The college football regular season has reached its halfway point, and not a ton has changed in the AP Top 25. So, will there be surprises to come? This week’s midseason edition of the column assesses where the rankings have been, where they are now and where they might be going.

1. I find myself with a revolving door of No. 1s. I switched from Georgia to Texas before the Longhorns promptly lost to Oklahoma. And now, after putting Georgia back on top, I made another change this week: Washington has earned my vote.

I considered it a few weeks ago, when I pushed the Huskies all the way up to No. 2, but they lacked a true resume-making win. Now they have it against Oregon, a result better than anything Georgia has done thus far, on top of having arguably the best offense in the country behind Michael Penix Jr. and a deep receiving corps.

Meanwhile, Georgia sleepwalked through much of the afternoon against Vanderbilt, and its lackluster performance against Auburn doesn’t look any better after the Tigers were pummeled by LSU. Throw in the fact that dominating Kentucky also looks less impressive after the Wildcats were soundly beaten by Missouri, and I don’t believe Georgia needs to be No. 1 by default.

I thought a few more people might join me in voting Washington No. 1, but just one other voter did it. Most of the Georgia defectors have gone to Michigan, which has 16 votes despite playing a schedule even more forgettable than Georgia’s to this point.

Not that I’m against thinking that Michigan is the best team. But we’ll get to that later.

2. Georgia was ranked No. 1 at midseason in each of the past two years en route to its national championships. Fourteen of the past 16 national champions were ranked in the top five in mid-October, the only exceptions being 2014 Ohio State (No. 13) and 2015 Alabama (No. 8). In fact, of the 38 national champions since 1989 (including shared titles), 32 were ranked in the top five of the AP poll at this point in October, including 11 at No. 1.

Since the poll expanded to 25 teams in 1989, the only national champions that were ranked outside the top 10 at midseason beyond 2014 Ohio State were both in 1990: split champs Colorado (No. 14) and Georgia Tech (No. 11). In other words, history says the national champion is almost certainly in the top 10 right now and is likely to be one of Georgia, Michigan, Ohio State, Florida State and Washington.

3. The word “parity” has been thrown around plenty when analyzing this season, including in this space. But parity can mean many things. In the case of this season, it’s the feeling that there are more national championship contenders than usual, or at least more than there have been for most of the College Football Playoff era. The top tier of teams feels bunched up, and nobody is running away from the pack.

But it’s also been a chalky season. Through seven weeks, there has been only one win by an unranked team against an AP top-10 team at the time of the game: Duke over Clemson on Labor Day. Substantial upsets have been few and far between, and we haven’t had anything resembling the semi-annual chaos Saturday we all know and love.

In each of the past two regular seasons, there were 12 wins by unranked teams against top-10 opponents. That included five in the first half of 2022 and seven in the first half of 2021. The gold standard for chaos was, of course, 2007, which featured 21 wins by unranked teams against top-10 opponents in the regular season, including four on Sept. 29 alone. On the other end of the spectrum: 1997, which featured just three such upsets.

This season has had a handful of memorable games — including Washington-Oregon and Oklahoma-Texas rivalry classics the past two weeks — but truly shocking results have been lacking.

4. Voters were historically terrible at predicting how the season would play out last year, when just 10 teams both started and finished the season ranked in the AP poll — a record low since 1989 poll expansion. That number was 13 at midseason.

Perhaps things are evening out. This midseason? Twenty of 25 teams that were ranked in August are ranked today after Tulane and Iowa re-joined the poll. The only preseason ranked teams not ranked in this week’s poll are Clemson, Kansas State, TCU, Wisconsin and Texas A&M. Clemson could easily find itself back in the poll next week, too, if it beats Miami.

On average, 15.5 preseason ranked teams have finished ranked in the Top 25 era, with a record of 20 doing so in 1996.

5. In each of the past three seasons, three preseason top-10 teams finished unranked. The average since 1989 is 1.7, though zero did so as recently as 2019. LSU and Clemson are the only two-loss teams that began the season in the top 10, though both do have multiple chances for losses down the stretch. I’d bet on at least one finishing unranked.

An average of 2.1 preseason unranked teams have finished in the top 10 since 1989. Five did so last year, with four in the top 10 at midseason. At midseason this year? The ranked teams that started outside the poll are No. 16 Duke, No. 20 Missouri, No. 21 Louisville, No. 22 Air Force and No. 25 UCLA. Some noteworthy breakthroughs, but not exactly a list of upstarts ready to challenge for a Playoff spot.

6. The season has also lacked marquee games, with Florida State-LSU, Ohio State-Notre Dame and Oregon-Washington the only top-10 matchups at the time of the game. Based on current rankings, there are three top-10 matchups the rest of the regular season, all in the Big Ten East: Penn State at Ohio State on Saturday, Michigan at Penn State on Nov. 11 and Ohio State at Michigan on Nov. 25.

No. 3 Ohio State hosts No. 7 Penn State on Saturday. (Matthew O’Haren / USA Today)

7. So when will unpredictability arrive? Keep an eye on Saturday, Nov. 4, a day that features plenty of potential traps, from Penn State at Maryland to Florida State at Pitt to Oklahoma at Oklahoma State to Kansas State at Texas to Georgia vs. Missouri, among others — on top of LSU at Alabama, Notre Dame at Clemson and Washington at USC.

The next week, Nov. 11, also brings plenty of intrigue, headlined by Michigan at Penn State, Ole Miss at Georgia, Miami at Florida State, Utah at Washington, USC at Oregon and what could be the first ranked matchup between Duke and North Carolina since 1939.

The first seven Saturdays have lacked depth, but be patient. November has a lot of upside for both high-profile games and chaos.

8. As always, plenty of history can be made in the final AP poll. A few milestones to watch:

• No. 16 Duke has achieved its highest AP ranking since reaching No. 16 in 1994. It hasn’t been in the top 15 since 1971 or finished in the top 15 since 1960.
• No. 10 North Carolina hasn’t finished in the top 10 since the final two seasons of Mack Brown’s first stint in 1996 and 1997.
• No. 12 Oregon State has its highest ranking since 2012. It hasn’t finished in the top 15 since 2000, when it was No. 4.
• James Madison is one spot outside the poll and trying to finish ranked in just its second season as an FBS team.
• No. 11 Alabama is trying to avoid finishing outside the top 10 since it was unranked in Nick Saban’s first season in 2007.
• Clemson is trying to avoid finishing unranked for the first time since 2010.
• No. 2 Michigan is trying to reach No. 1 for the first time since it won the 1997 national title. Nobody but Alabama, Georgia, Clemson and LSU has reached No. 1 in the past eight seasons.
• No. 24 Iowa is ranked and averaging 20.9 points per game. Since 1995, the only teams to score fewer than 21 points per game and finish ranked were 2015 Northwestern (19.5) and 2004 Wisconsin (20.8).

9. Who will make the Playoff? I made plenty of terrible bold guesses in the preseason, but the general rule I wrote about, based on history, says: “A decent bet for a CFP prediction is two teams in the preseason top four, another team from the next five and one wild card that starts 10th or worse.”

In August, I said preseason No. 1 Georgia, No. 2 Michigan, No. 6 USC and No. 11 Texas would make the CFP. I’ll stick with Georgia and Michigan. Part of me wants to stick with Texas too — it can run the table and beat Oklahoma in a rematch to get in — but I’ll go with preseason No. 8 Florida State and No. 10 Washington to round out the field.

10. Who will win the national championship? Michigan hasn’t beaten anyone of note yet, but it has the all-around talent to beat both Ohio State and Penn State and make a run in the Playoff. It has an experienced quarterback in J.J. McCarthy, a proven running game, one of the nation’s best offensive lines, another great defense … and on and on.

I’ll stick with my preseason pick of the Wolverines to win it all, even if I have Washington No. 1 now. They’re dominant at the line of scrimmage and the most complete team in college football. They’ll soon finally have opportunities to try to prove it.

(Top photo of Blake Corum: Gregory Shamus / Getty Images)