Case File 090 · Narrative and Game Design

How to Write a Believable Suspect

A writer's guide to suspects who feel human, stay consistent and keep a mystery fair.

By Murder Book Editorial TeamPublished 16 September 2026Updated 16 September 2026

A believable suspect is a character with a plausible reason to be suspected, a life that exists outside the crime, and reasons to behave awkwardly that are not necessarily guilt. In practice, that means giving every suspect a motive, an opportunity and a secret, then deciding what each one wants from the conversation with the detective. Innocent suspects should have something to hide, so suspicion falls on them for understandable reasons. The guilty suspect should appear early and behave consistently with the eventual solution. And nervous behaviour should never be written as simple proof: in real life, behavioural cues to lying are weak, and honest people can look uneasy when they badly want to be believed.

Key takeaways

Give every suspect a motive, an opportunity, a secret and a want in the scene.
Innocent suspects are more convincing when they are hiding something other than the crime.
The culprit should be introduced early and play a real part in the story.
Avoid "tells" that prove guilt; research shows behavioural cues to deception are weak.
In games, write each suspect so they could plausibly be innocent or guilty until the evidence decides.

What makes a suspect believable?

Readers and players believe in a suspect when three things line up. The suspicion makes sense: there is a reason to look at this person. The person makes sense: they have history, relationships and concerns that would exist even without a murder. And their behaviour makes sense: when they evade, lie or lose their temper, the reason fits who they are.

Weak suspects usually fail on the second point. They exist only to be suspected, so they have nothing to talk about except the crime, and they tend to act either obviously guilty or obviously innocent.

Which layers should every suspect have?

A simple worksheet helps keep a cast consistent:

Layer	Question to answer	Why it helps
Motive	Why might this person want the crime to happen?	Gives the detective a reason to suspect them
Opportunity	Could they have done it, and what is their account of that time?	Creates something concrete to test
Secret	What are they hiding that is not necessarily the crime?	Explains evasive behaviour in innocent suspects
Want	What do they want from this conversation?	Gives each scene direction beyond answering questions
Limit	What would make them tell the truth or stop talking?	Shapes how the scene can turn
Knowledge	What do they actually know, and what do they only believe?	Keeps their statements consistent across scenes

Writing the "knowledge" layer as a timeline is especially useful. If you know exactly what each suspect saw and when, their statements will stay consistent, and any inconsistency you add will be deliberate.

Why should innocent suspects have something to hide?

If only the guilty person lies, the mystery is solved the moment someone lies. Giving innocent suspects their own secrets, such as an affair, a debt, a small theft or simply embarrassment, means that evasiveness stops being a reliable signal. The detective, and the reader, must work out what someone is hiding, not just whether they are hiding something.

These secrets also make the cast feel human. People in real stories are rarely neutral witnesses. They protect themselves and others, shade the truth to look better, and leave out details they think are irrelevant.

How should the culprit be written?

The guilty suspect carries the solution, so fairness matters most here. The Golden Age fair-play rules still give good guidance. Ronald Knox's "Ten Commandments" of detective fiction, as reproduced in the Open University's OpenLearn course, say the criminal "must be mentioned in the early part of the story". S. S. Van Dine's 1928 rules add that the culprit should be a person who has played a more or less prominent part in the story.

Beyond those rules, a few practical habits help:

Write the culprit's true timeline first, then write the version they tell. The gaps between the two are your clues.
Give the culprit a secondary, honest motive for some behaviour, so not everything they do points to guilt.
Let them be likeable or sympathetic in places. A reveal lands harder when the reader did not want it to be them.
Keep them consistent with the solution. On a reread, their lines should make more sense, not less.

These choices feed directly into the ending; see What Makes a Satisfying Mystery Solution?.

How should nervous behaviour be written?

This is where fiction often drifts from reality. Shifty eyes, fidgeting and sweating are familiar shortcuts for guilt, but research does not support them as reliable signs. A meta-analysis by Bond and DePaulo (2006), covering 206 documents and 24,483 judges, found that people correctly classified lies and truths only 54% of the time on average. It also noted that people appear deceptive when they are motivated to be believed. A separate review by DePaulo and colleagues (2003) reported that many behaviours showed no discernible or only weak links to deceit.

For writers, that research is useful rather than limiting. An innocent suspect who is desperate to be believed can come across as suspicious for completely human reasons. A guilty suspect can be calm. Writing behaviour this way keeps the mystery honest and pushes the solution toward evidence and contradictions, where it belongs. Why There Is No Single Reliable Sign of Deception covers the research in more depth.

What changes when a game has suspects?

Games add two pressures that prose does not.

Suspects must survive repeated questioning. A player can ask the same character about many topics in different orders. Every answer has to fit that character's knowledge and secret, which is why a written timeline for each suspect matters even more in games.

Guilt may not be fixed. The Shapeshifting Detective (D'Avekki Studios, Wales Interactive, 2018) chooses its murderer at random at the start of each game, according to its Steam page. In a design like that, every suspect has to be written so they could be guilty or innocent, depending on the evidence in that playthrough.

Some games even put the player on the other side. In Overboard! (inkle, 2021) you play the murderer, and Steam notes that other characters remember everything they see and hear. That design shows how believable suspects depend on the world around them: a character only feels real if their story has to hold up against what others noticed. For how suspicion is built across a whole cast, see How Detective Games Create Suspicion.

How does Murder Book approach suspects?

Murder Book: The Art of Interrogation is an upcoming narrative interrogation game from Gungrounds, an independent studio in Zagreb, Croatia. Its cases centre on interrogations, and its press materials describe gameplay built around behaviours such as Fabrication, Avoidance, Narrative Gaps and Emotional Leakage. Development screenshots show an in-game casebook explaining behaviours like Stress Response and Acting Emotion, and a case select screen that includes a "not guilty" result.

That combination reflects the principle above: behaviour is a reason to look more closely at a story, not proof that someone did it. A suspect can be stressed, evasive or inconsistent and still be innocent. Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027. See How Murder Book Turns Interrogation Into Gameplay.