Branching narratives
Definition
Narratives that present alternative story developments that are generally incompatible with each other. They affect major story and characters elements.
Classic structures in choice-based games
From Sam Kabo's Ashwell all-time great:
| Structure | Representation | Notes |
|---|---|---|
| Time cave |  | No variables, qualities, or stats |
| Gauntlet |  | Can be friendly or deadly Low use of variables |
| Branch and bottleneck |  | High use of variables Often used to represent the passage of time and the world's and the player's growth Works better in big pieces |
| Quest |  | High use of variables Works better in big pieces Centered on exploration and environment Narrative is fragmentary |
| Open map |  | Centered on exploration and environment High use of variables Narrative is either fragmentary or loose-paced |
| Sorting hat | High usage of variables at start, and then not at all Lots of work: custom systems + branch-exclusive content | |
| Floating modules | Akin to storylets, same requirements of wide content and explicit (?) stats Grinding stats can be incorporated into the game's design | |
| Loop and grow |  | Notion of runs High use of variables Justified by the narrative Often backed by the theme of escaping some kind of imprisonment Can be complified by using main branches |
Small-scale patterns in choice-based games
| Structure | Representation | Function | Further opportunities |
|---|---|---|---|
| Confirmation-required choice |  | Offer several occasions to drop out of a dangerous / stupid decision | Measure the player's proneness to risk |
| Track switching choice |  | Emulates indecision | Measure commitment, proneness to changing their mind |
| Scored choice | Aggregation of same-importance choices | Decompose a big decision into many smaller ones to make a tendency emerge | |
| Re-enterable conversation node |  | Liven-up an exposition dump | Track what they seem to be most spontaneously interested in Gate options behind variables |
| Chapter one sorting hat |  | Offers early game replayability | / |
| Endgame time cave |  | Pay-off setup Seems most adapted to disjointed consequences, for global consequences use narrative states | / |
Embranchments don't matter
Choices don't have to be impactful, they just have to feel that way. What matters most is having a compelling story.
Players rarely replay their games, many branchs are not seen despite their number or quality.
Characteristics of a good main embranchment
- Produces changes in scenes or character's behaviors
- Allows unique ways of reaching one's goals
- Marks a turning point that may be remembered and reffered to later
- Doesn't produce disjointed narratives, only affects minor secondary elements (that can still be displayed in small scenes)
How to build an embranchment
- Can stem from the order of scenes in the determined narrative, or the order the player learns key info in
- Can stem from major conflicts ands character's goals
- Good practice: first embranchment asap, then on a regular rythm (25/30min according to Episode Interactive's own Cass Phillipps)
ex: Branching Ulysse's return to Ithaca
❗Not to be read as a geographical map, each segment can only be played once; to make the most of your content, you'd probably want your player to play every single segment once before getting to Ithaca.
💡 Scoring choices at the end can be efficiently used in conjunction with narrative states