Pac-12 Rankings After Week 1 Games

It could be argued that preseason rankings and predictions are based on almost nothing. And it can be argued that even one game is not enough to get a good feel for a team-not even when they play what should be expected to be a good opponent. None-the-less, now that each Pac-12 team has played a game it is time to look at how well those preseason predictions look and at what the impact has been from those games.

Polls

There are now 6 Pac-12 teams that are ranked in the two major polls. Here is a summary of where the Pac-12 teams are and the changes from from the preseason.

Most of the Pac-12 teams that moved up did so because of a couple of bad losses by teams that had been ranked higher (LSU and Clemson). Obviously Colorado moved into the top-25 with their upset win over a ranked team.

Three other Pac-12 teams received votes. UCLA received enough to be #27 in the AP and #29 in the Coaches poll. WSU received enough to be #42 in both polls. And Arizona received one point in the Coaches poll-good enough for #53.

One note on the ranked Pac-12 teams. Each of the teams that are currently ranked will be playing at least 4 games against the other ranked teams. Utah and Oregon each play all 5 other ranked teams. (Corrected to include Utah playing the other 5 teams. 9/7/2023) Compare that to the ranked SEC teams like Georgia who is only scheduled to play 2 of the 6 other ranked SEC teams (and neither game is before November), and Tennessee which is scheduled to play 3 of the other ranked SEC teams.

Overall Rankings

The picture below is a comparison of the final Composite Rankings from last season with the preseason rankings this season and the current rankings after the week 1 (and week 0) games. (https://masseyratings.com/ranks)

As you can see, there is very little change for most of the teams. At the top there has been some slight shuffling (although it is hard to see in that picture). Last year it ended with Utah, Oregon, USC, UW, and then Oregon State (in that order). Currently it is Utah, USC, UW, Oregon, and then Oregon State; they are 8, 9, 10, 12, and 15 overall.

There was some movement up between last season and the preseason for California, Arizona State, and Colorado. California and Colorado continued to move up after their week 1 games. Arizona State, on the other hand, dropped after their too-close-for-comfort win over a FCS team.

Even with the jump given to Colorado there obviously there is still a disparity between these rankings and the AP and Coaches polls (and some other rankings). Some rankings have them as high as #10 while others have them as low as #133. (I checked and the ranking is 133 even after the win over TCU, which the same ranking has at #45.) Neither of those is probably accurate at this point.

Many of the advanced stats rankings (like SP+, FEI, FPI, etc.) use data from the previous season for early season rankings. They do this because there is so little data from this season (so far) and most teams do not change significantly year-to-year. Obviously that is not the case with this year’s Colorado team. The changeover in their roster (and coaching staff) is throwing a wrench into most of the rankings that use data from the previous season. It will likely take another 5 or 6 weeks before those rankings use enough of this year’s data to offset the data from last year. (And I know that there are a number of fans out there that agree that rankings before 6 or 7 weeks into the season aren’t worth much. While Colorado proves this, Colorado is also an extreme that we’ve never seen before in college football.)

The human polls like the AP and Coaches polls, especially after games have been played, tend to use this year more than last year’s data. This shows up clearly when you compare some of the computer/advanced stats rankings with the human polls which you can see in the table below. (Although many fans believe that pollsters just assume teams will be similar from one year to the next-and they likely do to some extent.)

The table below compares one of the advanced stats rankings, F+, with a human poll (I used CBS Sports since they rank all 133 FBS teams). The F+ ranking is a combination of Brian Fremeau’s FEI ratings and Bill Connelly’s SP+ ratings.

I’d argue that the only two that are significantly different are Colorado (duh!) and Oregon State. While Stanford and WSU have greater differences, both are well outside of the top 25 (for now) and the differences aren’t that great relative to where they are ranked. With Oregon State I’d argue that there is a difference between barely being a top 25 team and being a solid top-25 team. All of this will sort itself out as more games are played.

ESPN FPI and Win Projections

One of the computer rankings in that referenced a lot is ESPN’s FPI. They rank the teams, project overall wins, and even project each game.

So far they have gone 12-1 in their projections of the Pac-12 games-missing only the Colorado-TCU game. (To be fair, I think most fans probably would have the same record.) Here is some of the updated ESPN FPI information for the Pac-12 teams and a comparison to their preseason information.

As a reminder, “Favored” indicates that ESPN’s FPI has the team as having a greater than 50% win percentage.

What is interesting in the table above is not the absolute values, but the changes since the preseason.

Only one of the Pac-12 teams is now favored in one of their out-of-conference games that wasn’t favored before: Washington against Michigan State. UW now has a 56.9% win percentage compared to 45% in the preseason. (Most Husky fans would probably argue that ESPN should have had UW as being favored against Michigan State even before the Boise State game.) On the other side, one Pac-12 team is now an underdog against an out-of-conference opponent when they were favored before: Arizona State against Fresno State. Arizona State was at 61.5% and now it’s 36.6%. (Some of the win percentages can change day-to-day so they may be slightly different when you see them.)

But as you can see, almost every Pac-12 team has a change in the number of games where they are favored. That means that there are some changes in the conference games. Here’s a look at those.

The big winners in conference games are Washington and Colorado. Each are favored in two more conference games than they were in the preseason. For UW, those two games are their games against Utah and Oregon State. (They are also now favored against Michigan State as mentioned above which is why the total is +3.) They went from 48.2% against Utah to 53.9%, and they went from 44% against Oregon State to 51.6%. So, in both cases, it is still expected to go either way. For Colorado, they are now favored against Arizona State and Stanford. Colorado went from 26.7% against Arizona State to 56.5%; that is a significant increase! And they went from 44% against Stanford to 52.5%; so still a close game.

The big loser is Arizona State. They had been favored in 3 of their conference games (Colorado, WSU, and Arizona) and now are underdogs in all of their games (all 9 conference games as well as their remaining out-of-conference games). They went from 57.1% versus WSU to 31.2% and from 51.9% to 29.6% version Arizona. So they had significant drops in all three of those conference games. They also saw some drops in all of their other games, like from 30.5% versus California to 10.7%.

The other changes in conference games are California over UCLA (they went from 35.4% to 50.9%) and Oregon over Utah (they went from 40.8% to 50.9%). Both games are expected to be close now.

Because most of the remaining games are against other conference opponents, you don’t see a lot of change in the overall projected wins. The obvious exceptions are Colorado and Arizona State as noted above.

The one other team (besides Arizona State) that saw a drop in projected wins is Oregon State-however it is just a small drop (8.6 to 8.4). Given how well they looked in their opening game, that might seem like a surprise. But, the reason has to do with how well the other Pac-12 teams have looked. As mentioned above, Oregon State is now the underdog to UW in their game. Their win percentage dropped by more than 1 percentage point against WSU, California, Colorado, Stanford, and Oregon. We’ll have to see whether those drops end up being correct.

ESPN Week 2 Projections

As mentioned above, the teams that were favored by ESPN’s FPI before last weekend’s games won 12 of the 13 games featuring Pac-12 teams.

After including the data from the first games, ESPN updated the win percentages for the remaining games. Here are their projections for all this week’s Pac-12 games:

Despite the increase for Colorado, they are still an underdog according to ESPN’s FPI even though they are now being favored on the betting sites that I’ve seen.

Two other teams improved their odds: Oregon and Utah. Both had good performances in their first games while their opponents lost their first games (Texas Tech and Baylor). It should be noted that both of those games are road games for the Pac-12 team which makes the improvement even more impressive.

UW Win Projections

Next we have the win percentages for UW’s remaining games from two different sources. The first is from the Massey Ratings. The next is from ESPN’s FPI. The final one is from Sharp College Football’s Beta Rank. (The ESPN FPI win percentages are rounded.)

The biggest difference is for the USC game. ESPN’s FPI doesn’t give UW much of a chance while Massey indicates that USC is slightly favored and Beta Rank has UW slightly favored.

Conclusions

While some fans may deride rankings, especially this early in the season, they are a part of college football whether we like them or not. As we see with Colorado, rankings that rely a lot on the performances from last year are not always good at predicting performances early in the season. But Colorado is still an outlier in terms of how much change there has been between one year and the next. For the teams that do not have much turnover, using last year’s performances can help project rankings until enough games are played this season.

Specifically for UW, there were several rankings that had UW down slightly in the preseason compared to how they finished last year. But with the win over Boise State, much of that drop has been erased and in many cases UW is now ranked ahead of where they were at any point last year. The highest ranking last year that UW received in any of the rankings that I found was #5 after UW had defeated Stanford (after Michigan State). Currently UW is ranked #5 in three different rankings and is #3 in another, and they are as high as #2 in the Billingsley ranking. Overall, UW is ranked in the top 10 in over half of the rankings in the Composite Ranking. So while there are still some rankings (like FPI) that have UW ranked below where many Husky fans think that UW should be right now, UW is getting some well-deserved attention in many rankings.