Part 5
A board game design process: Test early, test a lot
We all know the importance of testing our designs, of putting them in front of real users. This is doubly important in board games. There are lots of variables and interdependencies that affect how a game behaves, making it a bit unpredictable once you put it in front of people. One can never know how a player will react, the behaviours that the different rules will produce and how the mechanics will interact with each other.
My best recommendation to game designers is to make a rough prototype as soon as possible and test it sooner than later.
Testing formats
There are different instances and different types of testing, each responding to different needs and phases of the project. I’m going to be talking about:
- Solo playtesting
- Quick playtesting
- Guided playtesting
- Extreme playtesting
- Blind playtesting
Solo playtesting
The simplest and quickest way to test your game is by playing it… by yourself. In this type of playtesting you take the role of every player and try the game.
While this is not going to help you with groundbreaking discoveries since you’re not testing your mental model of the game against someone else’s mental model (mental model differences is where the most disruption occurs) it’s still a useful and quick way to check for glaring issues like text legibility, ergonomics and game-breaking mechanics.
It’s a good practice to test your own game often.
Quick playtesting
You don’t want to wait a long time between playtesting but organizing a fully-fledged playtest might become quite an ordeal. It’s nice to have close friends or family members that are willing to test your game. While the same friends and family members might become familiar with the game and thus limit what you can learn it’s still a useful test to detect issues and to adjust flow and dynamics.
In a quick playtest you have more freedom to freely change the rules mid game, stop the game and start over or even start mid-game in order to test specific conditions or mechanics. You can be less strict overall. This is not recommended for more formal playtests.
Guided playtesting
Once you have something that mostly resembles a game it’s a good idea to put it in front of people in a fully-fledged playtest. While the game it’s still in a design phase, where the game is still in a state of flux, it’s better for the designer to be present and help players by explaining the rules and dispelling any doubts the players have. Regular usability test etiquette still applies. Let players make mistakes, ask for their understanding of the rules and never, ever tell them how to play “the right way”. The designer should set the stage and then move out of the way, only making adjustments and clarifying rules when needed.
Extreme playtesting
Once the game is more or less settled playtesting becomes all about refinement. Extreme playtesting is an important type here, and the idea is to test the game for absurd, out of the box play styles. What we are looking for is ways in which the game can be broken by player interaction and to be prepared for play styles that we never accounted for in the design. Someone monopolising one type of resource stopping the economy for all other players. Someone that constantly runs away instead of facing an opponent. We want to find out ways in which the game becomes unenjoyable because of unexpected behaviour and then design ways to minimize or eliminate the effects of those playstyles.
Blind playtesting
Blind playtesting is usually done near the end of the development phase. We give players the game and the rulebook and we let them figure it out. The designer should only observe how the game develops no matter what. The focus of the test is on the rulebook as the game should be mostly finished by now. We need to get to the point where people understand the game by themselves as we can’t ship a designer with each box.
It’s a good time to see how to make the game more intuitive without resorting to written rules.
What to test
Games don’t have obvious tasks or jobs to be done, so it’s not easy to break them up for testing sessions, games are systems and we need to understand how they operate as a whole. Still, we can aim to test a specific mechanic or the ergonomics of components or any other characteristic, but we have to test them in the context of the whole game.
Since games are systems, small changes can have a huge impact in behaviour. Something like letting players take one action instead of two during their turn can make for two completely different experiences. Boardgame design ends up becoming a game of constantly tweaking and questioning previous design choices.
The first playtest
The first time a designer tests a game is a magical experience. I remember mine, after weeks of working on my first game I was sure I had covered all my bases and that the game was almost complete. We came, we played and we got incredibly bored and confused. We spent an hour barely hitting each other and then we finished by ramming our ships into each other. There was too much randomness, there were too many unconnected mechanics making it hard to keep track of everything and it was too easy to be a defensive player making the game impossible to finish. The experience was revealing.
Most of my recommendations in this series came from all the lessons I got from this playtest. My biggest mistake was to try to test all the mechanics without making sure the core was right, too much clutter and too little interaction between mechanics meant that I had to keep track of several rules that were there just to patch flaws in the system.
It’s interesting that in my rush to solve many of the issues found in the first playtest I went too far in the other direction and then I got a completely boring and soulless game that felt more like filling a spreadsheet than an exciting movie scene.
After these two initial playtests I took time to think about the game and went through a complete overhaul getting it closer to what the game is today.
Test test test
Games are complex systems and thus they can lead to pretty unpredictable behaviour. Testing is the only way to actually make sure that it works as intended.
My game today looks nothing like how it started except for a few components. Mechanically it has changed so much that I could start again from the first playtest and end up with another game altogether. If you’re not sure about the value of iterative design, then you surely have to design a board game in order to really appreciate it.
Entries in the series:
- Part 1: Learn your stuff
- Part 2: Synthesizing information
- Part 3: Defining the game and its scope
- Part 4: A game is a system
- Part 5: Test early, test a lot (You are here)
- Part 6: Lessons learned
