I’m going to switch it up and put my top 25 last, since that’s old news at this point. But you can always look up my new ratings early Sunday morning and all week long via that link.

Saban vs. Muschamp and SEC notes

Also, I don’t think we really learned anything new last week. I’m not saying I predicted every game to happen as it did, but there was nothing earth-shattering. I do think the SEC should have quieted a bit of the talk about it being some kind of fraud illusion. I’ll get back to that after mentioning the Muschamp/Saban dispute, both of whom I remember well as LSU coaches of course.

Florida is a solid #3 now in my top 25. If Alabama beats Georgia, I would honestly rather see the Gators in the title game to Notre Dame. At least the Tide would have a greater chance of losing. But a certain head coach doesn’t even think Florida deserves the Sugar.

“You play your way into the (SEC) championship game, which means you’re the best team in your division. …”

Does it always mean that? Really? We wouldn’t to give a one-loss team that didn’t win its division the benefit of the doubt in some cases, right?

And Florida actually tied for its division, kind of like Alabama would have done with Texas A&M (which would have been followed by Alabama watching from home and hoping for a Sugar invite) had Florida not beaten Texas A&M. I hardly think it a coincidence that the two of the top-6 SEC teams who are playing for the championship had the weakest interdivisional competition (Alabama played Tennessee and Missouri, while Georgia played Auburn and Ole Miss).

So let’s look at top wins and losses by either potential loser against Florida. I list all the wins over teams .500 or better in BCS-conference competition. And don’t forget that you’re adding a second loss to either Georgia or Alabama.

(you'll have to go here to check out the table I made)

How do you NOT pick Florida?

Muschamp offered to play in the game in lieu of Alabama, and implicitly would have let the Tide have the Sugar Bowl. I’d probably rather see that game, actually.

Of course, I indicated above two of the SEC’s wins over the ACC over the weekend. Also, South Carolina beat Clemson and Vandy beat Wake Forest. The fact that Vandy not only won but was expected to win and won easily shows how far they’ve come. Being in the middle of this league (#7 based on wins and losses in conference) is hard to do. Vandy did lose to Northwestern earlier in the year though. No other such losses by the SEC top 8 though.

Full post

There has been some big news in how college football is going to handle the end of the season starting in a couple of years (Also check out this post on my ratings site with top 4 lists from the past few seasons), but I can’t help but think about the end to the most-recent season.

As you might have guessed, I’m still slightly traumatized by the way college football ended 6 months ago. At first, I couldn’t even listen to “Sweet Home Alabama”. While I’ve gotten over that, I was still moderately offended by what appeared to be a houndstooth wall in a hotel room I stayed in recently, for example.

I also don’t like that Monarchy of Roses song by Red Hot Chili Peppers (which mentions a “crimson tide is flowing”), but that might just be because it’s not a good song.

It’s not just because I’m an LSU fan. It’s also because of the special regard I have for the University of Alabama. I’ll explain. Unlike most other teams, LSU does not have an unquestionable #1 rival. The most equally reciprocated rivalry is probably that with Arkansas, but I think Hogs fans would still rather beat Texas despite that being an irregular rivalry in the last couple of decades. Texas of course seems more interested in their rivalries with Texas A&M and Oklahoma (or I guess I should use past tense in the former case). I don’t know if I’m representative of most LSU fans, but if the Tigers could only beat one team all year, I would choose Alabama. I don’t really care if Alabama regards its rivalries with Tennessee and Auburn as more important. I’m generally happy to cheer for them to lose to the Vols or to those other Tigers too.

Something else that bothered me was that LSU had successfully navigated the great SEC without a single loss. This included 8 regular-season conference games and a game against a ninth team in the SEC championship. Also, the Fighting Tigers had beaten Oregon and West Virginia, who each went on to win BCS bowls, and had not lost out of conference.

Despite the change of heart by the voters in 2006 to avoid a rematch scenario, I knew it would happen one of these years, but for this team, my team, to have to play a team they had already beaten on the road in order to claim a championship, that especially wasn’t right, even before knowing the result. If LSU had lost a game to another team, then I would have had absolutely no problem with it. But in the ONE game to win a national championship (after already winning 13), they had to line up against this same team again? You can’t pretend that’s the same thing as playing and beating a new opponent.

I also wasn’t a stranger to history. I knew that although LSU beat Ole Miss in 1959 on Billy Cannon’s historic punt return, Ole Miss would win the Sugar Bowl (by a score of 21-0) in a rematch.

Billy Cannon on his way to the end zone on October 31, 1959.

Continue reading (I also cover situations in recent memory where a two-team solution didn't work)

I don’t anticipate having much blogging time this week as I haven’t been feeling well, and I don’t have the type of job where I can just call in sick with a cold, so that’s why I’m doing this now instead of on Monday (or later) as I have been doing.

I’ll start with an anecdote I found interesting. In my first mathematical rankings, Maryland was only one spot behind Ohio St. Now, they are separated by 68 spots, and Ohio St. isn’t even in the top 25.

I anticipated Alabama might fall a spot with the bye week, and that’s what happened, as Oklahoma St. jumped into the #2 spot.

Announcers talk about how “the computers love Oklahoma St.”, but computers do not have feelings. It just so happens that ULL has only one other loss (albeit to a weak team), and Tulsa has only two other losses (to Oklahoma and Boise St.; there are maybe ten teams in the country who wouldn’t have three losses against that schedule). Arizona and Kansas are quite bad, but those are two reasons LSU has managed to come out ahead of the ’Pokes in my ratings. On the other hand, Kansas and Arizona have the best two schedules thus far according to my preliminary SoS, so that can keep those two games from hurting Oklahoma St. very much, and as a result, it’s not surprising many ratings have the ’Pokes #1. It is possible they might be #1 in my ratings if Alabama beats LSU. It would depend upon how past opponents of Oklahoma St. and Alabama do. But I would anticipate that my ratings will still have the “right” top two either way. By the way, my ratings agreed with the BCS as far as the top two teams entering the bowls in 2008, 2009, and 2010.

There are a couple of things to keep in mind that cut against Oklahoma St. They will have a bye week, and any team which plays in the SEC Championship game will not have a further bye week, so that’s a chance for the SEC team to come out #1 even if they trail after next week. Also, it would not be a surprise if the value of the wins over Tulsa and ULL were to diminish based on performance in the upcoming weeks.

None of the above is to say that the ’Pokes win Bedlam anyway (or beat K-State and Texas Tech on their way there…I do think it’s safe to say they’ll beat the Cyclones though). But anyone can win LSU-Arkansas or Alabama-Auburn as well. Barring some kind of catastrophic injury or suspension situation, I struggle to imagine LSU or Alabama losing between next week and those respective games. (Recent LSU and Alabama teams have failed to show up in character for such games, but I haven’t seen that quality in these editions yet.) LSU will face Western Kentucky and Ole Miss, while Alabama will face Mississippi St. and Georgia Southern.

The SEC Championship game could also be interesting. It could be a South Carolina team with only a three-point loss to Auburn in Week 5 or a Georgia team on a 10-game winning streak. Since that last loss was to South Carolina, Georgia could finish with two overall losses and actually fail to make the championship game like it did in 2007. South Carolina plays @Arkansas this weekend, so that may chance this fact.

Oklahoma could leapfrog one or more undefeated teams by beating Texas A&M, although it would have potentially helped the Sooners more had the Aggies beaten Missouri. The following week, Boise St. and Stanford have what will probably be their best remaining point opportunities against TCU and Oregon, respectively.

I don’t think Clemson has much of a shot, but if they beat a once-beaten South Carolina team followed by a once-beaten (by Clemson) Virginia Tech team, Clemson could have an argument with some help.

What follows are 10 other teams, one undefeated (Houston) and the rest with one loss each. Those who have a chance to knock off the top 7 are in a better position than those who do not. The next 7 each have two losses, and #25 North Carolina has three losses.

Like with the BCS, it’s not always a good idea to pick the teams with fewer losses to win, but those teams do tend to be higher since it’s a system designed to pick the best two teams. If two teams have even somewhat similar schedules, you want the team with the better record to come out on top for #1 or #2. I don’t care so much about #14 really being better than #15, for example. By the way, I don’t think they are.

Top 25

Full 120

In case you missed it:

Week 6 Top 25 Blog

Full ratings, Week 6

Part 1 of this series

I won’t write as much for the other conferences, so I plan on combining some of those.

The SEC and ACC both arose out of the Southern Conference, which split before the 1933 season. I would take three current ACC teams and add them to the current SEC. Since Texas A&M seems in the SEC already, I included them as well. I did not include Missouri, because although they might make a good rival for Arkansas, they don’t really fit overall. The three ACC teams (Georgia Tech, who used to be in the SEC; Florida St.; and Clemson) would all be in existing SEC states, which doesn’t fit the expansion blueprint, but one thing I would explicitly want to do is keep conferences regional.

I put the most thought into this grouping, so I had a couple of different ideas. One would maintain the East/West idea, but what I did there was add Florida St. to the West (which apparently will already gain Texas A&M). That might not make sense at first blush, but the Western border of the SEC East already runs from Nashville to Gainesville. Both Auburn and Tallahassee (along with the rest of the SEC West) are to the South and West of that line. Ga. Tech (Atlanta is slightly to the North and East of the line I mentioned) and Clemson would be added to the East.

I have the teams in an order so that suggested interdivisional rivals line up. So this is the proposal maintaining the East/West system:

East-West

Florida-Florida St.

S. Carolina-Arkansas

Ga. Tech-Texas A&M

Georgia-Auburn

Clemson-Miss St.

Tennessee-Alabama

Kentucky-LSU

Vandy-Ole Miss

Florida/Florida St. is obvious. The only problem there is the traditional Florida-LSU rivalry (which has been played for 41 straight years…and more times total than Florida-Florida St.) wouldn’t be part of the system. But when the SEC started, each team had two interdivisional rivalries. LSU’s second rivalry had been with Kentucky, whom they had played for 51 consecutive seasons before the format changed. So I started that one again. I don’t think Kentucky’s current permanent rival, Mississippi St. (there have been 37 games total in that series) is a big deal. I’ve maintained Alabama-Tennessee, Ole Miss-Vandy, Auburn-Georgia, and Arkansas-South Carolina. The other new ones are Texas A&M-Ga. Tech and Miss. St.-Clemson. No particular reason for that, but I thought Texas A&M-Clemson was too far apart. Arkansas-South Carolina is too far apart too, but that has been played since the SEC first expanded to 12 teams, and nothing else jumps out as making a whole lot more sense. Arkansas could play Vanderbilt, and that would make for more similar travel distances, but Vandy/Ole Miss is a much better natural and traditional rivalry. Auburn-Ga. Tech (played every year but one from 1902 {or earlier} to 1987) used to be a big deal, but since that hasn’t been played annually for some time (there have only been 2 games in the series since), I thought it best to leave Auburn-Georgia intact.

As far as the ACC teams, Florida St. has only been in the ACC for the last 20 years, so I didn’t think it was too traumatic to move them away from Georgia Tech and Clemson, who are more natural rivals of each other of course. After Florida and South Carolina, Auburn is Florida St.’s most commonly-played SEC opponent (18 prior games). Florida St. has also played Miss. St. and LSU 9 times apiece.

I came up with a North-South format, which also might make sense if Missouri is added by the SEC...

Continue reading

Took me most of the day, but I got it all done finally. I am a little disappointed because I feel like it’s a step back from what I was doing with the subjective rankings (I’ll explain), but it was finally at the point where it was too hard to do a fair subjective ranking anyway. I had the teams arranged in a pretty neat way with the winning teams ahead of the losing teams, but now that’s gotten more complicated. It makes a lot less sense to have Temple between Penn St. and Maryland, for instance, and there is of course the triangle of impossibility with South Florida beating Notre Dame, who beat Pitt, who beat South Florida (handily). Those are just a couple of examples where I thought, “What would I do with this information?”

I don’t see any glaring errors (although I was able to find some), but there usually are some mistakes at this point. Let me know if anything seems ridiculously out of place (like an undefeated team being 80th or a team with one win being 40th or something of that nature). I have one area where I type in the record and another area where wins go in one set of columns and losses go in another, so if that doesn’t match up, it causes really strange results. Sometimes the ratings comparison gives me a heads up when I realize I have the highest or lowest ranking for a given team.

I wish teams with respectable losses were higher (and undefeated teams with bad schedules were lower), but it will come around. Right now in most cases if you have on loss, you’re 25% behind in the winning percentage. Two losses, you’re probably 50% behind or at least 40% behind. When it gets to be more like a 10 or 15% difference with each loss, that will allow some of the good record/poor schedule teams to move down.

These rankings are made with the emphasis on having the top 5 to 10 teams in the right order at the end of the season. I give teams some amount of credit based on winning FCS games, and it depends in part on that team’s record. It doesn’t amount to much at the end of the season, but with 1/3 to 1/5 of opponents being FCS for a lot of teams, it does count for more now, so there can be some weird results because of that. Also, if a team has a bye week and an FCS opponent at this point, that makes it more likely that team has gotten by without playing anyone. So even if it’s a team that will likely finish with 4 or 5 wins, they might look good mathematically right now. So not only do my ratings not predict future events, but future events are needed in order to make my ratings look better.

This also isn’t a good barometer yet for teams that have played and lost tough games. Oregon, for instance, would have been better off beating another FCS team than losing to LSU. The Ducks only have intra-subdivision wins over Nevada and Arizona. The only reason Oregon is as high as they are is because Nevada has had a good schedule. I don’t think Arizona will finish winless against FBS teams, so when they win a game, that will pick the Ducks up a little more, and Nevada’s record should improve at least to the vicinity of .500.

The system’s limitations on giving Oregon its due have also affected LSU. In the by-the-numbers ratings, LSU is 4th because the only win that comes across as being of very high quality is West Virginia. Along with Oregon, Mississippi St. doesn’t count for too much because the Bulldogs have only beaten Memphis and Louisiana Tech (in OT at home, which gives Miss. St. even less credit). But again, if they turn their record around and get some wins of higher quality, this will help out LSU.

I treat #1 as a special case, and as is typical, I leave the team I have at #1 unless there is something at least troubling that happens. An example is the game USC nearly lost at Washington in 2007 (the Trojans won, 27-24, and lost to Stanford the next week). LSU just beat Kentucky 35-7, with the 7 coming as the last score of the game, so nothing troubling there. And as I just explained, I think they’ve beaten quality teams, it just hasn’t come across in the numbers yet. But my ratings site is always going to be the exact numbers the formula gives me.

I’ll at least leave LSU there for a few weeks unless I think the team that rates #1 is either equally deserving or more deserving. If LSU loses, that will also cause me to lean toward the by-the-numbers #1. Not only do the Tigers face Florida next week, but Oregon will play Cal and Arizona St. in the next two weeks. Mississippi St. faces UAB (not a good team, but a needed chance for an FBS win) and South Carolina. Getting into the conference schedule already helps out the stronger conferences in general.

Continue to ratings...