Cal Football, 2000 to Now
Rise, Peak and Drift

For a few seasons in the mid-2000s, Cal was one of the best teams in college football. This is how it got there, why it slipped, and what has changed underneath the program since.

Six chapters, one question each, built from 27 seasons of game results, recruiting rankings, NFL draft outcomes, attendance and transfer records.

Two seasons in the top ten, then a long slide

Jeff Tedford inherited a 1-10 team in 2002. Two years later Cal went 10-2 and ranked 2nd of 119 FBS teams in schedule-adjusted margin, ahead of every program in the country but one.

The peak lasted about five seasons. Cal has not finished in the top 20 since 2008, and the low point came in the handoff between coaches: Tedford's last team went 3-9, and Sonny Dykes' first went 1-11, 97th of 125.

Cal got more out of its talent, then lost both

Talent is the best single predictor of results in college football. Across 2,549 FBS team-seasons, rosters built from stronger recruiting classes reliably finish higher. That splits every season into two questions: how much talent did Cal have, and what did it do with it?

How much talent Cal had

In 2005 Cal's roster ranked 12th in the country. It hasn't had a top-20 roster since 2008, and this season (2026) it ranks 67th.

What Cal did with it

The next chart compares where Cal finished with where a roster like Cal's usually finishes. From 2005 to 2008, Cal beat its prediction by 10 spots a season on average. In 2006, a roster expected to finish 29th finished 6th. Since 2009, Cal has finished ahead of its prediction in only 5 of 16 seasons.

So the decline has two causes, and they compound: less talent coming in, and less coming out of the talent Cal had.

The peak produced pros who outplayed their draft slot

One part of the peak can be measured long after the games: what Cal's players did in the NFL. For every draft pick since 1980, I modeled the career a player taken at that slot typically has, then measured how far each school's players beat it.

For the 2003 to 2013 draft classes, the players Tedford recruited and coached, Cal ranks first of 41 schools. Its 44 picks beat their draft slot by about one extra season as a solid NFL starter each, and the edge came from early-round picks, not late-round luck.

Cal's biggest outperformers (career value vs. a typical pick at the same slot)

The crowd came for the peak and never came back

From 2004 to 2008, Cal averaged 62,697 fans a game. The crowd followed the team down after that, but it didn't follow it back up: the 8-5 teams of 2015 and 2019 drew 48,800 and 42,433, and the 7-6 team of 2025 drew 34,991, about 56% of the renovated stadium's capacity.

The exception came on October 5, 2024, when ESPN's College GameDay came to Berkeley and 52,428 people watched Cal lose 39-38 to Miami. The demand is there for a big moment. It hasn't been there week to week.

The transfer portal made the roster a two-way door

Since 2021, players can transfer without sitting out a season, and since NIL rules changed the same year, they can be paid. In 2021, 10 players left Cal through the portal and 3 arrived. In 2026, 36 left and 32 arrived.

The players Cal develops now have options. Of the 140 who left since 2021, 53 went to other Power 4 programs, including the highest-rated departures below. The most famous is Fernando Mendoza, who left for Indiana in December 2024, won the Heisman Trophy and went first overall in the 2026 NFL draft.

Highest-rated players to leave Cal through the portal

The talent is back to mid-pack. The results aren't.

The portal cuts both ways, and Cal has used it to rebuild. Counting transfers, its 2026 roster ranks 8th of 17 in the ACC and 32nd of 67 Power 4 teams, close to the middle of both.

The results haven't followed. In its first two ACC seasons, Cal finished 10th and 13th of 17 conference teams in schedule-adjusted margin.

1. Talent sets the ceiling. Cal's best seasons came with top-20 rosters. It hasn't had one since 2008.
2. Coaching decides how much of that talent shows up on Saturdays. With the portal reshuffling rosters every year, there is little time for slow, multi-year development. What matters is a staff that gets the most out of whoever is on the roster this season and turns them into a coherent team. Cal's peak teams finished an average of 10 spots better than their talent predicted; since 2009, Cal has beaten its prediction in only 5 of 16 seasons.
3. The portal is now the fastest lever. A roster can be rebuilt in a single offseason, in either direction. Cal's is already mid-pack in its conference, so the next gain has to come from coaching that roster into a team that plays above it.

Data: College Football Data API (collegefootballdata.com), with attendance gaps filled from published box scores, and NFL draft data from nflverse. Team strength is the Simple Rating System (scoring margin adjusted for schedule). Roster talent averages the 247Sports recruiting rankings of the four classes on the roster; chapter 6 uses the 247Sports talent composite, which also counts transfers.