How to Write Action Scenes: Decisions Under Pressure, Not Just Spectacle

Quick definition

Good action scenes run on decisions under pressure. Pressure forces a choice. The choice changes the situation. The change creates new pressure. That loop—not the number of punches, bullets, or explosions—is the spine of an action scene.

In one line: Pressure → choice → changed situation → new pressure.

Spectacle gives you moments. Decisions give you a scene.

Scribblers naturally chase images: the big stunt, the impossible shot, the spectacular hit.

Those moments may be memorable, but by themselves they are fragments.

The underlying engine is harder to fake:

1. The character has an objective.
2. The easy solution is blocked.
3. Several options exist, each with a cost.
4. The character chooses.
5. That choice changes the state of the scene.
6. The new state creates the next problem.

This is the WANT Framework under physical pressure.

Read: The WANT Framework for Screenwriting →

Goals need teeth

“Win the fight” is not much of a goal. It is a placeholder.

A strong action objective can fail in interesting ways:

• Get the witness out before the building burns.
• Hold the position for exactly two minutes.
• Stay unseen while your cover is about to be blown.
• Capture the target without pulling the trigger.

Then add a second pressure:

• Protect someone who does not want protecting.
• Preserve a cover story while using skills that would expose it.
• Refuse to cross a line the character has never crossed.

Now the objective pulls in two directions at once. That tension drives behavior.

The source essay points to the final mission in Top Gun: Maverick. The objective is not merely “blow up the plant.” The physical constraints of speed, terrain, SAMs, G-forces, and a tiny target give the goal teeth. Every second creates decisions.

Kill the easy exit

If the direct solution works, there is no scene.

Remove the clean shot. Jam the safe exit. Make retreat cost something. Force the character to choose.

A useful action pattern is:

problem → solution → new problem

The character cannot cross open ground without getting shot. They improvise cover. That cover gets them farther but leaves them pinned. One solution creates the next problem.

That is momentum.

Character decides the move

Put two different characters in the same problem and you should get two different scenes.

One charges. One studies. One protects. One cuts loose.

The question underneath every beat is:

Why does this particular person make this particular choice?

The source essay uses the fanny-pack fight in Everything Everywhere All at Once. The scene works because the solution is specific to Waymond and to the logic surrounding him. Put a different hero in the same room and the entire scene changes.

That is character writing, not stunt description.

This connects directly to Wants Invent Opinions: the character’s objective shapes how they read the problem, and that opinion shapes the move they choose.

Read: Why Character Wants Create Opinions and Conflict →

Space is pressure

Location is not decoration. It should push back.

A narrow corridor forces commitment. Open space removes cover. A crowd scrambles the plan. Darkness reduces certainty.

Then let the environment evolve:

• lights fail
• fire spreads
• exits jam
• traffic moves
• footing disappears

The source essay points to the Arc de Triomphe sequence in John Wick: Chapter 4. The traffic circle is not background. Cars become moving cover, weapons, hazards, and timing problems. The space keeps forcing recalculation.

The environment should begin applying pressure before the character even acts.

Build in turns, not movements

Listing movements does not create momentum. Tracking change does.

Do not outline:

He punches the guard.

Outline:

The guard loses control of the doorway.

Do not outline:

She runs to the door.

Outline:

The exit becomes unusable.

The first version records choreography. The second records state change.

Each beat should alter something real: position, advantage, time, information, options, trust, injury, exposure.

If the situation is the same after the beat as it was before, the scene may be moving without progressing.

Use a hard pivot

A strong action scene does not merely escalate. It can become a different kind of problem.

• Stealth becomes chase.
• Escape becomes rescue.
• Survival becomes sacrifice.

The source essay points to the falling train carriages in Mission: Impossible — Dead Reckoning Part One. A horizontal fight becomes a vertical survival problem once the bridge is gone. The objective and the physical logic reset without stopping the momentum.

A pivot forces new decisions rather than simply louder versions of the old decisions.

Give action an interior beat

Action needs an interior beat: not a speech and not necessarily a pause.

Something goes wrong. The character absorbs the cost. Then they commit to the next move.

That beat connects spectacle to experience.

This is essentially a compressed Sequel inside an action run: reaction, understanding, resolve.

Read: Scene vs. Sequel vs. Sequence →

Control what the audience knows

Clarity carries tension.

The audience should understand the essentials:

• Where is the exit?
• What is at risk?
• What does failure mean?

Beyond that, uncertainty can remain. Enemy numbers may be unclear. Backup may be uncertain. The character may not know whether the plan can work.

Confusion is not suspense. Controlled uncertainty can be.

Real consequences

When the action scene ends, something should be different.

Time lost. Trust broken. Position compromised. A line crossed. A relationship altered. Something should become true that was not true before—or stop being true.

The gunfire and impact are delivery mechanisms. The state change is the story.

The action-scene diagnostic

Run the beats through this rubric:

1. What is the real objective?
2. Why can’t it be done the easy way?
3. What options are available?
4. Which option gets chosen, and why this character?
5. What changes as a result?
6. What new problem appears?
7. How does the space affect the choices?
8. What does this reveal about the character?
9. What is different when the scene ends?

Then ask the hardest question:

If you removed all the punches and gunfire, would the decision structure still be interesting?

If yes, the scene has a spine.

How action scales into larger structure

An action scene is still a Scene: objective, obstacle, conflict, choice, consequence. Several action and reaction beats can accumulate into a Sequence, and those sequences can serve the larger Reels of the Modular Story Method.

Read: The Modular Story Method →

Try it on your own scene

Scribbler’s Scene Doctor

Use the Scene Doctor to identify the objective, opposing force, decision structure, escalation, and state change underneath an action scene.

Analyze a scene →

See it in a real page

Not a stunt beat — a decision with its own logic. Turning his back to take the arrows, then releasing the whole flock as cover for the one bird that matters, is the choice that ends the scene, not a punctuation mark after it.

Two enemies at once is stated as explicitly unwinnable head-on — "kill the easy exit" as a literal design constraint, not just a screenwriting metaphor. The direct fight is taken off the table, so a specific tactical decision has to fill the gap.

Source note

This maintained reference page develops ideas first published in “Action Scenes Run on Character Choices” in Scribbler’s Toolbox on April 15, 2026. The original essay remains part of the editorial archive; this version is structured as a durable action-writing reference.

Related pages in the WANT knowledge cluster

• The WANT Framework for Screenwriting: How to Give Every Character a Playable Scene Objective — Cluster hub: defines the core WANT diagnostic used by the other four pages.
• How to Write Visually Without Directing the Camera — render consequential action through visible behavior, selective detail, and clear geography.
• Why Character Wants Create Opinions, Conflict, and Better Scenes — Character extension: explains how WANT generates opinion, behavior, and performance.
• Scene vs. Sequel vs. Sequence: How Screenplays Build Momentum — Structural rhythm: explains how scene objectives and consequences accumulate into larger dramatic units.
• What Is the Modular Story Method? A Four-Reel Approach to Screenplay Structure — Macro architecture: scales the cluster’s scene-level mechanics into four connected story arcs.

About the author

Jesse Alexander is a screenwriter, showrunner, and narrative designer whose work spans film, television, games, and transmedia storytelling. Scribbler’s Toolbox collects practical frameworks developed from professional story development and production.