Designer Diary: Québec

Part 1. Birth of the Game

Québec arose from the merging of a concept and a mechanism. One evening in 2007, following a game night at the home of Martin Bouchard from Filosofia, Martin suggested the idea of a game that would commemorate the four-hundred-year anniversary of the city of Québec.

Pierre Poissant-Marquis took the ball and ran with it, suggesting a game that showcases the different powers which have influenced Québec over its history: the church, which was so important in the colony's early years; the politics and wars which marked the 18th century; commerce, naturally; and the effervescent culture of the city's inhabitants. As a good historian, Pierre took note of the interactions of the different spheres of influence. That was when the idea came to me to have an area majority game in which victory in one region would have the consequence of improving your lot in a neighboring region. The concept of the cascade was born.

Two other principles were soon determined, and they are still present four years later in the final version of the game. Firstly, following my experience with the design of Panama, which had multiple phases, I wanted to design a game with only a single phase, i.e., a game in which players take actions without worrying about which phase they're in. (Think The Settlers of Catan versus Caylus.) In Québec, the flow of play is interrupted four times during the game for scoring, but between those interruptions are only single phase turns.

Since the beginning, also, I wanted to use a mechanism which I'd been desperately trying to perfect for several months. This mechanism – which I affectionately baptized "I offer and you play" – allows the players to choose the actions that their opponents will be able to play. In the first iteration of the game (see the illustration at right), the board contained several spaces upon which the players placed the building cards that provide special actions. On his turn, a player placed his architect token on a card, making the special action available to the other players; he himself, though, was not able to use it.

Shrewd strategists will initially propose an infallible strategy: Always offer the least useful actions possible. That strategy, however, soon reveals itself fatally flawed given that at the end of the game a player receives a large bonus for the buildings upon which his opponents have played.

From the first tests it was obvious that the "I offer and you play" system worked well and presented the players with agonizing choices: "I'd very much like that action. Should I wait for others to offer it to me, or should I offer it myself and be assured of completing the building?" In any case, other small details needed to be tweaked.

Image from an early prototype

Part 2. Bugs and Debugging

A game the size of Québec is rarely born perfect. Even though the original version, conceived in January 2007, soon appeared interesting, it was obvious that it could benefit from adjustments to its secondary mechanisms.

One important aspect of Québec is the passage of four centuries, which form the background narrative for the game. From the beginning we wanted the number of turns per century to vary to increase the tension as the game unfolded. To achieve that, several mechanisms were proposed, tested and rejected.

The most memorable was "the wheel of time", shown at left, which was introduced in the ninth iteration of the game in April 2007. After each action the players rolled a die which progressed a marker around the wheel. Each time an hourglass was passed, time progressed ten years. New hourglasses could appear on the wheel, bringing the end of the century closer. While the idea was interesting, it interrupted the rhythm of the game by requiring frequent die-rolling.

In fact, that problem was inherent with the addition of almost any supplementaty mechanism to manage the passage of time. We experimented with different methods of integrating the end of a century with the mechanisms of the game. In the final version, where the die has disappeared, a century can end in two different ways: by running out of buildings or workers. The result is more fluid and produces a tenser game because players can rarely predict in advance what will cause the end of the century.

Another problem we encountered was directly linked to one of the most interesting mechanisms: the cascade. Originally, the majority player was able to move half of his cubes into a neighboring region without no limitations. This rule, although obviously interesting, was crippling to new players who were surprised by the sudden influx of cubes into a zone of influence. Furthermore, between experienced players we saw merciless struggles for domination of the first zone, struggles which caused heavy losses to those who didn't win the majority. After a lot of research we observed that the simple constraint of limiting the number of cascading cubes to five could fix both problems. In the first instance, a majority of five cubes or more was now solid and couldn't be affected by the cascade. For the second, it became less rewarding to invest more than ten cubes in a zone because the extra cubes didn't increase the power of the cascade. A nice example of the simplest solutions sometimes being the most effective...

The last important problem concerned the construction of buildings. As shown in the image below, the building tiles have evolved a lot. At first, the game didn't use star tokens. Only the buildings with three compartments occupied by groups of workers were considered complete, and all others went back to the box. The difference had very important repercussions for the game. First, the players were constructing fewer buildings and the groups at the end of the game were much smaller. The worse problem, however, came once again from players new to the game. Those players tried to complete every one of their buildings, which was obviously impossible (but not to them) and they frequently lacked active workers, which forced them to displace their architect. The star tokens, introduced in the winter of 2009, proved to be the solution. They made it possible to add a building to your network even if it hadn't been entirely finished.

The three preceding examples demonstrate the work required to transform raw ideas into a solid game. The job wasn't done though! To end up with a high quality game it was necessary to balance the different actions and ensure that no combination was excessively powerful.

Part 3. Final Rules

At this stage of development, with the major problems solved, we knew we had a solid game – but we also knew it needed a lot more work. Repeated plays and tests on players new to the game revealed small faults: too powerful strategies, excessive and easy-to-forget maintenance, and rules that were still too complex.

In March 2010, when Le Scorpion Masqué took the game on, certain objectives were set: Reduce the play time to 90 minutes or less (instead of two hours); reduce the maintenance, particularly between centuries; streamline certain elements; and integrate the events into the game more harmoniously.

For example, the final rules allow the activation of three workers when starting a new building. Originally, though, the number of workers activated depended on the district, as shown in the illustration at left. This additional variable didn't really enrich the strategic aspects of the game, but it did cause headaches for players who had to take into consideration the action of the district and the number of cubes needed to play there, as well as the number of workers activated. By changing the rule to be three workers irrespective of the district, the game instantly became more fluid. We also decided to require players to activate three workers during the initial set-up. As a consequence, the number of errors in play decreased considerably.

The arrangement of the hexagons for districts and buildings on the final board is also the result of a long refinement process. The first version, shown on the left below, was playable but lacked some tension because there weren't enough opportunities for blocking. If you analyze the board you can see that most buildings have five neighbors, facilitating the creation of large connected groups of buildings; in the final version, shown on the right, most of the buildings have only four neighbors. Of the configurations tested, this offered the best balance between opportunities to connect and to block, while maintaining a symmetry that didn't advantage any player. Furthermore, adjacent buildings are rarely the same color so a player must offer different actions if he wishes to create a significant group. In the first version, a player wanting to develop his main group often had to offer the same action to his opponents, which turned out to be not very interesting. Even though the board is slightly less legible in this configuration, we decided that the advantages were worth it.

The final scoring of the value of groups of buildings was another aspect which needed lots of adjustments. At first, only the buildings in the player's main group scored points. However, that rule discouraged interaction as players concentrated on their main group without trying to block their opponents. Also, large differences could appear during the final scoring – if a player had his main group cut early enough, he no longer had any interest in completing his buildings. Furthermore, for the remainder of the game, players had to place their star tokens so that they would not be worth any points. Several attempts were made to fix this problem, the challenge being to maintain the tension of the connection game while ensuring it didn't dominate. The final solution, which was proposed by one of the playtesters, was to grant a bonus for buildings in the player's main group while also counting the others to a lesser extent.

In the first versions of the game, all of the sixteen additional actions were concerned with sending cubes to the four zones of influence. These actions soon became repetitive, and we looked to diversify them. The addition of the citadel and several brainstorming sessions allowed us to come up with a few new actions.

Multiple test sessions were necessary to select a suite of actions which were at the same time balanced, yet different enough to offer real choices to the players. One of the testers even maintained an Excel spreadsheet counting the utilization of each of the additional actions to ensure that each had a roughly equivalent tactical or strategic usage. It was a real puzzle, probably the most difficult task in the whole tuning process. To impose some order on the resulting collection of actions, we decided to group them thematically, giving them colors matching the districts and giving the players clues to find the action for which they were looking. The red political actions are those which allow you to gain influence by sending two workers at once to a zone. The blue cultural actions offer the chance to score victory points. The yellow economic actions accelerate the construction of buildings, and finally the violet religious buildings permit "intrigues" or more complicated actions.

The development of leaders also caused a lot of headaches. They needed to be useful or else nobody would ever bother taking them – but if they were too powerful, they'd become a required play, and the first turn of each century would simply be role selection as in many other games.

It was a challenge to balance, but I think we've done well. The last leader we proposed was the grey one allowing a player to send three workers to the Citadel. It introduced a subtle mechanism which opened several doors to us: an instantaneous action. Most leaders are interesting at the start of a century, as a player who takes one wants to benefit from him for as long as possible. The unique action of the grey leader, however, is most interesting at the end of the century. It helps maintain tension. This new option helped with the development of the two-player version (in which you find several other leaders with instantaneous effects) as well as in the development of the goodies which will be available at Spiel 2011 when the game debuts.

Development of the events was relatively easy, compared to the rest. The manner in which they entered the game caused the most difficulty. At first we wanted them to come into play in the middle of a century. Behind the building tiles were sand timers which set off the events. Even though that was thematic, it didn't work at a mechanical level due to too much bookkeeping and useless complexity. That's why we turned to a solution less novel but strangely more effective: A single event was revealed at the start of each century, and its effect lasted for the whole century.

As a designer, it's nice to see our players and testers arguing in favor of their favorite action or leader. To us, that suggests the game is both engaging and balanced. We believe you yourself will enjoy getting into the game and finding your own infallible strategy!

Philippe Beaudoin