ords of Waterdeep is a new strategy board game set in the world of Dungeons & Dragons, debuting this month. In this Euro-style game, players send off their Agents to recruit Adventurers and use them to complete Quests. Below, designer Peter Lee describes how this new D&D board game evolved from the original design into the final game.
When I'm doing any sort of creative process, I really prefer using "iterative design." While the term may be new to you, I'm sure you've done something in your life that used this process. Even the simple task of adding salt to your food is an example of iterative design.
An Example of Iterative Design
You sit down for dinner, take a bite of food, and discover that it's not quite right. You add a little salt and pepper, then taste the food again and consider if you've got the flavor right. Congratulations! You've just experienced iterative design.
This simple example illustrates the three steps you use in iterative design:
- Create or Refine
- Evaluate and Repeat
In the case of dinner, you have created the meal. Once the meal is ready, you test it—usually by tasting it. You then evaluate it—did you like it, or does it need a dash of seasoning? What makes iterative design powerful is that you now repeat the process. If you feel the food is lacking a little salt, you refine the food by adding a little salt, test with another bite, and evaluate if you've added enough seasoning. Creating a game is no different, except that each step takes a little more time (and isn't nearly as delicious).
Designing Lords of Waterdeep
When Rodney and I started work on what would become Lords of Waterdeep, it was a pet project. We didn't have a firm schedule for the game, so it had a leisurely development process. We sporadically worked on the game design for about nine months, putting in only a couple of hours of work each week.
Creating means developing a prototype. This entails getting your thoughts out of your head and onto paper, cardstock, playing cards, and other media. We'll go into more details on creating the prototype for a game next week.
Testing is the fun part: you play the game. This can be just a single play-through, especially early in the design when there are a lot of rough spots. Later on it takes a couple of games to see what works and what doesn't work for a new iteration. At the beginning, these games are for testing the big concepts. As the game progresses, you worry more and more about the fine details. Eventually, you start showing it to new people to see what areas other people may have trouble with that you might think are obvious.
Evaluation is the last step of the cycle, and also the most difficult. Unfortunately, it's not as easy as tasting food. You need to decide if a game is fun and how you could make it more so. It's helpful to have a strong idea of how you want the game to work. Without a clear direction, the changes result in a random walk; it's much easier if you have a goal. For Lords of Waterdeep, our self-imposed direction was:
- Euro-style mechanics with a strong D&D theme.
- 1 hour play time.
- The Action draft is how you get resources.
- The Quests are what you do with resources.
- The Intrigue cards are how you mess with other players.
After each game, we'd discuss what we liked and what we didn't like within that iteration. We'd usually come to a consensus on what to change, make those changes, and then try again. Whenever we'd come to a problem that didn't have an obvious solution, we'd pick a change that seemed likely to meet the game's goals and try it. If it succeeded, then the change stayed. If not, then we'd try a different solution until we found one that did.
With that in mind, let's look at various mechanics in the game and how they evolved throughout the design and development process.
The Fifth Round Agent
In Lords of Waterdeep, you take an action by assigning an Agent to a Building. The number of Agents you have is effectively the number of actions that you can take in a round. In any worker placement game, if it's possible for you to acquire an additional action, your best course is to get that additional action as soon as possible. We liked the idea of increasing your available actions, but we didn't like how purchasing it dominated strategy. It was Rodney's idea to just give everyone an additional Agent at the beginning of the fifth round. It took a couple of games to convince me, but I definitely think it was the right thing for this game. The fifth round Agent provides excitement in the middle of the game, it adds tension to the Action draft, but it doesn't dominate your strategy. It's been part of the game from the very beginning.
We didn't add any other additional Agent mechanics until the introduction of the Plot Quests near the end of development, as seen in Recruit Lieutenant.
Evolution of the Basic Buildings
The first version of Waterdeep had ten basic Buildings, but none actually made it into the final game unchanged.
The five Buildings that provide Adventurers and Gold originally accrued resources throughout the game. For example, the Field of Triumph was originally called the Arena. At the start of each round, you'd place two Fighters on the Arena. When a player took an action at the Arena, he or she would gain all the Fighters that had accrued at the Building. During development, we removed the accrual of resources as it slowed the game down between rounds. We saved the mechanic for a cycle of Advanced Buildings.
Castle Waterdeep was originally called the Meditation Gardens, and all it did was change the Starting Player. We also had a Building called the Lair, which provided two Intrigue cards. As we played, we realized that the first player space needed to do something in addition to getting the first player marker, and we were worried that the Lair was adding too many Intrigue cards into the game. We combined the two Buildings into one.
There was a time in development where we didn't have a Building that allowed you to take the first player marker, but instead the role of first player rotated throughout the game. I don't think the rotating first player space worked for two reasons: it didn't seem fair in a three player game that some players would have more chances to go first, and players would often forget to move the marker to the next player between rounds.
The Builder's Hall was originally called the Marketplace. The only difference between it and the final Building space was that when you'd put a new Building down, you'd put two VP tokens on it instead of just one. This change was made to streamline play.
Three basic Buildings were removed as the game evolved. I already mentioned the Lair, the space that was folded into the creation of Castle Waterdeep. The other two spaces were the Docks and the Library of Waterdeep. The Library provided either a Wizard or a Cleric. The Docks provided both a Fighter and a Rogue. For a while, the Library changed to the Yawning Portal, a basic Building that gave you one Adventurer of any type. These Buildings didn't accrue Adventurers, and they showed that the accrual mechanic was unnecessary. Once accrual was cut, these spaces were no longer special, so they were also removed. However, the Yawning Portal survived to become an advanced Building that provides two Adventurers of any type.
Two Buildings didn't exist at first because they weren't needed: Cliffwatch Inn and Waterdeep Harbor. They didn't exist because drawing Quests and playing Intrigue cards occurred instead of assigning an Agent. At that time, you could do one of three things on your turn:
- Play an Agent.
- Draw 3 Quest cards, keep 1, and put the rest on the bottom of the deck.
- Play an Intrigue Card.
This worked, but it made the game more difficult to learn and it strayed from the point of an action drafting game.
The Quest Hall was the first basic Building to be added. We were a little worried that the game would break down if the Quest Hall was always filled, so we made it three different action spaces. At first, it rewarded a varying amount of Gold depending on how soon you went to it. The first Agent gained 2 Gold and a Quest, the second Agent gained 1 Gold and a Quest, and the final Agent just gained a Quest. Future iterations worked to solve the problem that the Quest Hall would get clogged with Quests that weren't interesting to any of the players, so we added the ability to clear the Quests in the Quest Hall and replace them. We also realized that we wanted more Intrigue cards back to the system, so we added the Intrigue card space to make the eventual Cliffwatch Inn more interesting.
Waterdeep Harbor was added near the end of the game's development cycle. We liked how playing an Intrigue card delayed your action, but we didn't like people playing a ton of cards in a single turn. Waterdeep Harbor was our solution to simplifying the turn sequence yet keeping the same mechanical effect. We started with 5 spaces, but we cut it down to 3 after playing a few games.
Evolution of the Lords
The Lords took a while to get right. Early in the design, back when they were still Sorcerer-Kings, they gave players an asymmetric benefit, such as gaining a Fighter at the start of each turn or gaining 2 VP each time you play an Intrigue card. Once we moved the game to Waterdeep, it didn't take us long to associate these characters with the powerful ruling council of the city, the Lords of Waterdeep.
As the game developed, we started to notice of a handful of minor flaws. New players were finding it difficult to form long-term plans. We really liked the score track for showing who was in the lead, but the end of the game was anticlimactic, as you have a very good idea who's going to win. (We tried some Intrigue cards to solve this issue, but more on that below.) And the Lords were also lacking one primary thing: the right flavor.
In the Forgotten Realms, the Lords of Waterdeep wear masks to hide their true identities, so having the players' Lords be public information seemed wrong. As it happens, turning the Lords secret until the end of the game also provided solutions to both the long-term planning issue and the lack of tension in end of the game. We tried a couple of different planning options, but having each Lord gain points for completing a particular type of Quest ended up being the best.
Along the way, we learned that scoring based on leftover resources wasn't a great idea: if you gained points for Adventurers in your tavern or Intrigue cards in your hand, it created pressure to not use those resources and thus be inactive in the game.
One design variation had a Lord giving Victory Points to whatever player that completed the most of a particular type of Quest. For example, Khelben Blackstaff gave 10 points to the player who completed the most Arcana Quests, even if that wasn't the player who drew the Lord. There were some interesting games where someone other than the owner gained the points, but it happened so infrequently that it wasn't worth pursuing in the final game.
Intrigue Cards and Player Interaction
Intrigue cards had the right purpose from the beginning: these cards are the way you interact with other players. Originally, this just meant delaying them. Mandatory Quests are one such delay—an Intrigue card that you play on another player to slow him or her down.
We tried a few variations on Mandatory Quests, such as not allowing you to draw new Quests until you completed your mandatory ones, but this was a case where the first design really was the best thing for the game.
Near the end of our original design period, we added Intrigue cards that provided resources to you and another player. I was surprised by the first game that featured these cards. These cards caused the players to form alliances. Player 1 awarded player 2 a small boon, so when player 3 played a similar card, he gave his boon to player 4 (because players 1 and 2 had already benefited). I loved it. The rewards are asymmetric to make sure they are still a viable choice in a 2-player game.
As I mentioned earlier, we had a few Intrigue cards that gave you extra points at the end of the game. One example was a card that gave you 1 VP for each Gold you had at the end of the game. While that sounds perfectly innocent, it had some major problems—a player now had a reason to not spend Gold on completing Quests or buying Buildings, effectively short-circuiting the game. This card and similar cards were added to provide tension at the end of the game, but they didn't actually accomplish that goal because it wasn't assured that any of the players would have one. We removed the entire mechanic, which provided the space for the final Lord of Waterdeep role cards.
Finally, we had a few Intrigue cards that basically canceled other people's Intrigue cards as they were played. These cards were cut because they caused a lot of activity that resulted in absolutely nothing happening. This isn't fun in a game where there already aren't a lot of cards being played. These Intrigue cards were removed early in development.
Evolution of the Quests
We already talked about most of the changes to Quests in last week's column, so this will be relatively short. Here are a couple of minor changes we made to the Quests.
When we introduced the concept of Quest types, each Quest had two types. A Quest could be Skullduggery and Piety, or Commerce and Arcana, and so forth. The Lords gave points for only one type of Quest. We switched it to make it easier to identify Quest types, which made it easier to play the game. We were able to change the Lords of Waterdeep to each give rewards for two Quest types, nearly doubling the number of Lords we could include.
We tried a few Quests that forced other players to discard a Fighter when completed. When we had an assortment of these Quests, it caused people to concentrate too much on what other people's Quests did, slowing down the game. We realized that since Intrigue cards were something that were specifically intended to mess with your opponents, other game elements generally shouldn't. The game is about completing your Quests, not worrying about other players completing their Quests. In the end, the only remaining interactive Quests either allow you to play Intrigue cards or force you to provide assistance to another player for a greater VP reward.
Evolution of the Advanced Buildings
The Advanced Buildings also changed as we iterated through the design. At first, when a player used an Advanced Building, the owner would gain only VP. We found it was more fun when the owner could also gain Adventurers and other resources, so we made VP rewards occur only on a small fraction of the Buildings.
The early design had a small game board, and purchased Buildings would sit in front of the player who owned them. We tried a mechanic that gave an ongoing benefit once that Building was purchased. For example, the City Walls prevented the owner from suffering the effects of Attack Intrigue cards. The actual benefit when someone took the space was preventing Attack Intrigue cards from affecting you for the rest of the round, and as a result the Building was never taken. (This Building was intended to be used only in 4- or 5-player games.)
When Advanced Buildings sat on the side of the board, it was difficult to see where you could assign an Agent. We changed it so you'd place them on the board and attach an ownership marker. This meant some of the complicated Buildings like the City Walls had to be cut.
We had a few Buildings that were very interactive. For example, the Council of Lords allowed you to take an Adventurer from another player. The Shipping Lanes gave you either 5 Gold or let you take 2 Gold from each opponent. The Assassin's Guild House made one of your opponents lose an action that round. All of these were removed for the same reason that we avoided interaction on the Quests—direct player interaction is best done through Intrigue cards.
We had a few Buildings that just weren't good for the game. The original Port of Waterdeep allowed you to discard up to two Intrigue cards, gaining 4 Gold for each Intrigue card discarded. We didn't like using Intrigue cards as a resource, so the Building was cut.
The Manor in the Noble House was an action space that just gave you 6 VP when you took it. It was removed because it made Quests less exciting, but the idea of gaining VPs from Buildings came back when it was tied to a secondary benefit, as seen on the Heroes' Garden.
We had a lot of Buildings that accrued resources just like the base board did. Imagine New Olamn gaining a Wizard and a Rogue each round! They were all simplified when resource accrual was removed from the game, and only the most basic accrual Buildings were added back.
We noticed in some of our early playtests that the players who were winning didn't buy any Buildings at all. We knew that this had to change: if buying Buildings wasn't a good decision at the beginning of the game, it makes the 5th round frustrating when the players gain the additional Agents. We solved this in two ways: we increased the amount of money that players gained at the start of the game, and we reduced the cost of all but the best Buildings to make them more appealing purchases.
Next week, we'll shed some light on how we prototype games. See you then!