Critic Consensus vs. Quilty Score
How our greenlight model relates to films critics loved — the 50 top-consensus features of 2016–2026.
Critic averages tell us how a finished film was received. The Quilty Score asks a different question: would the underlying material have been worth greenlighting in the first place? This page puts both lenses side-by-side so the disagreements are useful, not embarrassing.
A note on sources
IMDb, Rotten Tomatoes, Metacritic, and the consensus average shown on this page were transcribed from a public "Best Movies of 2016–2026" graphic. They are unverified against the source APIs; treat them as directional. Quilty scores come from our production database and are calculated by the V5 ensemble.
8 / 50
films currently scored by Quilty
74
across the scored subset
Nope
+2.9 vs. consensus
| # | Film | IMDb | RT | MC | Avg | Quilty | Δ vs. Avg | |
|---|---|---|---|---|---|---|---|---|
| 1 | Parasite (2019) Drama | 8.5 | 99% | 97 | 94.8 | 76.1 | -18.7 | |
| 2 | Moonlight (2016) Drama | 7.4 | 98% | 99 | 91.5 | 76.8 | -14.7 | |
| 3 | Portrait of a Lady on Fire (2019) Romance | 8.1 | 97% | 95 | 91.4 | Not yet scored | — | |
| 4 | The Father (2020) Drama | 8.2 | 98% | 88 | 91.3 | Not yet scored | — | |
| 5 | Mad Max: Fury Road (2015) Action | 8.1 | 97% | 90 | 90.3 | Not yet scored | — | |
| 6 | Spider-Man: Across the Spider-Verse (2023) Animation | 8.6 | 95% | 86 | 89.9 | Not yet scored | — | |
| 7 | Aftersun (2022) Drama | 7.6 | 96% | 95 | 89.5 | Not yet scored | — | |
| 8 | Oppenheimer (2023) Biopic | 8.3 | 93% | 90 | 89.4 | Not yet scored | — | |
| 9 | The Zone of Interest (2023) Drama | 7.5 | 93% | 98 | 89.3 | Not yet scored | — | |
| 10 | Past Lives (2023) Romance | 7.8 | 95% | 94 | 89.3 | 80.4 | -8.9 | |
| 11 | Minari (2020) Drama | 7.4 | 98% | 89 | 88.1 | Not yet scored | — | |
| 12 | The Irishman (2019) Crime | 7.8 | 95% | 94 | 88.0 | Not yet scored | — | |
| 13 | Godzilla Minus One (2023) Action | 8.3 | 98% | 81 | 87.8 | Not yet scored | — | |
| 14 | La La Land (2016) Musical | 8.0 | 91% | 94 | 87.7 | 70.8 | -16.9 | |
| 15 | The Worst Person in the World (2021) Drama | 7.7 | 96% | 90 | 87.6 | Not yet scored | — | |
| 16 | Anatomy of a Fall (2023) Drama | 7.7 | 96% | 90 | 87.6 | Not yet scored | — | |
| 17 | Drive My Car (2021) Drama | 7.5 | 97% | 91 | 87.5 | Not yet scored | — | |
| 18 | Nomadland (2020) Drama | 7.3 | 93% | 93 | 87.1 | Not yet scored | — | |
| 19 | The Boy and the Heron (2023) Animation | 7.4 | 93% | 91 | 87.1 | Not yet scored | — | |
| 20 | The Banshees of Inisherin (2022) Drama | 7.7 | 96% | 87 | 86.9 | Not yet scored | — | |
| 21 | Soul (2020) Animation | 8.0 | 95% | 83 | 86.0 | Not yet scored | — | |
| 22 | TÁR (2022) Drama | 7.4 | 91% | 92 | 86.0 | Not yet scored | — | |
| 23 | Killers of the Flower Moon (2023) Drama | 7.6 | 93% | 89 | 86.0 | Not yet scored | — | |
| 24 | Poor Things (2023) Comedy | 7.8 | 92% | 88 | 85.9 | Not yet scored | — | |
| 25 | Dune: Part Two (2024) Sci-Fi | 8.5 | 92% | 79 | 85.8 | Not yet scored | — | |
| 26 | Sound of Metal (2020) Drama | 7.7 | 97% | 82 | 85.6 | Not yet scored | — | |
| 27 | Top Gun: Maverick (2022) Action | 8.2 | 96% | 78 | 85.4 | Not yet scored | — | |
| 28 | The Holdovers (2023) Comedy | 7.9 | 97% | 82 | 85.3 | 78.1 | -7.2 | |
| 29 | A Hero (2021) Drama | 7.5 | 97% | 82 | 85.2 | Not yet scored | — | |
| 30 | Decision to Leave (2022) Thriller | 7.3 | 94% | 88 | 84.4 | Not yet scored | — | |
| 31 | Everything Everywhere All at Once (2022) Sci-Fi | 7.8 | 94% | 81 | 84.3 | 67.5 | -16.8 | |
| 32 | Perfect Days (2023) Drama | 7.9 | 96% | 89 | 84.3 | Not yet scored | — | |
| 33 | Mission: Impossible – Dead Reckoning (2023) Action | 7.7 | 96% | 81 | 84.2 | Not yet scored | — | |
| 34 | Palm Springs (2020) Comedy | 7.4 | 94% | 83 | 84.1 | Not yet scored | — | |
| 35 | Licorice Pizza (2021) Comedy | 7.1 | 91% | 90 | 84.0 | Not yet scored | — | |
| 36 | The Mitchells vs. the Machines (2021) Animation | 7.6 | 97% | 81 | 83.5 | Not yet scored | — | |
| 37 | Another Round (2020) Drama | 7.7 | 93% | 81 | 83.2 | Not yet scored | — | |
| 38 | CODA (2021) Drama | 8.0 | 94% | 75 | 82.3 | Not yet scored | — | |
| 39 | John Wick: Chapter 4 (2023) Action | 7.7 | 94% | 78 | 82.1 | Not yet scored | — | |
| 40 | Suzume (2022) Animation | 7.6 | 96% | 77 | 81.9 | Not yet scored | — | |
| 41 | The Fabelmans (2022) Drama | 7.5 | 92% | 84 | 81.2 | Not yet scored | — | |
| 42 | Furiosa: A Mad Max Saga (2024) Action | 7.8 | 90% | 79 | 80.9 | Not yet scored | — | |
| 43 | All Quiet on the Western Front (2022) War | 7.8 | 90% | 76 | 80.6 | Not yet scored | — | |
| 44 | Challengers (2024) Drama | 7.4 | 88% | 82 | 79.1 | Not yet scored | — | |
| 45 | Society of the Snow (2023) Drama | 7.8 | 90% | 72 | 79.0 | Not yet scored | — | |
| 46 | The Northman (2022) Action | 7.0 | 90% | 82 | 79.0 | 63.8 | -15.2 | |
| 47 | Blade Runner 2049 (2017) Sci-Fi | 8.0 | 88% | 81 | 79.0 | Not yet scored | — | |
| 48 | The Batman (2022) Action | 7.8 | 85% | 72 | 76.9 | Not yet scored | — | |
| 49 | Civil War (2024) Action | 7.1 | 81% | 75 | 76.4 | Not yet scored | — | |
| 50 | Nope (2022) Sci-Fi | 6.8 | 83% | 77 | 75.6 | 78.5 | +2.9 |
Parasite (2019)
Moonlight (2016)
Portrait of a Lady on Fire (2019)
The Father (2020)
Mad Max: Fury Road (2015)
Spider-Man: Across the Spider-Verse (2023)
Aftersun (2022)
Oppenheimer (2023)
The Zone of Interest (2023)
Past Lives (2023)
Minari (2020)
The Irishman (2019)
Godzilla Minus One (2023)
La La Land (2016)
The Worst Person in the World (2021)
Anatomy of a Fall (2023)
Drive My Car (2021)
Nomadland (2020)
The Boy and the Heron (2023)
The Banshees of Inisherin (2022)
Soul (2020)
TÁR (2022)
Killers of the Flower Moon (2023)
Poor Things (2023)
Dune: Part Two (2024)
Sound of Metal (2020)
Top Gun: Maverick (2022)
The Holdovers (2023)
A Hero (2021)
Decision to Leave (2022)
Everything Everywhere All at Once (2022)
Perfect Days (2023)
Mission: Impossible – Dead Reckoning (2023)
Palm Springs (2020)
Licorice Pizza (2021)
The Mitchells vs. the Machines (2021)
Another Round (2020)
CODA (2021)
John Wick: Chapter 4 (2023)
Suzume (2022)
The Fabelmans (2022)
Furiosa: A Mad Max Saga (2024)
All Quiet on the Western Front (2022)
Challengers (2024)
Society of the Snow (2023)
The Northman (2022)
Blade Runner 2049 (2017)
The Batman (2022)
Civil War (2024)
Nope (2022)
What the deltas tell us
Where Quilty and the critic consensus disagree, the breakdown almost always points at a single pillar. That is where the model is doing useful work — and also where it is most testable.
Quilty rated higher than consensus
- +2.9
Nope (2022)
Quilty 78.5 · Consensus 75.6
- -7.2
The Holdovers (2023)
Quilty 78.1 · Consensus 85.3
- -8.9
Past Lives (2023)
Quilty 80.4 · Consensus 89.3
Quilty rated lower than consensus
- -18.7
Parasite (2019)
Quilty 76.1 · Consensus 94.8
- -16.9
La La Land (2016)
Quilty 70.8 · Consensus 87.7
- -16.8
Everything Everywhere All at Once (2022)
Quilty 67.5 · Consensus 84.3
How to read a gap
A negative delta does not mean Quilty disagreed with critics on quality. Story & Craft scores are usually high on this list — the gap typically comes from Commercial Viability or Production Reality, which the critic score does not measure. Expand any row to see which pillar moved.
How we use this dataset internally
This page is also a calibration asset. The scored subset gives the product team a fixed cohort to track when prompts, models, or weights change.
Prompt regression
Re-score the cohort after a prompt change. Material movement on these high-consensus films is a flag worth reviewing.
Weight tuning
Compare pillar deltas before and after weight changes. The cohort is small enough to inspect by hand and large enough to be useful.
Red-team divergences
Films where consensus and Quilty disagree the most are the cases we use to pressure-test the model and the caveat copy.
Yearly refresh
New top-consensus films get added each year. The script regenerates the data file from production scores so the page stays current.
42 films still need a Quilty score
We will run a controlled benchmark-scoring pass using the V5 orchestrator in historical mode (per release year) and update this page once the cohort is complete. No critic data is fed into the model — this dataset is for evaluation, not training.
Last regenerated: 2026-05-23. Source consensus values:
user-supplied graphic (unverified). Quilty scores: production quilty_scores_v5.