2026 season · Week 3
Over Expected: who beat expectations in Week 3
Every game’s biggest overperformer and letdown against our pregame projection, live on game days, then the week’s leaders in fantasy points over expected (FPOE): points beyond what a league-average player scores with the same targets, carries and throws. Big FPOE weeks rarely repeat; the usage behind them does.
Updated Tue, Oct 6, 9:37 PM ET
Game by game
- ATL ATL35@GB GB14FinalBeat expectations+16.2
Drake LondonATLWR28.4 pts · proj 12.2 · 98th-percentile game
9 catches for 194 yards on 10 targets
- LAC LAC16@BUF BUF24FinalBeat expectations+10.9
Keaton MitchellLACRB13.7 pts · proj 2.8 · 99th-percentile game
8 carries vs a projected 3
- CAR CAR18@CLE CLE21FinalBeat expectations+16.1
Harold Fannin Jr.CLETE24.1 pts · proj 8.0 · 99th-percentile game
9 targets vs a projected 5
Letdown−12.2
Tetairoa McMillanCARWR3.7 pts · proj 15.9 · 9th-percentile game
5 targets vs a projected 9
- NYJ NYJ24@DET DET31FinalBeat expectations+18.6
Kenyon SadiqNYJTE23.5 pts · proj 4.9 · 99th-percentile game
8 targets vs a projected 3 — Mitchell out
- HOU HOU17@IND IND19FinalBeat expectations+7.5
Keenan AllenINDWR18.3 pts · proj 10.8 · 87th-percentile game
6 for 63, 1 TD on 9 targets
Letdown−8.9
Jonathan TaylorINDRB9.2 pts · proj 18.1
No touchdown on 26 touches (23 carries for 68; 2 for 4 on 3 targets)
- NE NE6@JAX JAX35FinalBeat expectations+8.3
Jakobi MeyersJAXWR19.4 pts · proj 11.1 · 88th-percentile game
7 for 64, 1 TD on 8 targets
Letdown−12.3
Drake MayeNEQB3.8 pts · proj 16.1 · 6th-percentile game
2 interceptions (14/25, 199 yds, 2 INT)
- KC KC24@MIA MIA10FinalBeat expectations+12.4
Ollie Gordon IIMIARB14.5 pts · proj 2.1 · 99th-percentile game
17 carries vs a projected 3 — Achane left early
Letdown−16.0
De'Von AchaneMIARB1.7 pts · proj 17.7 · 4th-percentile game
Played 5% of snaps (projected 73%)
- TEN TEN7@NYG NYG12FinalBeat expectations+11.2
Wan'Dale RobinsonTENWR18.7 pts · proj 7.5 · 99th-percentile game
11 targets vs a projected 5
- CIN CIN27@PIT PIT30FinalBeat expectations+9.1
Aaron RodgersPITQB22.7 pts · proj 13.6 · 91st-percentile game
3 TD passes (19/33, 292 yds, 3 TD, 1 INT)
Letdown−8.7
Chase BrownCINRB8.9 pts · proj 17.6
No touchdown on 15 touches (13 carries for 61; 2 for 8 on 2 targets)
- SEA SEA31@WAS WAS33FinalBeat expectations+14.5
Jaxon Smith-NjigbaSEAWR35.4 pts · proj 20.9 · 91st-percentile game
2 touchdowns on 14 touches
- ARI ARI30@SF SF36FinalBeat expectations+15.0
Jeremiyah LoveARIRB21.9 pts · proj 6.9 · 99th-percentile game
21 carries vs a projected 6 — Conner out
- MIN MIN23@TB TB16FinalBeat expectations+11.1
Jordan AddisonMINWR20.0 pts · proj 8.9 · 96th-percentile game
A 41-yard touchdown (5 for 90, 1 TD on 9 targets)
Letdown−10.3
Justin JeffersonMINWR5.2 pts · proj 15.5 · 12th-percentile game
Played 12% of snaps (projected 92%)
- BAL BAL34@DAL DAL31FinalBeat expectations+9.2
Matthew HibnerBALTE10.3 pts · proj 1.1 · 99th-percentile game
2 for 23, 1 TD on 2 targets
- LV LV35@NO NO27FinalBeat expectations+18.6
Brock BowersLVTE27.6 pts · proj 9.1 · 99th-percentile game
13 targets vs a projected 5
- LAR LAR26@DEN DEN30FinalBeat expectations+16.3
Tyler HigbeeLARTE20.2 pts · proj 3.9 · 99th-percentile game
11 targets vs a projected 2
- PHI PHI7@CHI CHI27FinalBeat expectations+12.9
Kalif RaymondCHIWR21.0 pts · proj 8.1 · 99th-percentile game
A 41-yard touchdown (6 for 90, 1 TD on 7 targets)
Projections for weeks before kickoff snapshots began are the same model run as of that week from earlier games only (no live injury or depth inputs).
Week 3 leaderboard
Fantasy points minus our last projection before kickoff. The top and bottom of the list are shown; sort any column.
| Why | ||||
|---|---|---|---|---|
| 23.5 | 4.9 | +18.6 | 8 targets vs a projected 3 — Mitchell out | |
| 27.6 | 9.1 | +18.6 | 13 targets vs a projected 5 | |
| 20.2 | 3.9 | +16.3 | 11 targets vs a projected 2 | |
| 28.4 | 12.2 | +16.2 | 9 catches for 194 yards on 10 targets | |
| 24.1 | 8.0 | +16.1 | 9 targets vs a projected 5 | |
| 35.3 | 19.5 | +15.8 | 2 touchdowns on 31 touches | |
| 19.3 | 3.9 | +15.4 | 2 touchdowns on 4 touches | |
| 19.3 | 4.0 | +15.3 | 8 targets vs a projected 2 | |
| 41.4 | 26.2 | +15.2 | 3 touchdowns on 28 touches | |
| 21.9 | 6.9 | +15.0 | 21 carries vs a projected 6 — Conner out | |
| 35.4 | 20.9 | +14.5 | 2 touchdowns on 14 touches | |
| 25.9 | 11.6 | +14.3 | 17 targets vs a projected 7 | |
| 23.3 | 9.1 | +14.2 | 2 touchdowns on 8 touches | |
| 21.0 | 8.1 | +12.9 | A 41-yard touchdown (6 for 90, 1 TD on 7 targets) | |
| 26.2 | 13.3 | +12.9 | 2 touchdowns on 7 touches | |
| 14.5 | 2.1 | +12.4 | 17 carries vs a projected 3 — Achane left early | |
| 26.0 | 14.6 | +11.4 | 3 TD passes (31/37, 321 yds, 3 TD) | |
| 26.7 | 15.4 | +11.3 | 10 catches for 107 yards on 13 targets | |
| 31.3 | 20.1 | +11.2 | 4 TD passes (15/27, 297 yds, 4 TD, 34 rush yds) | |
| 18.7 | 7.5 | +11.2 | 11 targets vs a projected 5 | |
| 5.4 | 12.2 | −6.8 | 3 for 24 on 5 targets | |
| 0.0 | 6.8 | −6.8 | 0 targets vs a projected 5 | |
| 0.0 | 7.0 | −7.0 | 0 for 0 on 3 targets | |
| 8.5 | 15.5 | −7.0 | No touchdown on 16 touches (13 carries for 32; 3 for 23 on 3 targets) | |
| 4.0 | 11.1 | −7.1 | 2 targets vs a projected 6 | |
| 6.1 | 13.3 | −7.1 | 14/22, 118 yds | |
| 3.8 | 11.4 | −7.6 | Lost a fumble (2 for 38 on 3 targets) | |
| 7.8 | 15.8 | −8.1 | No touchdown on 19 touches (15 carries for 46; 2 for 12 on 4 targets) | |
| 1.5 | 9.6 | −8.1 | 2 targets vs a projected 5 | |
| 3.8 | 12.4 | −8.6 | Played 36% of snaps (projected 75%) | |
| 11.9 | 20.6 | −8.7 | 4 for 19, 1 TD on 8 targets | |
| 8.9 | 17.6 | −8.7 | No touchdown on 15 touches (13 carries for 61; 2 for 8 on 2 targets) | |
| 9.2 | 18.1 | −8.9 | No touchdown on 26 touches (23 carries for 68; 2 for 4 on 3 targets) | |
| 1.8 | 10.7 | −8.9 | 1 target vs a projected 7 | |
| 1.2 | 10.9 | −9.7 | 5 carries vs a projected 12 | |
| 5.2 | 15.5 | −10.3 | Played 12% of snaps (projected 92%) | |
| 1.6 | 12.3 | −10.7 | 3 targets vs a projected 6 | |
| 3.7 | 15.9 | −12.2 | 5 targets vs a projected 9 | |
| 3.8 | 16.1 | −12.3 | 2 interceptions (14/25, 199 yds, 2 INT) | |
| 1.7 | 17.7 | −16.0 | Played 5% of snaps (projected 73%) |
Team defenses vs projection
Season-long scoring (DraftKings alongside). Points allowed leave out the opponent’s own return touchdowns and safeties, as ESPN and Yahoo score them.
| Defense | Pts | Proj | vs proj | DK | Line |
|---|---|---|---|---|---|
| MINMIN D/ST vs TB | 17.0 | 6.8(2–12) | +10.3 | 17.0 | 6 sk · 2 INT · 0 FR · 1 TD · 16 PA |
| LACLAC D/ST vs BUF | 14.0 | 3.8(-1–9) | +10.2 | 14.0 | 4 sk · 2 INT · 3 FR · 24 PA |
| DENDEN D/ST vs LAR | 14.0 | 5.9(1–11) | +8.1 | 14.0 | 4 sk · 2 INT · 0 FR · 1 TD · 26 PA |
| WASWAS D/ST vs SEA | 11.0 | 4.3(0–9) | +6.7 | 11.0 | 0 sk · 2 INT · 1 FR · 1 TD · 31 PA |
| JAXJAX D/ST vs NE | 13.0 | 6.5(1–12) | +6.5 | 16.0 | 3 sk · 2 INT · 1 FR · 6 PA |
| LVLV D/ST vs NO | 12.0 | 6.5(1–12) | +5.5 | 12.0 | 2 sk · 1 INT · 3 FR · 1 BLK · 27 PA |
| CHICHI D/ST vs PHI | 11.0 | 5.7(1–11) | +5.3 | 12.0 | 2 sk · 2 INT · 1 FR · 7 PA |
| HOUHOU D/ST vs IND | 10.0 | 6.0(1–11) | +4.0 | 11.0 | 4 sk · 1 INT · 2 FR · 19 PA |
| ATLATL D/ST vs GB | 6.0 | 5.0(0–10) | +1.0 | 6.0 | 1 sk · 1 INT · 0 FR · 1 BLK · 14 PA |
| NYGNYG D/ST vs TEN | 7.0 | 6.2(2–11) | +0.8 | 8.0 | 0 sk · 1 INT · 1 FR · 7 PA |
| DETDET D/ST vs NYJ | 7.0 | 6.3(2–11) | +0.7 | 7.0 | 5 sk · 0 INT · 1 FR · 24 PA |
| KCKC D/ST vs MIA | 7.0 | 6.4(2–11) | +0.6 | 8.0 | 0 sk · 1 INT · 1 FR · 10 PA |
| TENTEN D/ST vs NYG | 6.0 | 5.7(1–11) | +0.3 | 7.0 | 3 sk · 0 INT · 0 FR · 12 PA |
| CLECLE D/ST vs CAR | 5.0 | 5.4(1–11) | −0.4 | 6.0 | 3 sk · 1 INT · 0 FR · 18 PA |
| TBTB D/ST vs MIN | 5.0 | 5.7(1–11) | −0.7 | 5.0 | 3 sk · 1 INT · 0 FR · 23 PA |
| BALBAL D/ST vs DAL | 4.0 | 4.8(0–10) | −0.8 | 4.0 | 1 sk · 0 INT · 1 FR · 1 BLK · 31 PA |
| PITPIT D/ST vs CIN | 5.0 | 6.3(1–12) | −1.3 | 5.0 | 1 sk · 0 INT · 2 FR · 27 PA |
| INDIND D/ST vs HOU | 4.0 | 5.3(1–10) | −1.3 | 4.0 | 3 sk · 0 INT · 0 FR · 17 PA |
| BUFBUF D/ST vs LAC | 5.0 | 6.6(2–12) | −1.6 | 5.0 | 2 sk · 1 INT · 0 FR · 16 PA |
| MIAMIA D/ST vs KC | 2.0 | 4.3(-1–9) | −2.3 | 2.0 | 0 sk · 1 INT · 0 FR · 24 PA |
| LARLAR D/ST vs DEN | 3.0 | 5.5(1–11) | −2.5 | 3.0 | 1 sk · 1 INT · 0 FR · 24 PA |
| NYJNYJ D/ST vs DET | 1.0 | 3.9(-1–9) | −2.8 | 1.0 | 2 sk · 0 INT · 0 FR · 31 PA |
| CARCAR D/ST vs CLE | 4.0 | 7.3(2–13) | −3.2 | 4.0 | 2 sk · 0 INT · 1 FR · 21 PA |
| CINCIN D/ST vs PIT | 3.0 | 6.3(2–11) | −3.3 | 3.0 | 2 sk · 1 INT · 0 FR · 30 PA |
| NONO D/ST vs LV | 2.0 | 6.5(2–12) | −4.5 | 1.0 | 3 sk · 0 INT · 1 FR · 35 PA |
| SEASEA D/ST vs WAS | 2.0 | 6.8(2–12) | −4.8 | 2.0 | 2 sk · 0 INT · 0 FR · 27 PA |
| SFSF D/ST vs ARI | 1.0 | 5.8(1–11) | −4.8 | 1.0 | 2 sk · 0 INT · 0 FR · 30 PA |
| NENE D/ST vs JAX | 0.0 | 4.9(0–10) | −4.9 | -1.0 | 1 sk · 1 INT · 0 FR · 35 PA |
| DALDAL D/ST vs BAL | -1.0 | 4.0(-1–9) | −5.0 | -1.0 | 0 sk · 0 INT · 0 FR · 34 PA |
| ARIARI D/ST vs SF | -1.0 | 4.1(-1–9) | −5.1 | -2.0 | 0 sk · 0 INT · 1 FR · 36 PA |
| PHIPHI D/ST vs CHI | 0.0 | 5.6(1–11) | −5.6 | 0.0 | 0 sk · 0 INT · 0 FR · 27 PA |
| GBGB D/ST vs ATL | -1.0 | 6.6(2–11) | −7.6 | -2.0 | 0 sk · 1 INT · 0 FR · 35 PA |
2026 season: over and under expected
Through Week 4, games with play-by-play only.
Scoring the most above what their usage usually produces (3+ games). History says about a quarter of that edge carries forward: sell-high candidates.
| 3 | 22.4 | 12.2 | +10.2 | Sell high | |
| 4 | 29.2 | 19.6 | +9.6 | Sell high | |
| 4 | 25.1 | 15.9 | +9.2 | Sell high | |
| 4 | 26.4 | 19.2 | +7.2 | Sell high | |
| 4 | 28.1 | 21.2 | +6.9 | Sell high | |
| 4 | 28.1 | 21.4 | +6.6 | Sell high | |
| 4 | 14.5 | 8.6 | +5.8 | Sell high | |
| 4 | 19.8 | 14.0 | +5.7 | Sell high | |
| 4 | 16.1 | 10.4 | +5.7 | Sell high | |
| 4 | 20.9 | 15.2 | +5.7 | Sell high | |
| 4 | 19.8 | 14.4 | +5.3 | Sell high | |
| 4 | 27.5 | 22.6 | +4.9 | Sell high | |
| 4 | 12.1 | 7.3 | +4.8 | Sell high | |
| 4 | 13.2 | 8.6 | +4.5 | Sell high | |
| 3 | 15.0 | 10.4 | +4.5 | Sell high | |
| 4 | 22.4 | 18.0 | +4.4 | Sell high | |
| 3 | 14.8 | 10.6 | +4.3 | Sell high | |
| 4 | 14.3 | 10.2 | +4.2 | Sell high | |
| 4 | 29.0 | 24.8 | +4.2 | Sell high | |
| 3 | 8.1 | 4.0 | +4.1 | Sell high | |
| 4 | 11.2 | 7.2 | +4.1 | Sell high | |
| 4 | 22.7 | 18.8 | +4.0 | Sell high | |
| 4 | 19.3 | 15.3 | +4.0 | Sell high | |
| 3 | 12.7 | 8.9 | +3.7 | Sell high | |
| 3 | 17.4 | 13.8 | +3.6 | Sell high |
Next: Ascension (rising usage) · Waiver wire · Projections
How this works
vs projection compares his fantasy points with the last projection we published before his game kicked off (mean, with the 10th–90th percentile range). While a game is on, the projection is scaled by the share of the game played, by quarter and clock.
Expected points (xFP) come from our own play-level model fitted on 2023–25 play-by-play: each target is valued by its air yards, field position and down, each carry by field position, down and whether it was a scramble, each pass attempt by air yards and field position. On 2024–25 games it was not fitted on, xFP averaged within 1–6% of actual points at every position and correlated 0.98 with ffopportunity’s public model.
FPOE = points − xFP. Across 2023–25 it barely carried from one week to the next (r = 0.05) while expected points did (r = 0.34): treat a big FPOE as regression risk, and big usage with a negative FPOE as a buy-low.
Sources: nflverse play-by-play and weekly stats, ESPN box scores (live)Updated Tue, Oct 6, 9:37 PM ET
Frequently asked questions
What does “over expected” mean here?
Two things. Versus projection: his fantasy points minus our pregame projection (the last one made before his game kicked off). Versus opportunity, or FPOE: his points minus what a league-average player scores with the same targets, carries and pass attempts from the same spots on the field (expected fantasy points, xFP).
Should I chase a big FPOE week?
Usually not. Across 2023–25 a player’s FPOE one week had almost no relationship with the next (correlation 0.05), and only about a quarter of a half-season’s FPOE carried into the second half. A big positive FPOE is a sell-high signal; a big negative FPOE with heavy usage is a buy-low. Expected points from usage hold up far better (week-to-week correlation 0.34).
What is the “pace” number during games?
While a game is on, his projection is scaled by how much of the game has been played (by quarter and clock), so a receiver with 12 points at halftime is compared with half of his projection. It is a rough guide: usage is not spread evenly across a game. Final numbers replace it after the game.
When do the final numbers land?
Live numbers come from box scores every 15 minutes during games. Expected points need play-by-play, which nflverse publishes the morning after; FPOE and the final leaderboard update then (Monday night games on Tuesday).