Leaving the Terracotta Warriors Behind

Today, I’m performing a bit of a post-mortem on a game design I’ve been working on for a while. Ed Mariott’s post about quitting Brooklyn Bridge was very insightful, and while it didn’t inspire me to quit, it gave voice to some of the same things I was feeling. The game was initially developed as “Terracotta Warriors”, a little over a year ago, as I was first showing New Bedford in public. The theme is about the Terracotta Army of the first Qin Emperor. It’s a fascinating subject that I was drawn to based on a PBS documentary.

At the heart was a great twist on the role selection mechanic. In the established genre, players select from the same handful of roles to perform a variety of actions based on the tableau they have built. Each role has a bonus that the person selecting it receives, while everyone gets to perform a basic action. My twist was to have each role provide a unique bonus. Players build a tableau of the role cards they have previously selected. Into this system, I added resource collection, so each of the basic actions is associated with a resource type, with the unique roles providing permanent benefits, one-time-use abilities, immediate gains in resources, or a scoring bonus. Each basic action was also associated with a paradigm of earning points: slow but steady point gain, occasional multiple-resources to big points, trading for cash worth points, and a bidding/auction for points. (The final category was no points, but had a lot of powerful special abilities) The unique roles were thematically tied to the fact the each terracotta warrior (out of thousands) was unique.
(Fidelitas, now on Kickstarter uses a similar idea of suit plus unique action, but has a much different core mechanic. It’s worth checking out.)

It was initially a bit cumbersome to use, because I had tokens for all of the resources. I eventually switched to a resource track for convenience, but that added some problems with the auction mechanics. I also had problems with setting up the game for various player counts (ambitiously 2-6). It was initially a completely random distribution from the entire deck of 100 cards, requiring a balance between specialization, which tended to be powerful, and generalization, which was important for mitigating risk. But the potential for runaway games based on luck of the draw was too great, so I had to rework setup into something more complicated, but still reasonable for a 90 minute game.

When testing with friends, they seemed to like the initial concept and where the game was going. These are serious gaming friends I can trust to tell me the truth about a game. But the theme just wasn’t working. I didn’t have a clear idea in my head of who players were and what they were doing. I think this is absolutely critical to designing a game. If you don’t know what players are doing, how should they. I attempted to focus the theme more around the actual production of the army, but this didn’t feel much better.I started considering alternate themes, but nothing felt quite right. Based on the relationship between the basic actions and resource types, something always felt out of place.

I spent a while retheming it to something a little more accessible, building a wild west town.  The roles all become unique characters that players select to come live in their towns. While this was more thematically interesting, it presented another problem. The theme was not very tightly integrated with the mechanics. There is nothing wrong with making a more abstract game with a loose theme, but I was having doubts whether that was even the type of game I wanted this to be. I was very fortunate to get this on the table in front of a number of excellent designers,. While the response was generally positive, they echoed some of my own private concerns while suggesting excellent ways to streamline and improve the game. I saw that I needed to simplify the choices to speed it up, adjust how some of the major mechanics work, but probably most importantly, trading needed to be available at any time, to keep players from getting stuck.

The issue with this big change is that I now need to replace that basic action with an entirely new mechanic. So far I have been unable to find something that fits in with the scheme relating actions to resources to scoring methods, that also fits within the larger theme of the game. And so I am left with a rather large hole in an otherwise functional game. As I look back, the game has come quite far from the original concept, but that concept now seems to be holding it back. So I am left in the position where I can drop the remaining parts of the original game to create something entirely new, or leave everything behind.

So I have decided to put the idea on my shelf. The resources need to be disassociated from the roles and scoring. Scoring may even need to be a separate abstraction layer. The current 1-to-1-to-1 relationship between them is too limiting. I think the core mechanic is still great, but I haven’t found the right use for it yet. I need to establish the basic behavior of a game, and then use the mechanic to put it in motion. It’s a great game design lesson that you need a lot more than just a great core mechanic. Even when you have a game that works, you have to ask yourself if it’s the best possible version of that game. If something doesn’t feel quite right, it probably isn’t, even if you can’t identify what specifically is wrong. Following that intuition is crucial to game development. Sometimes it is something you can solve, and sometimes you need to walk away.


  1. Great read Nat! This was especially interesting for me as I also had a design using a terracotta army construction theme up until late last year. When I was working on it I read somewhere that the Terracotta Army was one of the earliest instances of an assembly line being used. The concept I was working on has since developed into two unrelated ideas now but it is really a fascinating theme to explore.

  2. Rock solid point in that last paragraph. I’ve been trying to express that same concept in my blog (is something seems off, it probably is). Great read.

    As for your game, it reads like a fascinating Draft/deck building game. I haven’t played it, so I don’t know. It sounds, however, like a big portion of your trip up is the relation between the roles and the rest of the mechanics. If that’s right, why not remove the roles and allow it to be a more symmetrical game?

    Either way, I look forward to seeing where you go 🙂

    • You’re absolutely right, the problem is that everything is too tightly linked. I don’t want to completely eliminate the roles, because that takes the game too far from the original concept I was going for. There is still a good game in that, it’s just not the one I wanted. I think I can still use the original idea but I need to separate the core roles from the other systems. And right now, I’m too close, so I need time to forget it a bit.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.