Belle of the Ball in Development

Image: Belle of the Ball - Game Prototype
It's a busy time for my work schedule, but I still scrounged up some spare hours to playtest a totally new iteration of Belle of the Ball this weekend. After numerous playtests this weekend (it's a fast game), I think it's getting close to public beta. Initially, the game played as described here. Since that first draft, I went through a bunch of different changes.

Trimmed down the number of Belles to 12. Enough to come out pretty often, but they don't dominate the game as much. I took some of the extra Belle's effects and pasted them onto the normal guests. Now those end-round effects only occur if you possess that guest in your clique.

Each player has their own Belle. Instead of one Belle in the center of the table, each player may invite a Belle to their own clique. The Belle's powers only apply to that player. Furthermore, inviting a Belle allows that player to attract a couple, which is quite useful since couples are otherwise safe.

Revised Couples rules a few times. Now, only the top card's attributes are in play, including its point value. Couples are a pretty safe investment, but can be lured away by the Belles.

Revised win condition a bit. The round ends when a certain number of couples are in play -or- when the deck runs out. The deck running out is also the trigger for endgame. Keep track of scores each round and discard all the cliques and hands in play. This starts each round with a clean slate. The player with the highest score tallied across all rounds wins.

Added negative-point guests. This gave players a lot more reason to invite repellant guests as an offensive strategy. I also made sure that only the double-attractive guests were worth negative points. So, you must weigh the cost of that guest against the value of the guests he or she would attract.

Added Events. One-time changes to the game like "Spoiled Food" causing all the eating guests to leave the party. There were tons of new events in one playtest, which actually caused the deck to be a little bloated. Also, playing an event card was often too costly a use of your turn when you could instead invite a guest. So, I trimmed out any redundant events, cut out their effects and pasted them onto guests. Now, when you invite that guest, it causes that effect.

Players may invite guests to any player's clique. So, you could invite a repellant guest to another player's clique, as another way to get new guests. This, again, is an effort to make repellant guests a bit more useful.

Overall, the challenge has been fighting my urge for symmetry across the whole deck. When there is perfect balance between all possible permutations, it creates a generally blasé game experience. You want enough eccentricity to force players to choose their own individual strategies, but enough predictability so an early decision still has some bearing on the late-game outcome. There's more to say on that subject, but for now I've got work to do!


  1. mmm yes, public beta!
    I'm definately keen to see a post elaborating (or even demonstrating?) on your final paragraph, about finding not-quite-perfect balance/symmetry in a game

    As someone also developing a card game, I am also very interested in the changes you list above, but am also curious as to how you recognised the need for those changes, and how you worked out the best solution to them?

    Obviously these things may or may not translate from your specific ruleset, but I fear I'm doing mine mostly on intuition, which, while probably not *bad*, does always leave me with the fear that I'm not noticing problems, or choosing sub-par solutions...

  2. Overall, (I've been following your posts on this game idea) I like this game a lot. It is a very original theme and game. Looking forward to following your work on it.

  3. Well, I have 96 basic guest cards. This is a useful number because it is divisible into thirds, quarters and sixths. I divided each of those guests with three independent axes of information:

    Family: Each of those guests is divided into six families, with 16 guests in each family.

    Activity 1: 32 of the guests are flirting, 32 are snubbing and a 32 are doing neither.

    Activity 2: 24 are eating, 24 are drinking, 24 are dancing, and 24 are doing neither.

    I had these perfect distributions of information. I knew I need exactly these number of icons across the entire deck, but they didn't need to correlate with each other at all. So, I distributed them somewhat randomly. I literally laid out a map of 96 cards, then randomly scattered the necessary icons across the entire deck.

    Does that mean one family has more drinkers? If so, that's okay. Does that mean some dancers are rarely flirting? That's okay, too. The division is equal, the distribution is organic.

    When deciding who would attract/repel which guests, I followed the same distribution as Activity 1 and 2, with two rules for myself: 1) A guest cannot repel itself, so I made sure no guest was doing an activity that it repelled. 2) Roughly half the guests are attractors, while the other half are repulsors.

    Lastly, because it's most economical to print card games in 108 card decks, I had 12 cards left over. Those are the Belles. How lucky! I have 6 families. So, each family gets two Belles that have special effects on them alone.

    I followed this same pattern across the entire design process. Divide up the deck in a logical fashion, but distribute those differences organically.

  4. Awesome. Especially the comment response to Cameron. I love/hate those sort of distribution decisions and the work I've had to do on decks in my own games.

  5. Thanks! Glad to hear it. :D

  6. Right? Oh man, it's such a pain. I want these perfectly intricate mandalas, but those are no fun.


Post a Comment


Show more