Boardgame Design

Notes from a panel by Bill Fawcett at LibertyCon 2014

Number one question is to ask, "why will someone want to play the game I am designing?"
One day of market research will save you six months of anguish. Go to a hobby shop and ask the people there what they want to play. And recruit them as playtesters.

Create a game which is accessable. You need to be able to pick up the basics on ten minutes, and get all the minutiate by playing it a few times.
"Simple level planning" so you can plan your next move while someone else is acting.

Learn the value of constant positive reinforcement. Give a goal which is achievable within minutes, and opens another goal. You may want to have several goal tracks operating at once--for instance, in Steampunk Paladins, you might complete a quest which gives you one more step towards Airship Cruiser, then the next quest might be a step toward increasing your Fame, or maxing out your Valor, or creating Professor Gould's Patent Photonic Death Ray.

The game should reward your players often. Rewards are things like the ability to take an extra or special type of move; make an extra attack;  find valuable information; taunt another player (but not screw him too badly); or gather items. Contrariwise, penalizing players for success will kill your game.

The ideal game does not knock players out of the game early. They should be able to affect the game, even if they're behind. Example: in Carnage con Queso, even if your squad was demolished or you hadn't collected any cheese, you still had cards that you could use to harass the leader (or whoever else you felt deserved a random mortar barrage). A player who is behind should be able to surge, although a strong player should still be able to win. Anything harassing plays should set back the leader, not knock them out.

Start with an existing game that you like, and fix it.