SportsRatings: College Football

Will Saturday help sort out the Pac-12? Probably not

The Pac-12 has more legit Playoff contenders than any other conference, and it's not even close. Take a look:

Rank Team                   Conf  Rec  %Odds  This weekend • Line
  5. Southern Cal           P12   4-0   31.3 @Colorado      -21.5
  6. Washington             P12   4-0   30.6 @Arizona       -18.0
  9. Oregon                 P12   4-0   28.2  Stanford      -27.0
 11. Utah                   P12   4-0   12.2 @#13 Oregon St  +3.5
 13. Oregon St              P12   3-1    9.5  #11 Utah       -3.5
 20. Washington St          P12   4-0    3.7  idle
 24. UCLA                   P12   3-1    1.5  idle

USC, Washington, and Oregon are all legit candidates for best team in the nation, and Utah might not be far behind if they get Cam Rising back. Meanwhile Oregon State remains a Playoff contender despite their loss to upstart Washington State, and UCLA can't be counted out either.

With all these teams in contention,will Week Five help us sort things out? Unfortunately, the answer is mostly no.

USC, Washington, and Oregon are all huge favorites. So unless a crazy upset happens—Colorado beating USC, anyone?—nothing's going to move here.

Meanwhile the longer shots, Washington State and UCLA, are idle so the needle won't move on them.

That leaves one game that will give us some answers: Utah at Oregon State.

It's a big one, though. It either damages Utah greatly before they can reach their full potential with Cam Rising, or it eliminates Oregon State from the discussion. Two losses in a league with so many contenders is a death blow at this stage.

Oregon State is a slight favorite. If the Beavers win it revives their hopes a bit, but mainly crushes the Utes' hopes.

What can we learn from the other games?

A bit.

Is Oregon the #1 team in the nation? Or is it Washington? Our early power ratings (All-Division and FBS-only) put those two at #1, respectively. We won't be able to make any definitive declarations seeing them against Stanford and Arizona, but it's more evidence, especially if one or the other struggles.

Does Washington have the #1 offense in college football? Or does USC earn that distinction? Or maybe Oregon? They're 1-2-3 in Adjusted Offense so far. Maybe we'll get some separation.

How bad is USC's defense? It ranks in the 90s early on, how accurate is that? They can prove a lot vs. Colorado's strong offense—like Oregon did—or get into a shootout that proves they haven't made progress from last year.

Will Oregon have a letdown after last week's high-emotion win? Stanford is a solid rival but it's hard to imagine their coach giving the kind of pep talk before the game against the Cardinal.

Will Washington thrive in their first* road game (*against an opponent who isn't in shambles from losing their coach a la Michigan State) ?

The Pac-12 has proven itself against the rest of the country already, but now its time to sort things out internally. We won't learn much this week without some upsets, but it will at least get the ball rolling.

Will Georgia be exposed at Auburn?

The Georgia Bulldogs haven't been playing like a #1 team so far in 2023. That's pretty clear. But how good are they, and what will we learn from the Auburn game?

How Good (or Bad) Are the Bulldogs? A No-Bias Look...

Our Strength power rating for FBS games only puts Georgia at....#43. 

  #  Team                        rec   Overall      Strength     Success
 42. Georgia Tech                2-2   11.69 [ 43]  10.59 [ 42]   1.09 [ 56]
 43. Georgia                     4-0   13.90 [ 41]  10.27 [ 43]   3.62 [ 17]
 44. Coastal Carolina            2-2   10.27 [ 45]   9.35 [ 44]   0.92 [ 61]

Yes, #43.

That puts them in the company of Georgia Tech and Coastal Carolina. Seriously?

Let's take a breath here. This is after just four games—and, for almost every team, just three FBS games. And this is with absolutely no bias from pre-season rankings. There's not enough information to make a conclusive ranking for any team. Just a generalization, and a generalization that might be completely off.

So let's look at the All-Division Strength power rating, which includes every game. That gives four data points, which is much better than three.

Here, Georgia is #23.

  #  Team                        rec   Rating Offense[rnk]  Defense[rnk]
 22. UCLA                        3-1   63.18   52.72 [ 58]  -10.45 [  8]
 23. Georgia                     4-0   62.72   53.17 [ 55]   -9.55 [ 11]
 24. Louisville                  4-0   62.01   63.38 [ 14]    1.36 [ 67]

That puts the Bulldogs in the company of UCLA and Louisville. A bit better than Georgia Tech and Coastal Carolina, but still not the kind of company we'd expect. But at least they're in the top 25!

What is going on here? Well once again, it's a small sample size. Four games is much, much better than three, but every other ranking is still using prior information from last season and pre-season judgements to make their calls. I like to show early, unbiased power ratings to demonstrate how teams are doing independent of that bias—with the caveat that many of these "judgements" will be wrong.

Georgia is clearly one case. As you can see, their Offense is only ranked #55, while the Defense is #11 (which is still low compared to last year's #1 ranking)

But there is something to look at here. Georgia's offense has definitely not looked very good this year. They scored 45 points on Ball State, about the same as Kentucky and Georgia Southern did. They scored 24 on South Carolina, less than North Carolina and Mississippi State did. And against UAB they put up 49, but once again, Georgia Southern scored the same amount. They clearly aren't distancing themselves from the pack offensively. 

And in nearly every game, it's taken their offense some time to get going. They scored 7 in the first quarter on Tennessee-Martin; 0 points in the first quarter against Ball State; just 3 points in the entire first half vs. South Carolina; and just again in the first quarter vs. UAB.

That's just 17 total points in the first quarter, or 4.25 points on average in the first 15 minutes. That's bad.

And our early power ratings aren't the only ones flashing alerts. ESPN's FPI has an "Efficiencies" tab which clearly shows no pre-season bias to its calculations, as Georgia is only #21—very similar to our unbiased results—with the offense ranked #24 and defense #19. They also break out the special teams, which comes in at #124.

So is special teams a big part of the problem? Maybe, but most people point to injuries. The offense has been riddled with them since the start of the season and it's only gotten worse. Presumably, as they heal up the offense will improve. 

They're also breaking in a new quarterback, and we've seen many top teams struggle on offense this year with a new QB. And there's the issue of motivation as the Bulldogs face a poor early slate of opponents after winning back-to-back national titles.

So How Good Are They Really?

Early power ratings are much more accurate when they do retain pre-season bias. And there, the news is mixed but much better for the Bulldogs. SportsRatings Bayesian model, used for our College Football Playoff Projections, puts Georgia at #6. That's why they're at #4 in our Playoff Projections, and still favored in every game going forward. 

ESPN's FPI lists Georgia at #5, and Bill Connelly (also at ESPN) has the Bulldogs at #2.

But not #1, which is where they remain in the AP poll with 55 of 63 first-place votes.

I think it's clear that so far, Georgia has not played like the #1 team. They've played like a borderline top 5 team. And the Auburn game will tell us a lot about where they should be ranked.

Auburn's defense looks strong so far this year, ranking #6 in our early power ratings as well as the FPI's efficiency rankings. Georgia's offense can prove it's worth in its first road game vs. the Tigers.

A solid win over the Tigers on the road would answer a lot of questions. Of course, a bad performance, especially a loss, would call into question whether Georgia has a good enough offense to be ranked even in the top 10.

It could be the game that exposes Georgia. With two strong defenses and (so far) weak offenses, it could be a very low-scoring game, which could give Auburn a shot at winning near the end. Will this be the game that proves the nay-sayers correct about the Bulldogs?

Even a narrow win won't convince anyone that something is wrong. Get healthy, people will say, and they'll be ok. And the defense will carry them through most of their schedule anyway.

But five games in a row with poor offensive production can't be ignored. If Georgia doesn't come alive on offense against Auburn—a rival—then what will motivate them?

(Note: A shitty AI article on CBS Sports claims that last year's Auburn game was a 63-56 Georgia win, wtf CBS Sports, you suck. It was 42-10, where the hell did you get that wrong score? Oh, it's the overall record in the series. Get an editor ffs!)

The other possibility—a Georgia blowout win—would assuage a lot of worries in Athens. Don't think the fans don't see the struggling offense. And they aren't buying the injury excuse, either. Many teams have injuries but still dominate their opponents, especially if they have national title hopes. To be fair though, there have been some big-time teams with poor results on offense this season.

For example, Michigan's first four games aren't exactly amazing on offense, scoring just over 30 points each time out. And Utah has been terrible, averaging 22 points in four wins. Michigan was without Jim Harbaugh on the sidelines for the first three, and looked a bit better vs. Rutgers, but the jury is still out. Utah is without starting QB Cam Rising, so they have a good excuse but still should be doing better than they are.

But Georgia is supposedly the #1 team and they aren't playing close to it. I'm not convinced. They have two more games to show that they're up there with the best before I start believing the power ratings, and Auburn's defense—on the road—is the kind of test they need to show that they haven't faded.

Georgia is a 14 point favorite at Auburn. If they beat that spread then it will be convincing, because our early power rating—the one where they are #23—only shows them scoring 14 points vs. Auburn!

Something's gotta give. Either the power ratings will be forced to re-adjust...or, people's expectations for Georgia will need an adjustment.

2023 Week 5 Preview: no FCS, but many idle, and many mismatches

Week Five is upon us...and along with better games due to conference play, it's the debut of the dreaded "idle" opponent. Also known (after the fact) as DNP, 6 of our top 25 "Playoff Most Likely" teams have an open date this weekend, including two of the top three. Even those who play don't look to be in much trouble, as the entire top ten is favored by double digits. But there are some good games, even among the projected blowouts. 

Ohio State and Florida State don't play. Still, their odds will change based upon what other teams do, which will be interesting to see and analyze. It would be better if they played a game though. #2 Penn State plays at Northwestern and is a near 4-TD favorite, so they are more likely to ding their odds slightly. But a huge performance might put them past the Buckeyes for the #1 spot. Which hardly matters now since the teams play later in the year.

Georgia vs. Auburn is an interesting game because it will show what the Bulldogs do against an opponent with a pulse and on the road. The first four weeks have been underwhelming and that's why the Bulldogs are Playoff-borderline right now, leading a six-team logjam for that 4th spot. Georgia is a 2-touchdown favorite.

Another interesting game with a big spread is #5 USC at Colorado. Everything Colorado does is interesting this year and USC's defense might make it more interesting. They allowed 28 points by Arizona State which sets the stage for a potential shootout game. #6 Washington at Arizona has less fun potential, obviously but it will be good to get more data on the Huskies, who rank #1 in the inaugural unbiased FBS Overall Power Rating. 

Sportsratings Playoff Odds for 9/23 with 9/30 weekend opponent and line

Rank Team                   Conf  Rec  %Odds  This weekend • Line
  1. Ohio State             B10e  4-0   41.8  idle
  2. Penn State             B10e  4-0   41.4 @Nwestern      -27.0
  3. Florida St             ACC   4-0   36.4  idle
  4. Georgia                SECe  4-0   33.0 @Auburn        -14.5
  5. Southern Cal           P12   4-0   31.3 @Colorado      -21.5
  6. Washington             P12   4-0   30.6 @Arizona       -18.0
  7. Texas                  B12   4-0   30.1  #25 Kansas    -16.5
  8. Oklahoma               B12   4-0   28.7  Iowa State    -20.0
  9. Oregon                 P12   4-0   28.2  Stanford      -27.0
 10. Michigan               B10e  4-0   16.5 @Nebraska      -17.0
 11. Utah                   P12   4-0   12.2 @#13 Oregon St  +3.5
 12. Alabama                SECw  3-1   11.3 @Miss State    -14.5
 13. Oregon St              P12   3-1    9.5  #11 Utah       -3.5
 14. Mississippi            SECw  3-1    6.0  #18 LSU        +2.5
 15. Duke                   ACC   4-0    5.1  #16 Notre Dame +5.5
 16. Notre Dame             IND   4-1    4.8 @#15 Duke       -5.5
 17. Syracuse               ACC   4-0    4.7  Clemson        +7.0
 18. LSU                    SECw  3-1    4.5 @#14 Ole Miss   -2.5
 19. North Carolina         ACC   4-0    4.2  idle
 20. Washington St          P12   4-0    3.7  idle
 21. Louisville             ACC   4-0    3.2 @NC State       -3.0
 22. Tennessee              SECe  3-1    2.9  S Carolina    -12.5
 23. Kansas St              B12   3-1    2.4  idle
 24. UCLA                   P12   3-1    1.5  idle
 25. Kansas                 B12   4-0    0.8 @#7 Texas      +16.5

Texas hosts #25 Kansas; both teams are 4-0 but the Longhorns are big favorites. It's a big opportunity for the Jayhawks whose Odds% will soar (to 10% maybe? hard to say) if they win, and Texas' odds would probably fall to 10% if they lose. Fellow Big Twelve team Oklahoma hosts Iowa State, a disappointment so far this year. This looks like it could be a 45 point blowout but half the Sooners games have been 27-10-ish and that could happen here too. #9 Oregon hosts Stanford, an even worse mismatch on paper. Oregon and Oklahoma are #1 and #2 in our early, unbiased, All-Division Strength Power Rating.

And #10 Michigan is, like Georgia, struggling on offense, or at least struggling to put up the kind of gaudy numbers expected of them. The defense is killin' it, though, and given Nebraska's struggles it could be a shutout. Our power rating has it 21-0, Wolverines.

Now we get to the good stuff: #11 Utah at #13 Oregon State.The Utes have struggled a bit without Cam Rising at QB and their odds are barely higher than the Beavers' despite being undefeated. And Oregon State is the slight favorite, too. The Beavers need this game to remain alive in the Playoff race, obviously. It's hard enough for a 1-loss Pac-12 team to make the Playoff let alone 2-loss; though the conference is solid this season, 2-losses has never made the cut.

Alabama goes on the road vs. Mississippi State with their offense averaging 22 points over the last three games. They need to avoid their 2nd loss obviously, as do #14 Mississippi and #18 LSU who meet in Baton Rouge with the Tigers a slight favorite.

Duke and Syracuse will try to remain undefeated and stay tied at the top of the ACC but both have tough assignments. The #15 Blue Devils host Notre Dame and are a 5 1/2 point underdog, while Syracuse hosts Clemson and are a touchdown dog. Does the dream end here for both? These are some of the only undefeated  Cinderellas left in the country so let's root for them. They don't play each other so they'd both be 12-0 in the ACC title game! Add 4-0 Louisville to that list, too, though they play Duke so they can't all be 12-0. Note that Miami could be 12-0 too, which seems like a big flaw in the no-division system, if you can have three 12-0 teams somehow and one is left out of the conference title game. I guess they're just assuming the tiny odds of that occurring means they don't have to worry about it.

Incidentally Miami, #26 in Playoff Odds, will remain undefeated this weekend as they also face the dreaded "idle".

Week 4 Unbiased Power rating: When will Georgia (#23) surge?

Four weeks into the college football season and the power ratings are starting to come together—on the whole. A few key teams, however, are lagging so far. Either because the power ratings aren't mature enough to capture their quality...or, maybe these teams aren't as good as their reputation.

Let's take a look, comparing the top teams to last week's ratings.

  #  Team                        rec   Rating Offense[rnk]  Defense[rnk]
  1. Oregon                      4-0   86.00   73.01 [  3]  -12.98 [  4]
  2. Oklahoma                    4-0   83.26   72.09 [  4]  -11.17 [  7]

Oregon, last week's #3 team, is on top after butchering Colorado 42-6. That really helped their defensive ranking, which was #57 a week ago. Note that early in the year there are BIG jumps in the ranking components; due to the small number of games, every contest has huge implications. For example, Oklahoma debuted at #1 last week and though they fall to #2 their own defensive ranking improves a lot, from #40 to #7 after beating Cincinnati 20-6. Their offensive ranking only fell from #2 to #4. Both teams are looking very good now, with top ten offenses and defenses. It's still early though, and the bigger tests are down the road.

  3. Washington                  4-0   81.08   75.59 [  1]   -5.48 [ 27]
  4. Penn State                  4-0   78.80   70.40 [  6]   -8.40 [ 17]

Washington held its lofty spot pretty well, too, moving up from #4 to #3. Often the top teams are exposed and drop a lot early on, but the Huskies beat Cal 59-32 which gave them the #1 offense and didn't ding their defense much. Penn State also moved up a spot after beating Iowa 31-0 at home. You'd think that would help their defense rank but it fell from #8 to #17! I guess shutting out Iowa's offense isn't a needle-mover. It's obvious that Penn State's previous foes were the cause of that; neither Illinois, West Virginia, or Delaware scored 30 points last week. Note that Penn State is far and away #1 in turnover margin at +2.75 per game, something that probably won't continue. It's the equivalent of 11 points per game, and if it goes away the Nittany Lions will only be about #15.

  5. Texas                       4-0   76.48   68.13 [  9]   -8.34 [ 18]
  6. Alabama                     3-1   74.34   54.70 [ 45]  -19.64 [  1]

An interesting pairing at #5 and #6. Texas and Alabama both did well last week and moved up 8 and 19 spots respectively. That's right, Alabama was only #25 last week but beat former #8 Mississippi 24-10. That in turn helped Texas move up. The Tide's offense dropped while their defense shot to #1 (Ole Miss was averaging over 50 points).

  7. Miami FL                    4-0   73.52   64.32 [ 12]   -9.20 [ 15]
  8. Texas A&M                   3-1   73.09   69.16 [  8]   -3.93 [ 34]

Another pair of teams that played earlier in the season, with the winner one spot ahead of the loser. Miami beat Temple 41-7 and that was good enough to hold their #7 spot—a good sign since Miami has a recent history of faltering once they're declared to be "back". Texas A&M jumps from #16 to #8, more evidence that they, also, are back.

  9. Ohio State                  4-0   72.86   55.87 [ 35]  -16.98 [  3]
 10. Mississippi                 3-1   72.85   71.45 [  5]   -1.40 [ 47]

Both these teams dropped a few spots from last week, Ohio State due to their too-narrow win while Mississippi lost to Alabama. The Buckeyes offense only scored 17 against Notre Dame but their ranking went up (#44 to #35) and the defense, despite holding the Irish to 14, fell from #1 to #3. This is just part of the overall adjustments due to more connections being made. Interestingly, Mississippi's offense remained at #5 while Alabama's D jumped from #35 to #1. Why didn't it downgrade Ole Miss's offense and keep Alabama's D constant? The preponderance of evidence in the other game scores made the power rating "decide" to do what it did. Normally both are affected equally, but this is how things worked out in this instance.

 11. Southern Cal                4-0   70.35   75.54 [  2]    5.18 [ 97]
 12. Kansas St                   3-1   70.30   66.91 [ 10]   -3.39 [ 36]
 13. Michigan                    4-0   69.32   50.87 [ 75]  -18.44 [  2]
 14. Duke                        4-0   68.58   60.45 [ 21]   -8.12 [ 22]
 15. Notre Dame                  4-1   68.25   62.32 [ 16]   -5.93 [ 25]

USC beat Arizona State 42-28 and the Strength power rating didn't like that, dropping the Trojans from #2 to #11 (this kind of downgrade is common early on, as I mentioned above). Their offense remained about the same (#1 to #2) but the defense—the only question for this team—just got a lot worse, falling from a mediocre #59 to an awful #97. But just as quickly as that happens, they can redeem themselves when they face Colorado Saturday. 

Michigan is also in this cluster, moving up from #24 to #13 after beating Rutgers 31-7. That allowed their offense to make a decent jump from #99 to #75, and that movement will probably continue—it has to if the Wolverines are really a Playoff team. Duke held steady at #14, and Notre Dame moved up 4 spots for playing Ohio State so close. #12 Kansas State is the highest-ranked team with a loss to a lower-rated team (#37 Missouri). Other than that they've been fantastic, but can they be consistent?

 16. Oregon St                   3-1   66.04   56.64 [ 33]   -9.40 [ 13]
 17. Florida St                  4-0   65.19   59.94 [ 23]   -5.25 [ 28]
 18. TCU                         3-1   64.37   61.96 [ 17]   -2.41 [ 41]
 19. Central Florida             3-1   64.31   61.02 [ 20]   -3.28 [ 37]
 20. Syracuse                    4-0   64.29   56.05 [ 34]   -8.24 [ 20]

Oregon State lost to Washington State and fell from #9 to #16 but remain higher than the #29 Cougars. Florida State's overtime win at Clemson cost them 7 spots, and Syracuse also dropped despite beating Army 29-16. UCF moved up a few spots despite losing to Kansas State, and #18 TCU is re-establishing themselves following the opening loss to Colorado; the Horned Frogs beat SMU 34-17. 

 21. Auburn                      3-1   63.42   51.82 [ 66]  -11.59 [  6]
 22. UCLA                        3-1   63.18   52.72 [ 58]  -10.45 [  8]
 23. Georgia                     4-0   62.72   53.17 [ 55]   -9.55 [ 11]
 24. Louisville                  4-0   62.01   63.38 [ 14]    1.36 [ 67]
 25. Cincinnati                  2-2   61.35   52.00 [ 64]   -9.34 [ 14]

Now we come to the real question in our power rating: when will Georgia be properly ranked? Or, are they properly ranked right now? Based on last year's team the Bulldogs are clearly underperforming but to be fair they haven't had a lot to get motivated about. The Bulldogs looked pedestrian again in beating UAB 49-21, falling from #17 last week to just #23. Their offense—#76 last week—moved up quite a bit with the win but their defense took a hit.

If our power rating for Georgia needs correcting it should happen this weekend when Georgia visits #21 Auburn. If the Bulldogs win by two touchdowns like the oddsmakers expect, there will be a solid correcting force in Strength next week. If, on the other hand, Auburn wins by 4 points as our current ratings say, then it will be pretty clear that Georgia isn't the same team any more. I suspect that we'll see the real Georgia emerge as they face a familiar opponent (though they didn't do all that great against South Carolina).

UCLA lost to Utah but remains ahead of the undefeated Utes, who are surprisingly only #32. Undefeated Louisville and 2-loss Cincinnati round out the top 25.

 28. LSU                         3-1   61.06   63.11 [ 15]    2.04 [ 72]
 30. California                  2-2   60.92   56.75 [ 30]   -4.16 [ 33]
 36. Air Force                   4-0   57.73   52.71 [ 59]   -5.02 [ 31]

Three teams dropped from the top 25, with LSU's 3 point win over Arkansas not impressing the algorithm, Cal losing too big to Washington, and Air Force falling from #20 to #36 after beating San Jose State 45-20. The Falcons' 3 FBS opponents are now 2-10 combined, so that was probably an overall re-evaluation due to new information and more connections.

 27. North Carolina              4-0   61.16   55.39 [ 38]   -5.76 [ 26]
 29. Washington St               4-0   60.96   69.85 [  7]    8.89 [141]
 31. Tennessee                   3-1   59.21   59.93 [ 24]    0.72 [ 61]
 32. Utah                        4-0   58.72   49.68 [ 88]   -9.03 [ 16]
 48. Florida                     3-1   53.02   46.90 [116]   -6.11 [ 24]

The five teams above are in the AP top 25 but not in the Strength power rating top 25. Most of them are close, with Florida being the outlier. The Gators only beat Charlotte 22-7 last week which hurt them, but they beat Tennessee and didn't embarrass themselves vs. Utah. All three should move up as schedule strength gets sorted out. 

 33. Colorado                    3-1   58.41   59.98 [ 22]    1.56 [ 68]

Of course I have to take a look at Colorado, who despite a 42-6 drubbing by Oregon moves up from #41 to #33. Much of that is due to TCU's continued good play, and Nebraska and Colorado State also won last week which more than offset the Buffs' loss. Obviously the Buffaloes offense ranking took a hit, falling from #10 to #22, but they surged on defense (#148 to #68). Oregon was averaging 58 points per game, so limiting them to 42 at home counts as good defensive work! Next week they host USC, whose defense looks questionable so it could be more of a shootout than last week's execution. 

 51. Iowa                        3-1   52.77   39.81 [184]  -12.95 [  5]

Many teams in the early power ratings have "skewed" offense and defensive rankings as the numbers tend to push one way or the other. But Iowa is the real deal when it comes to bad offense and great defense. After losing 31-0 to Penn State Iowa's offense dropped from #96 to a horrific #184, while the defense remained about the same. Normally I would chalk this kind of discrepancy up to a small sample of games, but with Iowa the numbers might not be too far off.

 82. Texas-Permian Basin         3-1   45.19   65.57 [ 11]   20.37 [275]

Last week's poster boy for "yes, it's too early in the season to really trust unbiased power ratings" was Texas-Permian Basin who came in at #27 due to a couple of crazy blowouts (96-0 and 86-7). Last week they edged West Texas A&M 41-40 in overtime and naturally that dropped them over 50 spots. But they're still #82, much higher than they objectively should be. It will take some time for their offensive ranking to fall to fair value (it was #4 last week!) while their defensive rank might already be roughly correct. Maybe they should play Iowa?

2023 week 4 Playoff projection: Big Ten East holds top 2 spots; Georgia, USC battle for #4

Hard to believe 1/3 of the season is already over, but most of the big games remain as most Playoff candidates have played only 1 or 2 Power Five schools so far. Week 4 was a big step forward with very few FCS opponents, and the start of conference play for most of the big teams.

 Ranking                             Curr | Projected     | Playoff    | Bowl
 LW 9/23 Team                   Conf  Rec | W-L  W-L(Cume)|%Odds  %Chg |%Odds  %Chg
  3   1. Ohio State             B10e  4-0  13-0  11.3-1.7   41.8  +7.9   99.9  +0.6
  2   2. Penn State             B10e  4-0  11-1  10.8-1.2   41.4  +7.3   99.9  +0.1
  4   3. Florida St             ACC   4-0  13-0  11.4-1.6   36.4  +3.2   99.9   0.0
  5   4. Georgia                SECe  4-0  13-0  11.1-1.9   33.0  +4.0   99.7  -0.2
  1   5. Southern Cal           P12   4-0  12-1  10.3-2.7   31.3  -7.7   99.5  +0.2

At the top of the charts are Ohio State and Penn State, both from the Big Ten East. Usually this situation—two conference teams at the top—means that one or the other will get a bid, especially when they're from the same division. Only one can make it to the Big Ten title game but both look good in terms of beating Michigan. That could mean that the Big Ten East once again gets two teams in the Playoff. Ohio State's nail-biter win over Notre Dame helped their cause a lot, and Penn State's 31-0 blanking of Iowa increased their stock too. Florida State outlasted Clemson in overtime which is a great win to get behind them but their penchant for ultra-close games means they gain only marginally in Playoff Odds. And Georgia slips into the final spot at #4, back after beating UAB 49-21, but once again their performance wasn't dominant enough to move the needle very much. It was USC's tepid showing against Arizona State (a 42-28 win) that cleared a spot for the Bulldogs, as the Trojans dropped almost 8% in Playoff Odds. 

 LW 9/23 Team                   Conf  Rec | W-L  W-L(Cume)|%Odds  %Chg |%Odds  %Chg
  6   6. Washington             P12   4-0  10-2   9.7-2.3   30.6  +3.4   99.5  +1.3
  9   7. Texas                  B12   4-0  12-1  10.7-2.3   30.1  +9.5   99.7  +1.6
  8   8. Oklahoma               B12   4-0  12-1  10.9-2.1   28.7  +1.6   99.9  +1.2
  7   9. Oregon                 P12   4-0  11-2  10.2-2.8   28.2  +1.1   99.7  +1.0
 12  10. Michigan               B10e  4-0  10-2   9.8-2.2   16.5  +2.5   99.8   0.0

USC is the top team in the Pac-12 but there are a lot of good teams behind them, case in point Washington who racked up 59 points on California, and Oregon who crushed Colorado 42-6. Both had high Playoff Odds already and the wins didn't change a lot, in fact Oregon slipped a few spots due to the Big Twelve's Texas and Oklahoma surging. Especially the Longhorns who dominated Baylor 38-6 and are also seeing their stock rise due to their opponents' wins (they are #1 in Success this week). Texas gains almost 10 percentage points this week. Michigan gains after beating Rutgers but as noted, they are starting to look like the odd man out in the Big Ten East.

 LW 9/23 Team                   Conf  Rec | W-L  W-L(Cume)|%Odds  %Chg |%Odds  %Chg
 14  11. Utah                   P12   4-0   8-4   8.8-3.2   12.2  +2.1   99.6  +2.6
 18  12. Alabama                SECw  3-1  11-2   9.6-3.4   11.3  +5.5   98.4  +3.6
 10  13. Oregon St              P12   3-1  10-2   8.6-3.4    9.5  -9.6   97.4  +0.1
 11  14. Mississippi            SECw  3-1  10-2   8.9-3.1    6.0  -9.2   98.9  +0.5

A big part of Texas' gain is Alabama's 24-10 win over Mississippi, which added 5.5% to the Tide, making them the 1-loss team with the highest Playoff Odds (11.3%). A loss hurts more than a win helps, which is why Ole Miss dropped from 15.2% to 6.0% odds. A loss to a team in your own conference division is the worst, as it blocks you from conference title game contention. Alabama's loss—to Texas—leaves their path clear to the SEC title and with it, the Playoff. Two more Pac-12 teams show up here: Utah, who beat UCLA, and Oregon State, who lost to Washington State. They essentially switch positions with Utah now in front, but the Utes have a lot of close games coming up where they aren't the favorite (note their 8-4 straight-up projection).

 LW 9/23 Team                   Conf  Rec | W-L  W-L(Cume)|%Odds  %Chg |%Odds  %Chg
 16  15. Duke                   ACC   4-0   8-4   8.8-3.2    5.1  -1.9   97.1  +3.2
 13  16. Notre Dame             IND   4-1  10-2   8.6-3.4    4.8  -8.4   97.9  +0.5
 15  17. Syracuse               ACC   4-0  10-2   9.4-2.6    4.7  -2.5   99.2  +1.1
 17  18. LSU                    SECw  3-1   9-3   8.6-3.4    4.5  -1.6   96.5  +1.5
 20  19. North Carolina         ACC   4-0  11-1   9.2-2.8    4.2  +0.3   98.4  +4.4
 24  20. Washington St          P12   4-0   9-3   8.5-3.5    3.7  +1.3   97.5  +8.3
 23  21. Louisville             ACC   4-0  11-2   9.5-3.5    3.2  +0.2   98.7  +3.0

The ACC beyond Florida State is pretty fragmented, and now that Clemson is largely out of the picture there are many teams challenging the Seminoles. Duke and Syracuse both moved to 4-0 this week but lose some ground in terms of odds (shooting to the top four and being ranked #15 means your odds should decrease each week). North Carolina and Louisville also are 4-0 after winning conference games and both manage a meager increase. Things won't really change until they start playing each other. Honorary ACC team Notre Dame suffered a big drop, as will happen with your first loss. They lose about 2/3 of their odds and with no conference championship possible, the Irish will be behind any 12-1 Power Five conference champ as well as any 13-0 team obviously. 

LSU's tight win over Arkansas didn't improve the Tigers' outlook, while Washington State's win over Oregon State helped but the Cougars are expected to go 9-3 or 8-4, which won't get them too close. They move up to 3.7% odds.

 LW 9/23 Team                   Conf  Rec | W-L  W-L(Cume)|%Odds  %Chg |%Odds  %Chg
 25  22. Tennessee              SECe  3-1   9-3   8.7-3.3    2.9  +1.2   98.2  +2.6
 28  23. Kansas St              B12   3-1  10-2   8.9-3.1    2.4  +1.4   97.2  +4.0
 19  24. UCLA                   P12   3-1   9-3   8.1-3.9    1.5  -2.5   95.7  +2.4
 31  25. Kansas                 B12   4-0   7-5   7.4-4.6    0.8  +0.2   84.1 +10.6

Tennessee and Kansas State rebounded from bad losses last week, with the Vols crushing UTSA 45-14 and K-State beating UCF 44-31. Both add a small amount to their odds but they'll need to win their big games—and probably need to win out, as no 2-loss team has every made the Playoff. That's why UCLA drops from 4% odds to 1.5%—a loss means you have no more leeway. Kansas moves up mainly because they are still undefeated, while other teams around them drop like flies. They still have less than a 1% chance to make the Playoff but unlike the other teams in this group the Jayhawks control their own destiny.

 LW 9/23 Team                   Conf  Rec | W-L  W-L(Cume)|%Odds  %Chg |%Odds  %Chg
 27  26. Miami FL               ACC   4-0   8-4   8.1-3.9    0.7  -0.4   92.7  +5.1
 32  28. Fresno St              MTN   4-0  13-0  10.4-2.6    0.6  +0.2   99.7  +1.4
 33  29. Maryland               B10e  4-0   9-3   7.9-4.1    0.6  +0.3   91.1 +16.8
 37  30. Wisconsin              B10w  3-1  10-3   8.2-4.8    0.4  +0.2   92.6 +14.2
 26  34. Clemson                ACC   2-2   8-4   7.0-5.0    0.2  -1.5   79.0  +1.5

Some other teams outside the "top 25" include Fresno State, the highest-odds Group of Five team. Now favored in every Mountain West game, the Bulldogs crushed Kent State but face long odds even if they do manage to go 13-0 as their projected SOS is around #119 of 133 teams. 

Two Big Ten teams won big and at least made big gains in their bowl game odds. Maryland crushed hapless Michigan State 31-9 and Wisconsin beat Purdue 38-17. And just by being in the Big Ten and being pretty good, they have a prayer for the Playoff yet.

Meanwhile Miami, despite a 41-7 win at Temple, falls from 1.1% odds to 0.7% odds. I can't really explain it except that the 'Canes are still expected to go about 8-4, and it's a long road from 8-4 to the Playoff. Their odds will only really improve with wins in the games they're expected to lose—UNC, Clemson, FSU, and Louisville. Speaking of Clemson, they still show a meager chance—0.2%—to make the Playoff despite their two losses. 

 LW 9/23 Team                   Conf  Rec | W-L  W-L(Cume)|%Odds  %Chg |%Odds  %Chg
 50  40. Kentucky               SECe  4-0   5-7   6.4-5.6    0.1  +0.1   67.2  +8.5
 34  48. Wake Forest            ACC   3-1   6-6   5.7-6.3   <0.1  -0.3   55.2 -20.0
 63  49. South Carolina         SECe  2-2   7-5   6.4-5.6   <0.1   0.0   69.2 +17.7
 41  64. Cincinnati             B12   2-2   8-4   6.9-5.1    0.0  -0.1   72.9  -3.8
 58  78. Colorado               P12   3-1   5-7   5.0-7.0    0.0   0.0   34.2  -2.4

Kentucky rises from the ashes, the only team to go from near-zero to actual 0.1% odds this week. The Wildcats beat Vandy 45-28 which somehow brought them back above water. More likely it was their 4-0 record, which means they have a theoretical chance of running the table still. Conversely, Cincinnati is the only team to go from detectable odds to 0.0 this week, with the Bearcats loss to Oklahoma their 2nd of the year, likely putting them in too big of a hole to get out of.

South Carolina still registers at well under 1 in 1000 odds despite having two losses, but the Gamecocks' win over Mississippi State is more important for their bowl odds which jumped from about 50/50 to almost 70%. Going the other direction was Wake Forest, whose loss to Georgia Tech not only effectively blanked their Playoff odds but dropped their bowl odds from 75% to 55%.

And of course any run-down has to include Colorado. The Buffaloes never got to a position where they had even 0.1% odds in our system, but you'd think that a 42-6 loss would sink their bowl game hopes a bit more than 2.4%. In this case though the Buffs bowl odds were still catching up with their starting point, which was last year's team. And this year's team is very different. So as new information is digested week by week Colorado's new picture is starting to emerge, and even a 42-6 loss doesn't really offset that. They were projected to be 1-11 at the beginning of the year and are now seen to be 5-7 with a 1 in 3 chance of making a bowl game. Is that still too low? Next week's game vs. USC could radically increase that number, even if the game is a close loss.