Roadblock is a series of articles where I interview other designers, developpers, and others involved in the industry, to do a deep dive into a specific issue they’ve dealt with in a project. The goal is add concrete examples to the mass of game design advice out there.

JV: First, can you give us a bit of background on the game at that stage?

I had just recently signed Tumble Town from the designer, Kevin Russ. He’d only been working on the game for a few weeks, but I could immediately see the potential of the game. Tumble Town was always a town building game set in the Old West, but initially, it didn’t have any engine building or spatial puzzle aspects. Just simply building a town out of dice in the West. 

JV: What was the problem, and when did you first encounter it? 

The end game of Tumble Town was an issue at the start of the design process. Kevin had made the game a specific number of turns and I had never liked that in games. I really enjoy when I don’t exactly know when the game will end and when my choices in the game can make it go on longer or end it early. 

JV: Had you ever encountered a similar problem before? Why was this one different?

Every game has to end in some way, but each game is different and has different actions and components that can lead to the end game happening. Tumble Town needed to go on long enough that players could build up an engine and feel successful in doing so, but I didn’t want the engines to get too out of hand where it meant that you had only one real path to victory. 

JV: Can you talk about the process of solving it? What worked? What didn’t?

The first idea was supposed to be a fix of two problems; adding in dice mines. For the dice mines, there would be dice that you didn’t have to roll that you could take if you took a building plan from that specific row. Having dice that you didn’t need to roll meant you could plan your turn a little more and the game would be a bit less random. When two dice mines run out, then the game would be over. 

The dice mines worked to solve those two problems, but they introduced a new problem; players were confused on what dice to roll and what dice not to roll and would often forget or do the actions backward. I try to make games as intuitive as possible, so this definitely was something I wanted to solve. 

JV: So did you go back completely on the non-rolled dice?

Yep! There’s no dice that you gain that you don’t roll. I found that it’s so much easier to get players to roll the dice, if all the dice have the same rules and are treated the same. Making things easy is definitely something I prioritize. 

JV: What else did you try?

The next solution was to use Plan End cards. These cards would be placed under a specific amount of building plan cards based on the player count and the game would end if two of the plan end cards were visible. This worked better in that players weren’t confused on how things worked, especially since the Plan End cards had the game end trigger written on them. However, the problem with the Plan End cards was that the game end was very variable. If players all took from the first row, then went to the second row, no one ever got to the third row of cards. I like when games can end early, but not when players haven’t experienced a third of the game. That’s a bit too early. 

The solution that actually got a good end game was having the dice supplies run out. There’s four different kinds of dice, with gold associated mainly with level three building plans, brown with level one, and gray and black both with level two. With two colors for the level two building plans, it meant that even if most of the players did focus on level one and level two buildings, the game would only end early if the players someone only went for black OR gray buildings. As most players tend to base what building they want on other factors, usually the gray and black dice are taken at about the same rate and the end game is triggered when the gold and brown dice run out. 

JV: So both of those last two options give players some control over game length, but an unusually short game is now something that skilled players do on purpose, rather than new ones doing by mistake?

Yep, that’s exactly right. I think it’s more interesting as I always like for things to happen when players deliberately choose to do something, instead of when they don’t realize that they’ve done something. If I want to try to rush the game, if I succeed, I’m more happy with the experience, even if I don’t end up winning, as the plan I had worked out. I also like when there’s several different strategies in games, as you now have a strategy based on a long game, average game, and a short game. Players will have to play a number of times to see what works best for each game length. 

JV: When working on end game conditions, what do you set as a goal? Do you aim for a specific length of time? Do you aim for a specific moment in-game? Do you base it on the game arc?

I usually try to make the game less than an hour and also for the game to end before players want it to end. If players are still really engaged when the game ends, they’re way more likely to play again than if the games ends a turn or two after a player is bored of it. 

For Tumble Town in particular, I wanted the game to end after a new player can build a level three building (the level with scoring conditions based on the other buildings in your town) and still get a chance to build another building or two which will gain extra points based on the level three building they built. 

JV: How do you feel about “play one final turn” in games? Do you default one way or the other when you design?

I’ve done both! In the past, I’ve usually done equal amounts of turns and finishing out the round. Recently, I’ve tried where all other players but the player who triggered the game end gets another turn and I really like that for games that are based on player count. It means that you have to be in the lead to trigger the game end, but you might not end up the leader after everyone takes their last turn and any hidden scoring is added. 

For Tumble Town, I went with finishing out the round, as having two colors of dice run out means that half the buildings most likely can’t be built and the other colors of dice might also be close to running out so even having an extra round might mean that no one can do anything, unless they have stored dice. 

JV: Personally, I usually push off designing an end game trigger: I’m not planning on finishing the first few tests anyway, so why bother? Are you the same way? If so, do your testers also complain about not finishing the game?

I usually have a non ideal end game trigger in the beginning; either a number of rounds, point value, or a deck running out. Sometimes the end game stays, sometimes it doesn’t. For the very first test, I might even not tell players what the end game trigger is until we play a few rounds and I can kind of see where the game is going. 

JV: Can you think of a published game with a particularly well-designed end game trigger? 

Rajas of the Ganges has a really great and unique end game trigger that I love. There’s two tracks in the game, money and points. When one player’s markers cross, it triggers the end game, so you can focus in one aspect, the other, or try to do both simultaneously.

JV: Well thank you for your time Carla! I think that knowing how to end a game is a difficult skill that is much about nuance, and I’m sure sharing your experience will help many others! 

Tumble Town is currently on Kickstarter! If that sounded interesting, go back it here!