Our 2025 College Football Playoff predictions model is live. Here are 5 key takeaways

Can Ohio State defend its national championship or will Arch Manning and Texas walk into Columbus and put the college football world on notice in Week 1? Will James Franklin and Penn State follow the blueprint of their Big Ten rivals and give the conference its third national championship in a row behind a plethora of experience? These are among the most pressing questions as we head into the 2025 college football season, ones that my College Football Playoff predictor can help answer.

Just like last season, our College Football Playoff predictor, which launched Thursday and will be updated throughout the 2025 season, uses my College Football Projection Model to simulate the season 100,000 times to find out how likely each team is to win the national championship, win its conference championship and make the College Football Playoff, in addition to how many games it can be expected to win. My ratings are based on an offensive and defensive projection for every college football team using various metrics, such as Expected Points Added and Success Rate, and I also created an algorithm to predict which 12 teams the Playoff committee will choose based on more than a decade of committee decisions. (A full explanation of the methodology can be found at the bottom of our predictor.)

Using these projections, let’s take a look at some surprising outputs as well as some storylines about why the model’s outputs are what they are.

Ohio State favored to go back-to-back

I’m not going to sit here and say Ohio State doesn’t have question marks heading into the season. The Buckeyes replace both coordinators — offensive coordinator Chip Kelly joined the Las Vegas Raiders, and defensive coordinator Jim Knowles left for Big Ten rival Penn State — in addition to 14 NFL Draft picks. The turnover includes a new quarterback (Julian Sayin) and a new defensive line. But although Ohio State will have a drop-off at the coordinator positions, this is still one of the most talented teams in college football, and the model sends the Buckeyes to the Playoff 80.33 percent of the time (just barely ahead of Clemson by 0.04 percentage points) with a 14.3 percent chance to win the national title (edging Penn State).

Wide receiver Jeremiah Smith and safety Caleb Downs might be the best offensive and defensive players in the country, respectively, and players of that caliber can make your life easier when it comes to overcoming inexperience. Having a player like Smith leading a deep group of receivers and tight ends will help Sayin adjust.

All in all, Ohio State has had the highest floor in college football over the past few decades, so reloading shouldn’t be much of a problem. Factor in that its two toughest games — Texas and Penn State — will be played in Columbus, and Ohio State is well-positioned to return to the Playoff. After all, the Buckeyes have won at least 10 games in the regular season in 17 of their past 18 full seasons (excluding 2020).

Our bracket features the teams with the best chance to make the Playoff and to win it all, and Ohio State remains atop in both categories.

Is Clemson back?

Clemson hasn’t felt like a true national title contender in several years, one could argue in the NIL era (post-2021). Sure, the Tigers won the ACC last season and got into the Playoff, but they went 0-3 against the SEC and were never seen as a true championship threat.

I think this year is different.

Clemson has the second-highest odds to make the Playoff at 80.29 percent, per my projections, and is sixth in my team ratings. Cade Klubnik ranks No. 1 in The Athletic’s preseason QB Tiers and is firmly in the Heisman race to kick off the season, and our NFL Draft expert Dane Brugler has four Tigers defensive players in his preseason top 50 prospects. That seems like the build of a team that is ready to get back to contending on the national stage.

The Tigers will benefit from a weaker ACC, as they’re projected to win the conference and secure an automatic bid just over 50 percent of the time — easily the highest Power 4 conference champion odds — but they also have a chance to make an early statement against LSU from the SEC in Week 1.

Could the SEC really not get a bye?

The top five conference champions still get automatic bids, but the bracket is moving to a straight seeding format in which byes aren’t reserved for league winners. SEC fans may notice that in my model’s projected bracket that no team from the SEC is one of the top four teams in the country, which means no first-round bye in the Playoff. Some additional context here is needed.

The SEC has four teams in the projected Playoff, all seeded No. 5 through No. 9, as well as two teams on the bubble. It’s ranked as the toughest conference in my model. Why doesn’t it have a bye here? The issue is that the conference is too good, and the selection process really values records. The gripes from the SEC about strength of schedule not being valued enough are fair. If there is any indication through the committee’s rankings in November that they’re starting to value strength of schedule more — one day this will happen, maybe as soon as this season based on announced changes — then I will tweak my Playoff algorithm to reflect that.

But for now, the SEC’s depth, with nine teams boasting at least a 10 percent chance of a Playoff bid, makes it a greater challenge for any one team to get a bye, especially in the model’s preseason projections.

College football is top-heavy

There are eight teams in my projections that have a greater than 64 percent chance of making the Playoff. The next team is Boise State at 50 percent, followed by a drop to 39 percent with LSU. Those 10 teams are followed by 20 teams with anywhere from a 10 to 38 percent chance to make the Playoff. So while I’m not entirely convinced that Michigan (34 percent) is truly the first team out on the bubble, just below last-team-in Miami, it has a favorable enough schedule (no Penn State, no Oregon, plus Ohio State at home) to give it a chance.

I do believe that the transfer portal and NIL/revenue sharing have given us more parity in college football, but that might show up more in how the 12-team bracket is filled out — see Indiana making the Playoff last year — rather than who wins the championship. I agree with my model that 85 percent of the time the national champion is coming from the top seven teams.

Boise State owns the Group of 5

Just about every Group of 5 conversation starts with Boise State. The Broncos have been the most recognizable brand outside of the power conferences for two decades now. Last year, they played their way to the No. 3 seed in the College Football Playoff, but losing Heisman runner-up running back Ashton Jeanty is a major blow. So can Boise State hold onto the top spot and get another automatic bid as one of the five highest-ranked conference champions?

My model believes the Broncos will do it, as they have a 50 percent chance to make the Playoff heading into the season. The next highest-ranked Group of 5 team is Tulane at 12 percent, and I’m not sure I want to back the Green Wave after needing to go into the portal for quarterbacks post-spring. In fact, my model projects Boise State to be about four points better than Tulane.

The fact of the matter is that the Group of 5 has become weaker since the transfer portal era began. And yet Boise State has bucked that trend (just look at the fact that it held onto Jeanty last season). Though an Oct. 4 road trip to Notre Dame will be difficult, Boise State is projected to be a double-digit favorite in every Mountain West game this season before joining the revamped Pac-12 in 2026. Maybe the Broncos wouldn’t be as fortunate in The American, as my model projects that to be the best G5 conference, but Boise State remains the standard outside the Power 4.

Keep checking the College Football Playoff predictor page for regular updates throughout the season, all the way through the national championship game.

(Photo of Jeremiah Smith: Harry How / Getty Images)