9 Auburn SEC 0.59 19 1
10 Utah Mountain West 0.59 95 8
11 Oklahoma Big 12 0.58 8 9
12 Virginia Tech ACC 0.58 57 23
13 Ohio State Big Ten 0.56 64 11
14 South Carolina SEC 0.52 11 20
15 Louisiana State SEC 0.50 12 12
111 Louisiana-Monroe Sun Belt -0.59 97
112 Rice C-USA -0.60 99
113 Memphis C-USA -0.66 83
114 Buffalo MAC -0.67 76
115 North Texas Sun Belt -0.69 117
116 Louisiana-Lafayette Sun Belt -0.71 114
117 New Mexico Mountain West -0.76 86
118 Eastern Michigan MAC -0.78 109
119 New Mexico State WAC -0.81 100
120 Akron MAC -0.93 106

League Rating
Pac-10 0.35
SEC 0.28
Big 12 0.26
Big Ten 0.12
ACC 0.11
Indep 0.08
Big East -0.01
Mountain West -0.01
WAC -0.03
C-USA -0.22
MAC -0.42
Sun Belt -0.49

Some thoughts on the list:

1) Last week, I posted the compu-picks top and bottom ten, and this week I'm expanding to the top fifteen and bottom ten. I will continue to 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. You can even see this in the scores, where teams 7 through 12 are basically tied. It's even closer further on towards the middle.

2) Virginia Tech at #11 is still overrated. However, I'm a good deal happier now that there aren't any really egregious examples of teams that are unfairly ranked below them. In addition, I think this is a good time to clarify how much they're really overrated. To date they've had seven 1-A games, and an overall rating of 0.59. If we guesstimate their rating from the James Madison loss as -0.5 (they were home, they lost a fairly close rating, and Sagarin rates JMU as equal to a slightly below average 1-A team; given all that, a game rating of somewhere between -0.3 and -0.7 seems a reasonable range), then you get a new average of about 0.45, a drop of 0.14. That's a material drop, over a 0.1 change... but it still puts Virginia Tech in the top 25 (though not by much). So we're not exactly talking about a team that ought to be #40 or something wacky like that. Still overrated, and still not a good thing to see.. but it's becoming increasingly less substantial as the overall resume continues to develop. Barring something interesting popping up next week, I don't anticipate revisiting this issue.

3) It's certainly surprising that Auburn could win a big game against LSU and stay put at #9 on the list. Two teams (Arizona, Alabama) passed them on the way up, and two teams (Oklahoma, Virginia Tech) passed them on the way down. Since I doubt anyone is complaining about the teams that passed them on the way down, I'll focus on two things: first, why Auburn's rating didn't jump following the LSU win, and second, why those three teams' ratings jumped enough to pass the Tigers.

First, Auburn. Their win over LSU was a good one, and it did improve their rating. However, a 7-point home win against a team rated close to them didn't move the dial enormously. It was a quality win, but not a huge eye-opener. And on the downside, three of their earlier opponents (Miss St, UL-Monroe, Kentucky) had especially bad showings. The net effect of the two factors (positive for LSU win, negative for those three prior opponents) largely cancelled.

As for their opponents, Alabama's rating jumped largely because of their impressive win at Tennessee (who has been far from good, but still nearly won at LSU and generally hasn't been so bad that 41-10 isn't fairly impressive). Like Auburn's win, it wasn't enough to really move the dial, but it helped a bit, and Bama didn't really have any prior opponents' ratings plummet. Arizona's rating jumped due to two factors: first, they had an impressive blowout of Washington; and second, two of their prior opponents (Cal and Washington St) had especially good weeks that raised their ratings.

4) A second note on Auburn, since I'm sure this is going to be a popular point of discussion. Ignoring the question about lack of movement from last week to this week, why does Compu-Picks have them so much lower than the BCS? It's pretty obvious that the computers LOVE Auburn, and the voters, while not quite as high on them, still like them a great deal. So what, exactly, is the difference? It's not schedule, as compu-picks notes that Auburn has had the toughest schedule of all the unbeatens (though not by an enormous margin).

Instead, the reason is domimance. Specifically, an utter lack of it on Auburn's part. This is a team that, week after week, has barely skated by. Yes, they blew out a couple of Sun Belt teams, but other than those two (and any top 10 team should blow out Sun Belt teams at home), they've played six 1-A games, and five of them have been single-digit wins (Arkansas was a 65-43 shootout). They've had an 8-point win (SC), a 7-point win (LSU), a pair of 3-point wins (Miss St and Kentucky), and a third 3-point win that was in overtime (Clemson). If you think that a win is a win is a win, and that this simply doesn't matter, then yes, you should probably vote Auburn #1. But if you think that this consistent pattern of barely holding on should be factored in, it's difficult to justify having them at #1. Is a team that has won three of eight by a field goal, four of eight by a TD or less, and five of eight by single-digits REALLY the #1 team in the country? And if so, then how exactly was this different from Oklahoma getting placed into the top spot last week? The Sooners pretty much had the exact same type of warts (way too many way too close wins), and their placement was treated with almost universal disdain.

It's also worth noting that Auburn has been very questionable on the road; only twice have they left the friendly confines, and both times they escaped with 3-point wins against teams that are clearly not at the top of the league (Miss St is probably about average for the league, and Kentucky is solidly below-average). This is factored in by the model.

Now, that's not to say that Auburn isn't a very good team (pretty clearly they are), but as far as I'm concerned, the evidence doesn't really suggest that they're actually the best team in the country. Does compu-picks underrate them at #9? Probably... but the BCS is overrating them at #1 as well. As usual, my conclusion is that the "fair" value is about in the middle of the two.

5) It's interesting to note that Oregon has closed the gap between them and the top by about a third, which is pretty remarkable since it was one of the Pac-10's weakest teams, UCLA, that they blew out (and it was at home). If (and it's a big if) they can keep this up in the tougher games, especially on the road, they can absolutely close the rest of the gap before the end of the regular season. Ditto for Nebraska and Missouri (both jumped a bit from last week to this, and if one team wins their week 9 matchup handily they'll make a solid move upwards).

6) I've been asked specifically about Missouri. One thing compu-picks recognizes is their impressive level of dominance (only two games decided by single digits, and one was this past week's 9-point win over Oklahoma). It's not quite on the level of a few others, especially Oregon, Boise, Utah and TCU (the top 4 by a solid margin right now), but they're still undefeated with most of their games not being especially close. It's also worth noting that they've been very hot lately, with their last three wins (Colorado, Texas A&M, Oklahoma) all rated highly, especially the 21-point win at A&M, which currently rates quite well on the model's list of good wins.

7) I've also been asked about league ratings. It's still fairly early to really give the league ratings much weight... but I don't mind digging into the logic a bit. This is going to be a gross over-simplication, but I think it's still somewhat instructive. I'm going to focus on three specific leagues, the Pac-10, SEC, and Big Ten. This is far from the whole picture, but let's look at each league's five best and five worst non-conference games. Starting with the best:

Arizona @ Toledo: W 41-2
Cal vs Colorado: W 52-7
Stanford @ Notre Dame: W 37-14
UCLA @ Texas: W 34-12
USC @ Hawaii: W 49-36 (note that a close win against a better opponents - Arizona vs Iowa - wasn't very far off the top five)

Bama @ Duke: W 62-13
Bama vs Penn St: W 24-3
Bama vs San Jose St: W 48-3
Auburn vs UL-Monroe: W 52-3
Miss St @ Houston: W 47-24
(note that a couple closer wins against better opponents, such as Arkansas vs A&M, LSU vs UNC, and LSU vs WV, weren't very far off the top five)

Big Ten:
Iowa vs Ball St: W 45-0
Iowa vs Iowa St: W 35-7
Michigan @ Notre Dame: W 28-24
Ohio St vs Miami: W 36-24
Ohio St vs Ohio: 43-7
(basically the Big Ten just has three noteworthy OOC results - Iowa vs ISU, and Ohio St's two results, and the rest really are just filler)

And five worst:

USC vs Virginia: W 17-14
Washington @ BYU: L 17-23
Washington vs Nebraska: L 21-56
Wash St @ SMU: L 21-35
Wash St @ OK St: L 17-65

Miss St vs UAB: W 29-24
Tenn vs Oregon: L 13-48
Tenn vs UAB: W 32-29 (OT)
Vandy vs NW: L 21-23
Vandy @ Uconn: L 21-40

Big Ten:
Indiana vs Arkon: W 35-20
Indiana vs Ark St: W 36-34
Minn vs NIU: L 23-34
Purdue vs Notre Dame: L 12-23
Purdue vs Toledo: L 20-31

In terms of the five worst games, the model rates the Pac-10 and SEC as having little difference. The Big Ten's five were rated worse, mainly because all five were quite bad, while for the Pac-10 (USC's win vs UVA, Washington's loss to a very good Nebraska team) and the SEC (Miss St's regulation win vs UAB, Tenn's loss to a very good Oregon team) two of the five really weren't THAT bad.

In terms of the five best games, again the Big Ten falls well behind, since really they only have three of note. The model rates the Pac-10's five best over the SEC's five best because the Pac-10's list of five was against tougher opposition (ULM, SJ St and Duke are all quite bad, and neither Penn St nor Houston were especially noteworthy foes). If you'd rater make a list of the top five opponets that the SEC team (probably UNC, WV, Clemson, A&M and Louivsille), then the difference was that all five of those were within a touchdown, including a 3-point overtime win, while the Pac-10's list generally consisted of appreciably more convincing wins. It's also worth noting that for both the list of five best and five worst, there were a substantial number of road games (which is, in fact, true of the Pac-10's slate in general; compared to the SEC, Big Ten and Big 12 their proportion of OOC road games is MUCH higher), and that also gets factored in.

I should re-emphasize that this is a VERY simplistic analysis of the results. However, it does represent some of the system's logic, and hopefully helps clarify what it's doing a bit.

8) The following teams are ranked much higher by the model than the BCS: Nebraska, Arizona. Both of these are teams with very tough schedules, and who have had close losses and blowout wins. The model rewards both types of things, because those types of teams tend to be a lot better than their records.

9) The following teams are ranked much lower than the model than the BCS: Auburn (BCS 1st), Michigan St (BCS 5th), Wisconsin (BCS 10th). Auburn has already been covered (see notes 3, 4) so I won't go back into that one. Michigan St has had a schedule rated far below most of the top teams, right around 60th. This shouldn't be surprising, since their non-conference slate was really bad (only Notre Dame was at all decent; FAU and WMU were cupcakes, and that's not even counting their AA game - moreover, all of their OOC games were at home), and the Big Ten doesn't look very good this year. And in case you think that the compu-picks rating is out there, remember that the Spartans are nearly full touchdown underdogs to Iowa. It's pretty clear that Vegas doesn't take this team very seriously either.

Wisconsin's issue really isn't the schedule (it isn't fantastic, but it's in line with most of the upper-level teams right now), but with their dominance (or lack thereof). They've had not one but two 1-point wins (ASU and Iowa), as well as a 13-point win against SJ St (which is the closest SJ St has come to a 1-A win so far this year). In fact, their most dominant win was only by 20 points, at UNLV. Instead of consistent dominance, with perhaps the occassional close win, Wisconsin has been consistently close to mildly dominant, without a single really dominant showing.

10) When it comes to the worst team in the country competition, Akron is just kicking ass and taking names. They've combined a crappy schedule and getting their asses kicked week in and week out. The difference between them and #119 New Mexico St, while not insurmountable, is pretty wide for this stage in the season. However, Akron still does have games against fellow cellar-dwellars Ball St and Buffalo, so there could conceivably be some movement if they can pull an upset. Based on what we've seen from Akron so far, though, I'm thinking that's a pretty big reach.

11) Boise and TCU flipped after arguably Boise's worst performance of the year. A 23-point win over Louisiana Tech is a good showing... but far from a great one, which some of Boise's previous results have been. Before, they were basically tied, with Boise having a slight edge. Now, they're basically tied, with TCU having a slight edge. Ultimately, though, the real story is that both remain substantially above the rest, as opposed to which team is slightly ahead this week.

12) 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 cfn_ms@hotmail.com

Related Stories
Stoops happy with team
 -by WildcatAuthority.com  Oct 26, 2010
UCLA by the numbers
 -by WildcatAuthority.com  Oct 26, 2010
Alabama Football Goes Back To Work
 -by BamaMag.com  Oct 26, 2010