I played Troll's Dilemma with a small group of six players, using red and blue chips for the binary choice instead of heads/tails.
We played two games, five rounds each, with the highest individual scorer being the winner. Some players employed the "honestly, always choose the same color" strategy while others randomized blindly, while still others went blindly random, and others kept a strategy of switching periodically.
With this small of a group, it's hard to say whether any of our results will scale up. It seemed to some that random choice is only slightly worse than an actual conscious strategy. Scores were within five points of each other at the end of each game, but there was always a clear winner, even if they only won by 1 point.
Some tweaks to consider:
- Formalize the discussion phase so it's a simple statement: "I choose red/blue." I think I'd go so far as to give each player a badge or card that they can visibly flip to indicate their public choice.
- Use star stickers so everyone knows how many points you have. Possibly color-code the stickers so you can tell which points were earned by deceit.
- Randomize game length so you're not quite sure when the last round will be. Somewhere between 5 and 10 rounds might be the sweet spot.
All in all, it was a fun experience and a game I think I'll explore further with a bigger group. I appreciate any public playtesting you can offer!