This game. It looked amazing in my head – sleek, elegant, black, sophisticated, mysterious,… I could go on. I couldn’t wait to make it. But there was no getting around step one: the delicate balancing act of trying to make the rules work. I felt like whenever I plugged one hole, water would leak in from another. And I was starting to feel the pressure of the time limit.
As an English Literature undergraduate, I was never able to quickly whip out imperfect draft after draft of an essay, each one slightly better than the last in the journey towards a satisfactory final product. Instead, my approach towards writing was painfully slow, because I felt every sentence I wrote had to be perfect before I moved on. Unwittingly, I held on to this attitude whilst designing this boardgame. I forgot that play does not need perfection – the evolution of a child’s game isn’t motivated by the fear of failure (if the creation of a game with unbalanced rules can be called such), but rather by curiosity. What if there are two robbers this time? What if you have to find me first? What if…?
Luckily though, this curiosity was finally sparked again after a coursemate who playtested the game remarked that it reminded him of the Mao Kun Map in “Pirates of the Caribbean”. Consisting of several concentric rings, the map could be rotated to reveal mystical locations otherwise concealed from the reader by its intricate design. At last, I felt there was a way to incorporate the AP (action point) system represented by the phases of the moon (the moons could be placed on the outermost ring) while simultaneously adding a new, engaging mechanic to the game that would make it necessary for players to visualise the board several moves ahead if they wanted to succeed.
I researched whether such rotating boards had already been implemented by games before, and I came across “Everdark” by Walrus Games (I have written a case study in a separate post dedicated to board games that inspired mine). In “Everdark”, the board consists of four rings around a small centre, each of which is divided into twelve segments marked by white lines. One rotation of a ring is complete when these lines meet. With the phases of the moon on the outermost circle, it made sense to divide my board into eight slices – each corresponding to one phase of the moon – and like in “Everdark” make a 1/8th rotation cost one AP (represented by a star token). This meant that I could reduce the amount of constellations on the board for a less cluttered design, as single APs were no longer useless later in the game but could instead be used for rotating constellations into place.
As this idea required a rotating board, the process of creating a paper prototype became more complicated. Firstly, the constellations needed to be arranged so that that they still intersected with one another, but they also needed to be distributed on the different rings of the board in a way that required players to twist them into place to complete constellations. I began by making one prototype from cardboard with two rings around a centre, but quickly realised that this was not a viable option as it was just too complex – especially if I was going to add a third ring with the phases of the moon on it. The next two prototypes both only had one ring with constellations around the centre, but varied in size and type of zodiac sign depicted on the board. There were still minor constellations which represented safer investments, but now only had three rather than six larger zodiac signs on the board as volatile but lucrative investments to make the design less cluttered.
Moon Phase Integration
Although the addition of the rotation mechanic provided some strategic depth that I felt was previously lacking, there were still persisting issues with the AP system. I knew I wanted a system that was somehow connected to the phases of the moon, as I felt they added to the design aesthetics. Additionally, I also wanted to avoid having a constant number of APs for each turn, and instead felt there should be some connection with the number of APs per turn and the moon phases so as to integrate the moons as more than just decoration. The asymmetrical approach I had tried in the first week had not worked (see post about Week 1 for more information), but this was perhaps due in part to the fact that it had involved both a rise and fall in APs over the course of eight turns. What if the number of APs per turn only grew?
The first time I playtested the game with steadily rising APs (2 per turn, 4 per turn, 6 per turn, etc) was the first time the game was described as enjoyable by a playtester, and I had to agree. The rapidly rising number was not as much of an issue with the addition of the turning mechanic, as many points were spent on rotating rather than occupying the board. In addition, as the board started filling up, players also had more APs at their disposal to remove stars from the board. The game also became quite fast-paced, as players could eventually place over eight stars on the board at once, which gave me hope that it could be kept within the 10 minute timeframe limit we had been given for the project.
Although the shorter time-frame was desirable, the ridiculously large number of APs by the last turn was not. It became difficult to track what had been done in a turn – not to mention that the sheer amount of tiny tokens was annoying and impractical – which made it clear that the number of APs could not exceed eight per turn. The first player was also still at an advantage in this configuration , so I cut down the amount of turns to four rather than eight, and tried the following AP system: 1-2, 3-4, 5-6, 7-8. However, by the last round, the second player had had four more actions than the first, which gave them too much of an advantage.
To try and get around these issues, I experimented with making the different moon phases objectives that give certain rewards. In other words, players would not automatically progress from one moon phase to the next after each turn, but instead had to gather a certain amount of points to progress to each phase, which then would reward them with more APs than they previously had. In one iteration of this idea, the first moon phase gave players two APs, and the second rewarded them with three, but players also needed at least a score of three to progress to this second phase. The third moon phase required a score of nine, and gave four APs, and the fourth and final moon phase (at this point the board was split in two and each player inhabited one half of the phases of the moon) required a score of thirteen to reach, and thereby win the game.
I hoped this approach would help balance first/second player advantage by making APs dependent on strategic thinking and skill rather than simply on turn order, but instead I had created a feedback loop. With each success, the player was given more recourses in the form of APs, so it was easier for them to succeed again. This is an example of a positive feedback loop – a self-reinforcing process which “puts emphasis on the early game, since the effects of early-game decisions are magnified over time“. It also amplifies players’ successes and losses by making it difficult for a player recover once they have fallen behind because the winning player is awarded momentum that helps them leap ahead.
A negative feedback loop, on the other hand, serves to balance progress between two players, making it more difficult for a wide score difference to develop. To create one, I had to make it more difficult for players to achieve success after success, for example by making players have less rather than more APs after completing a constellation. To do this, I changed the rules to have players remove stars that were used for completing constellations from the board, meaning they effectively relinquish their control of the board with every success.
The game was immediately felt much more balanced, as it was easier for players lagging behind to take advantage of the empty board and catch up to the more successful player. Although this version of the rules worked fairly well, there was still not enough interaction between the two players. Players were in competition with one another in the sense that they were each trying to score constellations, but because constellations were only considered complete if every point was occupied by only one player, it felt more like they were racing alongside each other rather than actually directly engaging with each other.
Inspired by the famous game “Reversi” (otherwise known as “Othello”), in which players capture enemy pieces by flanking them on opposite sides, I decided to change what I thought was a core aspect of my game: the way in which constellations are conquered by players. Instead of needing to occupy every point of a constellation, now players simply had to be the one to make the final, completing move in order to win it. For example, even if the black player owned every point in a constellation apart from one, the white player could still win it by placing the final star – or by just rotating the constellation into place.
This mechanic encouraged players to complete zodiac signs using each other’s stars, because it was more cost efficient than using one’s own. As a consequence, the game became more cut-throat, giving players the ability to create traps for each other, and thereby forcing them to be cognisant of the amount of their opponent’s available APs and attentive to what they were used for.
Although this was the closest I had come to creating the type of strategy game I had wanted, this did slow the game down, as players spent up to one minute considering each move. In order to add some speed, I reverted the AP system to a simple rising pattern, which I felt also made more sense with the narrative I had in mind for the game. Having split the moonphases in half (one half ending with the New Moon and the other with the Full Moon), there were four rounds in total. The first player had one AP in the first round, while the second player had two to mitigate first player advantage. In the second round, both players had three APs, five in the third round, and seven in the last.
The two players were thus working towards opposite sides of the board, with one player gaining APs as they approached the New Moon, and the other as they approached the Full Moon. On a narrative level, this translated easily to two witches – one Black and one White – who each gained power under different moons, battling against each other for dominion over the night sky.
I also ended up returning to six constellations on each side of the board (I planned on building it in a way that it would be reversible, with the remaining six constellations on the other side), because safe investments in the form of small constellations were no longer necessary now that constellations could be completed using the opponent’s stars. As a result, one side of the reversible board represented the Northern hemisphere, while the other represented the South. All that remained now was simply to manufacture the game – the part I was most excited for.