HM
HARDMATTEVisual Analytics
FilmsCompareGameJournal
HardMatte

Decoding the art of cinema through advanced data analytics and AI-driven insights.

Product

  • Help
  • Metrics Guide
  • Browse Database
  • Compare Films
  • Cinemetrics Game
  • Pricing

Journal

  • Latest Analysis
  • Visual Patterns

Studio

  • About Us
  • Manifesto
  • Terms of Service
  • Privacy Policy
  • Contact

© 2026 HardMatte Inc. All rights reserved.

  • About HardMatte
  • Metrics Guide
  • Cinemetrics FAQ
  • Our Principles
  • Contact Us
Understanding the Data

Cinemetrics Guide

Every film tells a story through numbers. This guide explains what each metric means, why it matters, and how to use these insights to deepen your understanding of cinema.

Quick Reference: Pacing by Era

Classical Hollywood
>15s
Long takes, theatrical staging
New Hollywood
8-15s
Rhythmic editing, genre awareness
Modern
5-8s
Commercial pacing, MTV influence
Contemporary Fast
2-5s
Action-driven, quick cuts
Hyperkinetic
<2s
Extreme pacing, music videos

Core Shot Metrics

The fundamental measurements of film editing rhythm

Average Shot Length (ASL)

What it is

The mean duration of all shots in the film, measured in seconds.

Why it matters

ASL is the fundamental metric of film pacing. A low ASL (2-4s) indicates rapid cutting typical of action films, while a high ASL (10-20s+) suggests contemplative, slow-burn cinema.

Example

Mad Max: Fury Road has an ASL of ~2s, while 2001: A Space Odyssey has an ASL of ~20s.

Fast: <5sModerate: 5-10sSlow: >10s

Median Shot Length

What it is

The middle value when all shot durations are sorted.

Why it matters

More resistant to outliers than ASL. If a film has a few extremely long shots but mostly quick cuts, the median gives a better sense of the "typical" shot duration.

Shot Count

What it is

Total number of shots (cuts) in the film.

Why it matters

More shots = more visual information and faster pacing. A 2-hour film with 3000 shots feels very different from one with 500 shots.

Cuts Per Minute

What it is

Average number of cuts per minute of runtime.

Why it matters

Normalizes shot count by duration. Useful for comparing films of different lengths. Action films often have 10-15+ cuts/min, while art films may have 2-4 cuts/min.

Cinematography Metrics

Visual characteristics of lighting, color, and camera work

Luminance Timeline

What it is

Average brightness of each frame over time, sampled every 2 seconds.

Why it matters

Reveals lighting patterns. Dark films (noir, horror) have low luminance. Bright comedies have high luminance. Changes show day/night transitions or mood shifts.

Contrast Timeline

What it is

Standard deviation of pixel brightness in each frame (how much variation between light and dark).

Why it matters

High contrast = dramatic lighting with deep shadows and bright highlights (film noir, Blade Runner). Low contrast = flat, even lighting (sitcoms, some digital films).

Color Temperature

What it is

Ratio of warm colors (reds/oranges) to cool colors (blues) over time.

Why it matters

Warm tones feel intimate, nostalgic, or aggressive. Cool tones feel detached, futuristic, or melancholic.

Example

The Matrix is heavily blue-tinted (cool), while Mad Max: Fury Road is orange-tinted (warm).

Camera Movement

What it is

Frame-to-frame difference measuring how much the image changes.

Why it matters

High movement = kinetic energy (action scenes, handheld camera). Low movement = static, composed shots (Wes Anderson, Kubrick).

Color Analysis

Understanding the film's visual palette and color identity

Dominant Color Palette

What it is

8 most common colors in the film, extracted via K-means clustering on sampled frames.

Why it matters

Reveals the film's visual identity. Wes Anderson films have pastel palettes. Blade Runner 2049 is orange and blue. The Godfather is brown and gold.

Color Variety Score

What it is

Standard deviation of hue distribution (0-1).

Why it matters

High variety = colorful, diverse palette (Pixar films). Low variety = monochromatic or limited palette (noir, some period pieces).

Warm/Cool Ratio

What it is

Percentage of warm-toned pixels (reds, oranges, yellows) vs. cool-toned (blues, greens).

Why it matters

Warm films feel energetic, nostalgic, or aggressive. Cool films feel detached, futuristic, or melancholic.

Average Saturation

What it is

Mean color saturation (0 = grayscale, 1 = fully saturated).

Why it matters

High saturation = vibrant, stylized (musicals, fantasy). Low saturation = muted, realistic (dramas, war films).

Intensity Analysis

Measuring the emotional and kinetic energy of scenes

Intensity Timeline

What it is

Combined score (0-1) of motion, visual energy, and cut density at each moment.

Why it matters

Creates a "tension curve" showing the film's emotional/kinetic peaks and valleys. High intensity = action, chaos, climax. Low intensity = calm, contemplation.

Formula(motion × 0.4) + (visual_energy × 0.3) + (cut_density × 0.3)

Peak Moments

What it is

Top 10 highest intensity points in the film.

Why it matters

Identifies the most visually/kinetically intense scenes — likely action sequences, climaxes, or shocking moments.

Climax Timestamp

What it is

The single highest intensity moment in the film.

Why it matters

Often corresponds to the narrative climax. Useful for analyzing story structure.

Average Intensity

What it is

Mean intensity score across the entire film.

Why it matters

High average = relentless pacing (action films). Low average = contemplative (art films).

Narrative Structure

How the film's pacing maps to story structure

Three-Act Breakdown

What it is

The film divided into Setup (0-25%), Confrontation (25-75%), and Resolution (75-100%), with metrics for each act.

Why it matters

Classic screenplay structure. Shows how pacing changes across the story. Act 2 often has faster cutting (rising tension), while Act 3 may slow down or speed up depending on the climax.

Rhythm Variance

What it is

Statistical variance of shot durations.

Why it matters

High variance = unpredictable rhythm (experimental films, thrillers). Low variance = steady, metronomic pacing.

Acceleration/Deceleration Points

What it is

Moments where the cutting rate significantly increases or decreases.

Why it matters

Identifies rhythm changes. Accelerations often mark action sequences or tension build-up. Decelerations mark emotional beats or resolution.

Visual Complexity

Analyzing the density and detail of visual compositions

Edge Density Timeline

What it is

Percentage of pixels that are edges (detected via Canny algorithm) in each frame.

Why it matters

High edge density = visually busy frames (cityscapes, crowds, detailed sets). Low edge density = minimalist compositions (deserts, empty rooms).

Average Complexity

What it is

Mean edge density across the film.

Why it matters

Reveals overall visual style. Blade Runner is high complexity. The Tree of Life is low complexity.

Movie Barcode

What it is

200 vertical stripes, each representing the average color of a moment in the film.

Why it matters

Visual fingerprint. You can "see" the film's color journey at a glance. Useful for comparing films or identifying color patterns.

Example

The Matrix barcode is green-tinted. Mad Max is orange. Moonlight shifts from blue to pink to blue.

How to Use These Metrics

For Film Analysis

  • • Compare ASL across a director's filmography
  • • Use intensity timeline to identify climactic structure
  • • Compare color palettes across genres or eras

For Discovery

  • • Find films with similar pacing to ones you love
  • • Discover outliers (slow action films, fast art films)
  • • Explore visual styles through color palettes

For Education

  • • Teach editing principles with ASL comparisons
  • • Demonstrate color theory with palette analysis
  • • Support arguments with quantitative data

Limitations to Keep in Mind

•Shot scale is estimated — true scale detection requires AI to identify subject distance
•Transitions are estimated — based on brightness changes, not frame-by-frame analysis
•No semantic understanding — metrics don't know what's happening in the story
•Frame sampling — some metrics sample every 2-4 seconds rather than every frame

Shot detection: PySceneDetect (content-aware) • Color clustering: K-means (8 clusters) • Edge detection: Canny (50/150 threshold) • Sampling: 2-4 second intervals

Cinemetrics Analyzer v2.1.0