The Save the Cat Beat Sheet, Explained With Data

The Save the Cat beat sheet is fifteen story beats that recur in well-structured films, and in successful 2015+ titles they cluster around predictable runtime positions. Here is the structure and what the data shows.

Blake Snyder's Save the Cat gave screenwriters a shared vocabulary for structure: fifteen beats that recur, in order, in well-built films. It is not a formula that guarantees a good script — plenty of great films bend it — but as a diagnostic it is hard to beat, because the major turns it describes tend to land at consistent positions in films that work.

Below is the structure, with what the runtime data tends to show. Percentages are positions in the runtime, drawn from the macro pattern across successful 2015+ titles; treat them as a reference range, not a rule.

The 15 beats

1. Opening Image (~0%) — the tone and the "before" state in one shot.
2. Theme Stated (~5%) — someone names what the film is really about.
3. Set-Up (0–10%) — the hero's world and what's missing.
4. Catalyst (~10%) — the inciting event.
5. Debate (10–20%) — should they? The hesitation.
6. Break into Two (~20%) — the hero commits; the story proper begins.
7. B Story (~22%) — the relationship that carries the theme.
8. Fun and Games (20–50%) — the "promise of the premise"; the trailer moments.
9. Midpoint (~50%) — a false win or false defeat that raises the stakes.
10. Bad Guys Close In (50–75%) — pressure mounts, internally and externally.
11. All Is Lost (~75%) — the low point.
12. Dark Night of the Soul (75–80%) — the hero processes the loss.
13. Break into Three (~80%) — the solution, fusing A and B stories.
14. Finale (80–99%) — the hero acts on what they learned.
15. Final Image (~100%) — the "after" that mirrors the opening.

What the data shows

The striking thing is the consistency of the big turns. Across well-structured successful films, the commitment to the story (Break into Two), the stakes-raising Midpoint, and the low point (All Is Lost) cluster tightly around the 20%, 50%, and 75% marks. That regularity is exactly why the beat sheet works as a diagnostic: if your Break into Two arrives at 40% of the runtime, your first act is probably overlong, and a reader will feel it before they can name it.

The reverse is also true: rigidly hitting every beat at its "ideal" percentage does not make a script good. The beats are a map of how attention and tension are usually managed, not a guarantee of either.

How real films hit (and bend) the beats

You can read the structure live in recent successes. Parasite (Cannes 2019, Best Picture 2020) lands its Break into Two almost exactly where the model predicts — the Kim family's infiltration commits the story near the 20% mark — then detonates a true Save the Cat Midpoint reversal at roughly the halfway point that flips the film's genre under you. Get Out (2017) front-loads its Catalyst the way horror does, but still snaps its All Is Lost low point into place around the three-quarter mark. Everything Everywhere All at Once (2022) bends the model hardest, fracturing the Fun and Games into a multiverse — and yet its emotional act breaks still fall where the beat sheet says the audience needs them.

The pattern is consistent: the more a film succeeds with a wide audience, the more its major turns honor the act-break positions, even when the texture is wildly unconventional. The beats describe attention management, and attention does not care about genre.

How do recent award films treat the beats?

Structure is a baseline, not a cage. Everything Everywhere All at Once hits recognizable beats - catalyst, midpoint, dark night of the soul - yet won seven Oscars including Best Picture in 2023 on the strength of how it bent them. Anatomy of a Fall, the 2023 Palme d'Or winner with five Oscar nominations including Best Picture, largely ignores the commercial beat map. The data shows beats correlate with broad appeal; they do not guarantee it.

How to use it

• As a diagnostic, not a generator. Write the story, then check where your turns actually fall. Outliers are worth a second look — not automatic problems.
• Watch the act breaks first. The 20% / 50% / 75% turns carry the most structural weight; get those right and the smaller beats tend to follow.
• Let the genre flex it. Horror front-loads its catalyst; slow-burn drama earns a later Break into Two. The model bends by genre.

What does the beat data tell you, and what does it not?

Mapping a script against the Save the Cat beats is useful because it surfaces structural problems early — a midpoint that lands late, a second act with no real turn, a finale that arrives unearned. Across successful films of a given type the major beats do tend to fall in a recognisable range, and a script that drifts far outside it usually has a pacing issue worth examining. But the beat sheet is a diagnostic, not a guarantee. Hitting every mark on time produces correct structure, not a good film; the beats are scaffolding for the things that actually move an audience — voice, character, and a premise worth ninety minutes. Use the data to find where your story sags, then solve it with craft rather than by forcing a page count. Structure that serves the story is the goal; structure for its own sake is just a tidy way to be boring.

In summary

Structure is not the same as story, but a script that fights its structure usually fights its reader. The beat sheet, read against real 2015+ data, will not write your film — but it will tell you where to look when a draft feels off, and it will tell you with the same accuracy a reader's instinct does, only earlier. Use it as the map of where attention is usually won, then earn the turns in your own voice.