Reflection on War of Zones

It was hard. It wasn’t just the game that was hard.

Working on this project taught me many things, the challenges of modeling a system within a physical game, the challenges of working in a team with many conflicting ideas, and most importantly — play-testing.

Play-Testing The Mechanics

First of all, there is no such thing as enough play-testing. It’s never enough. But the question here is rather, “When do you stop testing?”. In our game, we took a look at different cities across the world to build a system of zoning. One mechanic I’d like to talk about is the evolution of the event cards (based on real-life events across cities of the world) that affect the board in various ways.

Initially, we decided to go with a global event, drawn at the end of every round, that stays in play for the next round and gets removed after all players have played. Frustrated, our testers expressed their confusion on when the events are to be played. Are event statuses played as a separate phase before the turn phase? Or directly before their turns?

We tried again, with the status cards played directly before each player’s turn. “Not fair”, people complained, “the first player with the status gets a much better deal in this.” Furthermore, we observed some bored expression while they wait on others to complete their turns with the effects. “We need more interaction with the event cards,” said one during feedback.

Again, we iterated. This time with event cards to be drawn on each turn, instead of once per round. We definitely saw much more engagement with the events now. However, “This card is so broken”.

There was some balancing to do.

Now, we added variability to the number of event cards, instead of unique events only in the event deck. We make powerful cards rarer and weaker cards more common. Still, “There’s too many homeless!”.

More balancing to do. We’ll reduce them, again.

Again, we tried again. “There’s still too many homeless”. “We need more rezoning cards”.

More balancing.

However, other than the balancing complaints, there was no other. Players were significantly more engaged with the events. “Flash flood! Someones dying today!” We noticed many players getting very excited about the cards they draw, especially during the final round before scoring.

Looking at this, we see a pattern here. From many complaints to fewer complaints, from bored to excited. Balancing problems remain the same.

So, when do you stop testing the mechanic?

First, your gut feeling from observing. Do you think people liked the mechanic you were testing for or do they not?

Next, the emotions in players. Are they engaged in the particular mechanic you were testing for?

Lastly, the complaints. Have people stopped complaining about the same mechanic you are testing for?

With the above, consider the benefits of doing more testing on this particular mechanic. How much more would this mechanic and the overall game improve if you were to keep testing this specific mechanic?

A note on balancing.

The truth here is you can never truly reach a completely balanced state. It’s an ever-changing process that needs to be tested indefinitely. One metric to determine when more balancing tests aren’t worth the time is when the complaints on balancing have reduced to a very small amount.

Conclusion

Play-testing is the most important phase of any game you make. There are many parts to a game that models a system, and it’s important to test out each mechanic well. You may think you have a great time playing the game yourself during development, but it’s not about you. It’s about the people playing your game, and it’s important that they like it.