Thursday, 17 October 2013

Being your own Guinea Pig

One exceedingly common problem for board game designers is a lack of playtesters. If you are lucky, you have friends who are willing (or even enthusiastic) to play your games on a semi-regular basis, or some sort of game design circle that will play your game in exchange for you playing theirs.

Usually though, such arrangements are weekly at best, and having a week iteration time makes for slow progress. [In this context an iteration is one cycle of Playtest > Rework > Playtest].

It is often beneficial to play against yourself in order to test tweaks and variations - small or large - in order to get a feel for the effectiveness of your changes. Self testing should be the first line of defence when evaluating your game.

I can't even count the number of times I've had a idea for a game that sounds great on paper, but as soon as I went to try it, I lost all enthusiasm for it. This has happened for cards, turn structures, restrictions, rules, items - basically anything that it could happen for. Putting it to the simple self-test forces you to change your mindset, from designer to player - and ultimately what the player thinks is far more important.

It can also help you figure out what is too complicated or does not work within the confines of your game. It is far too easy to forget the restrictions of a board game and overcomplicate things. In a video game, you can add on complexity, and the computer will handle all of your variables in the background - in a board game, the players must do everything for themselves, and this can result in a drastically slower experience if you add on too many things.

It isn't a perfect solution, and there are several things that complicate playing against yourself:

  • Limited information is no longer limited. You know exactly what is in your opponent's hand, because you have seen it. To counter this, figure out the rational decision based on the information that each player should have; alternatively, make the decision at random if that is reasonable.
  • It becomes a lot harder to keep track of what each player is doing. It takes a lot more mental effort to keep track of what every player is doing, instead of just one. To compensate, you can cheat a bit; don't be afraid to write down each player's strategy, or counts of key things in the game. You can even look through decks (so long as you shuffle them back afterwards).
  • The metagame becomes irrelevant. This is probably the biggest problem for me; each player's strategy will always be based on what *I* think is the right move. My best way to compensate for this is to play an occasional game with a player I designate the "psychopath", who will commit to an unusual strategy and ignore my usual priorities. In addition, I try and pursue a different strategy with each player, though this is not always possible or practical based on the game.
  • Anything based on social interaction is completely irrelevant. This isn't usually an issue for me, based on the types of games I usually make, however, this may not be the case for you. The best advice I can give for a social game is to use the "random" method for decision making.
    • A useful statistic I can offer is that in The Resistance, where one team must deduce the identities of the other, the win rate should favour the team aiming to hide information, so that when played randomly, they should win 65-70% of the games. This is because human deduction is usually good enough to compensate for the difference. (I can't find the source for this, but this was based on a quote from the designer)

This method doesn't work for every game, but is a powerful tool that every designer should employ. It can greatly improve the speed of your development, and the quality of the playtesting sessions when playing with other people.

No comments:

Post a Comment