We’re making a creative social deduction game called Ghostwriters where players are given themes and take turns adding to a story with their themes, but two players (the ghostwriters) have the same theme, and everybody is trying to guess which two players those are (or who their partner is).
Does the social deduction aspect add to the experience?
We discussed the possibility of simply leaving out the guessing part and focusing more on rewarding good storytelling through a variety of mechanics. It’s important to answer this question because social deduction is one of our core mechanics, and we need to decide early whether to commit to it or pivot off of it – if we meander around it we may end up creating an unfocused or disjointed game. To identify what contribution, if any, social deduction mechanics make to our game we could make a role prototype that has all roles known and incorporates an alternative scoring mechanic, like having players extend the story in rotating pairs then having everybody else vote on which of the pair extended it better. I would guess that people will prefer the social deduction mechanics, as they seem more interesting to me.
Should we focus more on guessing players or guessing themes?
One of our drafts focused more on the concept of players avoiding having their themes guessed by other players, rather than trying to avoid having their identity guessed. This is another core mechanic that decides the direction and feeling of the game, so it’s important to figure out early. To answer it, we can make another role prototype that uses the theme guessing mechanic for scoring instead of forcing players to identify the ghostwriters. I would guess that guessing players might work better, as players of social deduction games will likely find it to be a more familiar mechanic – though I personally prefer the theme guessing concept.
Is dealing from a deck a good way to assign roles?
We’ve been assuming that themes will be dealt out from a deck, with special care taken beforehand to ensure a duplicate exists, but this might be somewhat inconvenient to do. We can check this with an implementation prototype which uses a deck of themes, to see how it feels to play. I would guess that it’s adequate, and most players probably won’t mind the inconvenience of finding a duplicate to deal out.
