How to host a dubbing party
The failure mode of a voice game night is not technical. It is four people enjoying themselves and two people quietly deciding they are not going to have a turn.
A dubbing game asks people to be deliberately ridiculous out loud in front of their friends. For roughly half of any group that is the appeal, and for the other half it is the reason they are hoping their turn does not come. Almost everything about hosting one well is about that second group.
The shape of an evening
- Four to six people. Two works and is a different, quieter thing. Above about seven the wait between turns gets long enough that people start looking at their phones, and a phone in this game is very hard to come back from.
- Ninety minutes, not four hours. This is a high-energy format and it does not sustain. Stopping while everybody still wants another round is how you get asked to do it again.
- Three or four scenes, run several times each. Not twelve scenes once each. The second and third performance of the same scene is where the evening actually is, because everybody now knows the lines and is listening for what this person does differently.
The first twenty minutes decide it
Whoever goes first sets the ceiling. If the first performance is tentative, everybody calibrates to tentative and the evening never recovers. So the host goes first, and goes too far — visibly, unmistakably too far. It gives everybody else permission and it costs you nothing, because you already know how the game works.
Then hand the second turn to whoever is most obviously enjoying themselves, not to whoever is next in the circle. You are building a floor before you ask anybody nervous to stand on it.
Choosing scenes that work
- Two people arguing. An argument hands every performer something to push against; a monologue asks them to invent a shape from nothing, which is a much harder job.
- Sixty to ninety seconds. Long enough to build something, short enough that a room happily watches four people each do it.
- Faces you can see. Performers are watching mouths and expressions to know where to land. A scene shot from behind, in the dark, or in constant motion gives them nothing.
- Silences left in. Do not pick something edited to wall-to-wall dialogue — you have removed the room's best material before anybody opened their mouth.
Getting the quiet ones performing
- Offer pairs. Two people splitting a two-hander is dramatically less exposing than one person doing both voices, and it is often funnier.
- Assign the voice instead of asking for one. 'You are a weather presenter who has had enough' removes the hardest decision and the one most people freeze on.
- Let somebody be the reader. A person who genuinely does not want to perform can call the scenes, run the playback and keep score, and that is a real job rather than a consolation.
- Do not comment on anybody's voice afterwards, even kindly. Praise pulls attention to the fact that performances are being evaluated, which is exactly what the reluctant half is worried about.
If people are joining from elsewhere
Everything above still applies and one thing changes: on a call you lose the room. Nobody can see anybody else react, and reaction is half of what makes this work — so performances land into silence and performers assume they have failed.
Two fixes. Ask everybody to keep cameras on, which sounds like a small thing and is not. And use something built for people being apart rather than screen sharing a desktop game, because the host of a screen share ends up driving menus all evening instead of playing.
PartyDub is the browser version of this idea: everybody joins a room with a code, dubs the same clip without hearing each other, and then it is played back together and the room votes. It is free, there is nothing to install, and it is made by the same person who runs this site — said plainly here for the same reason it is said everywhere else on it.