Duke Blue Devils

Duke is No. 39 of 138 in the model after week 6 (4-1, 1-1 in conference play). Its projected final record is 7.8-4.2, with a 1% chance to win out and a 2% chance to make the playoff.

Updated · See Duke in the full rankings

2026 at a glance

  • #39in the model (▼6 from last week)
  • 7.8-4.2projected final record
  • +6.1points vs. an average FBS team, neutral field
  • 2%chance to make the playoff

Scouting report: Duke's best unit is its run defense (No. 24 in FBS); its weakest is its pass defense (No. 91). Overall, the offense ranks No. 83 and the defense No. 60, adjusted for opponents. These ranks rest on 5 games, so expect them to move week to week.

Résumé: No. 42, by wins above what a typical top-25 team would expect against Duke's schedule. That measures what the record is worth; the power rank (No. 39) measures how good the team is.

Deserved wins: playing the way it has, Duke would usually have won 4.5 games; it has won 4.

Postseason odds: 2% to win the ACC, 2% to make the 12-team playoff, 0% to earn a first-round bye and 0% to win the national title. 96% to reach 6 wins. (Model estimates; how the playoff odds work.)

Rank by week: Wk 1: 52 → Wk 2: 45 → Wk 3: 28 → Wk 4: 31 → Wk 5: 33 → Wk 6: 39
Week 1: No. 52Week 2: No. 45Week 3: No. 28Week 4: No. 31Week 5: No. 33Week 6: No. 39

Rankings for weeks 1-6 were rebuilt after launch from games through each week; later weeks are published live.

Season stats

  • 34points per game
  • 14.4points allowed per game
  • 44%offensive success rate
  • 39%success rate allowed

Success rate: share of plays that keep an offense on schedule (half the yards needed on first down, 70% on second, all of it on third or fourth). Garbage time excluded.

Leaders

  • Passing: Walker Eget, 980 yards, 7 TD, 1 INT
  • Rushing: Nate Sheppard, 743 yards on 118 carries, 8 TD
  • Receiving: Jeremiah Hasley, 19 catches, 234 yards, 0 TD

Schedule and win odds

WeekOpponentResult or win chancePlayed like
1vs TulaneW 17-394% win
2at IllinoisW 31-2798% win
3vs StanfordW 35-799% win
4vs William & MaryW 62-7>99% win
6at Georgia TechL 25-2856% win
7vs North Carolina59% to win
8at Virginia55% to win
9vs Boston College86% to win
10at NC State68% to win
11at No. 4 Miami13% to win
12vs Clemson63% to win
13at No. 22 Wake Forest33% to win

Win chances come from the model's current ratings, including home field. Played like: how often a team wins a game it played the way it played this one, based on the box score. Rankings shown are the model's.

ACC in the model

#TeamRecordConf.Projected
4Miami5-03-010.6-1.4
22Wake Forest5-13-19.5-2.5
32Louisville3-32-26.7-5.3
36Pittsburgh5-12-18.5-3.5
43North Carolina3-21-16.7-5.3
44Florida State3-31-26.1-5.9
50SMU4-12-17.7-4.3
52Clemson3-22-17.0-5.0
53Virginia Tech4-21-27.1-4.9
59Virginia4-22-16.6-5.4
72Syracuse2-30-34.7-7.3
79California3-32-15.4-6.6
84Georgia Tech2-31-14.1-7.9
91NC State3-31-25.4-6.6
110Boston College2-30-23.3-8.7
112Stanford2-41-34.2-7.8

This week's picks and rankings

Every week's top 25, game of the week and picks are in the weekly breakdown. Latest: Week 7 college football picks and power rankings: the model's case for LSU.

How the model works: each team starts from a preseason rating (last season, returning production, roster talent), and every game moves it using the final margin and play-by-play efficiency, adjusted for opponents. Projections simulate the rest of the season 10,000 times. More on the method. Data from CollegeFootballData.com.