Designer Diary: Figuring Out How to Reach The Heavens of Olympus

The Heavens of Olympus is the culmination of my first major attempt at designing a game. Because of this, I learned a lot of hard lessons about game design while iterating from version to version. Looking back, it's obvious now how many of my early ideas were doomed before they started from a sound game design standpoint, so feel free to learn from my experiences as I give you a complete tour (warts and all) of the evolutionary history of this game.

In The Beginning

The genesis point for this game came from trying to combine two simple concepts:

1. a bell curve, and
2. the diminishment of an ignored resource.

With respect to the second concept, two games I would consider influences are Power Grid and an old game I played once called The Stock Market Game, due to the prices of items rising and falling in both games. However, I was thinking of a "diminishing mechanism" along the lines of the advertising world paradigm in which one puts money into a promotion, but over time interest in the promoted product dies without continued investment. This idea would manifest itself in a manner similar to those other games in that something was rising and falling, but the marker would fall on its own and rise only when the players caused it to.

I set up a basic game board with columns of varied heights in the shape of a bell curve and had an advertising marker that advanced and fell, causing spaces to be active. Pieces were placed on the board in the squares of the columns. The advertising marker moved on smaller squares between the columns. At various times in the game, the marker would either drop to a lower space or fall back to a space to the left. I also had a separate round tracker as the number of rounds in the game differed for differing numbers of players (so as to somewhat equalize the total number of turns). Players scored points for having the most pieces in a column and for connecting their pieces in horizontal or vertical lines.

Theme, Theme, Where's the Theme?

These early designs had barely any suggestion of a theme. Instead, they were very much mechanical manifestations of an attempt to simulate some sort of abstract "business" growth...and an advertising venture...thingy. My first attempt at ascribing theme was to say that the pieces on the board represented business locations. (Selling what I didn't know.)

Early Problems and Lessons

I quickly ran into the difficulty of effectively scaling a game for differing numbers of players. If I made a board big enough for a larger number, it was too big for a smaller number and vice versa. My early solution was to introduce different boards to the table for multiple players. This was an atrocious idea because having different boards resulted in players focusing on only one board and ignoring the others so as to maximize their efforts. If a player tried to divide his efforts, he would lose. I also attempted to say – very embarrassingly, I might add – that the different bell-curved boards represented different cities in which the players were building.

Initially, the advertising marker caused spaces to be active, referring to the spaces being available for the placement of new markers (which I quickly realized was too restrictive). I then changed it to refer to whether an item was eligible for scoring purposes.

At this time, the design had only one advertising marker instead of each player having his own. This created a problem with players not wanting to be the one to raise the marker's value because all of the players after them would benefit from their efforts.

The columns each had two numbers at the bottom: the left number showing how much a player earned for placing a piece in that column, and the right showing how much it cost to buy out another player - that is, to remove one of the opponent's pieces from the board and replace it with one's own color. Over time, I realized how problematic it can be to give players the ability to remove another player's piece because whoever got picked on the least was often the winner and not necessarily the player who planned the best.

These early problems gave me a crash course education on basic game design principles.

A Theme! (Sort of...)

I eventually decided that having different column heights with some spaces being shaded would be how to scale the game. (This decision was somewhat influenced by how scaling is handled in Ingenious.) I put everything on one board. I portrayed the columns as different cites on their own and the pieces on the board as "Toy Stores" (which was a weak attempt at infusing some sort of fun in a themeless game). I put in a score track and included the round tracker. By now, the game had taken on the name "Market Share".

After a while I became convinced that I needed to get serious about a better theme, so I looked for various things in the world that were shaped somewhat like the bell curve columns as a source of inspiration. I decided on the Hawaiian island of Oahu, and the game became a travel agency game called "Agents of Paradise".

This gave rise to the pieces on the board being "tourists" and I came up with various and sundry explanations (most of them weak) for what was going on in the game, such as calling the scoring of connections of pieces "positive customer feedback". Additional actions were added, with players having a dial to select an action (not unlike the dials in Maharaja). By this time I had learned that the economics of the game promoted a runaway leader problem, and I implemented a wonky fix called a "Tax Bracket" system (because, hey, taxes are high in Hawaii). I also bought knick knacks to add for thematic purposes and put together a more formal prototype.

Game Night Games and The Board Game Designers Guild of Utah

Throughout this time, I was frequenting (and eventually working as a part-time employee at) a local board game store in Salt Lake City called Game Night Games. The manager of the store at that time, Greg Jones, gathered contact information from customers interested in game design and organized an initial meeting for these people in January 2007.

We got together at the store and discussed forming a club for the purposes of playtesting each other's designs. (By this time, my game bore the Hawaiian travel agents theme.) The group would eventually be known as The Board Game Designers Guild of Utah. Little did we know what kind of group we had formed. Several people at that initial meeting would go on to become published designers, with Sean MacDonald's Pastiche and Alf Seegert's Trollhalla being the most recent releases from this group.

Meeting Jay

After several meetings of BGDG, Greg happened to be at a game convention where he talked with Jay Tummelson of Rio Grande Games and informed him about what was going on with our group. Jay decided to attend a summer board game convention in Utah in 2007 and look at many of our game designs with the understanding that, a year later, he would come back, hold a formal contest, and publish the winning design.

When Jay came in 2007, he looked at a number of our games. After sitting down with me and letting me explain how my game worked, he asked me a series of questions that went something like this:

Quote:

Jay: "Why is it that a tourist in this city would just so happen to be able to communicate with a tourist way over here in this other city such that 'customer feedback' increases?"

Me: "............."

After a few more questions like this to which I had no viable answers, he left. I was devastated.

I was also in denial for a while afterwords. Instead of accepting Jay's criticisms and acting on then, I continued to pitch the game as it was. Zev Shlasinger from Z-Man Games took a look at the design, but turned it down due to the game, as he put it, having too many components compared to the level of play it offered (among other things). Again, I was devastated. After all my hard work and hours of toil working over the numbers and pondering ideas, two different publishers had each found separate fundamental problems with my game.

Beginning to See the Light

After some time, I accepted that my theme needed to change and that I needed to simplify the component count. I also met my future wife during this time. Her maiden name is Starr, and that name inspired me to imagine that the connections between similar pieces in the columns could actually be construed as constellations. That realization led to more changes and the introduction of the game's final name: The Heavens of Olympus.

Despite that breakthrough, I was still in denial and mired in the paradigm of my own design. I was thinking about nothing more than cosmetic changes to components and the theme. The problematic runaway leader fix was still in place, players still had six different actions from which to choose (which was overly complicated for this type of game), and the columns on the board hadn't really changed.

Decision Time

Jay's questions continued to eat at me as I thought about him returning to Utah and judging the upcoming game competition. I finally came to an important decision:

Every aspect of the game would have a plausible thematic explanation – even if I had to scrap all of my efforts and start from scratch.

This lead to me asking the following: "What kind of board would plausibly involve stars and constellations?" The answer was a solar system-style board, so I began with a brand new board concept. I kept the idea of earning points for having the most things in an area, only now players wanted circular orbits instead of vertical columns. This scrapped one of the two core concepts I had started with - the bell curve - but the game was finally starting to come together.

The lesson here:

Be willing to cut something fundamental from the game if it interferes with the game becoming good.

Zev's feedback also continued to eat at me, and I eventually accepted that I needed to start looking at my game from the standpoint of it being a potential product and not just a game - meaning that I needed to both simplify the components and make sure the components were easier or less expensive to produce from a publisher's perspective. I also tried to make sure everything in the game was handled with symbols instead of text to facilitate potential multi-lingual considerations.

With this new board concept came the next obvious step, which was to represent planets with small discs instead of the cubes I had used in previous versions of the game. (After all, punch-out discs would be a lot less expensive than cubes if the game were published.) I then looked at the different actions and stripped out those that were creating complications. This simplified the design to what are now the four basic actions, a change that allowed me to use four cards as the player interface instead of a more complicated/expensive dial.

With respect to scaling, I decided to divide this new solar system game board into five regions and use a number of regions equal to the number of players (similar to how Power Grid uses more or fewer sections based on the number of players).

During different rules iterations, I had used various sets of coupled numbers at the bottom of each column to show money gained or spent. For this new board, it occurred to me that I could also use a left and a right number - but to set up a scaling amount of points based on most and second-most planets in an orbit and have the right number double as a minimum requirement for earning the higher points.

In previous versions, I had set up varying costs for actions based on others selecting the same action in the round. In this new version, I simplified the costs to be a one-for-one payment for every other player who selected the same action for the round, a system much easier to remember. Also, when a player places a new planet, he now earns money based on the number of planets in that region of the board.

I set the game at five rounds and made each round a "day" comprised of three phases (morning, afternoon, and evening). Scoring occurs during the night phase when Zeus comes out, looks upon the heavens, and awards prestige points. This change simplified the rounds and phases of the game, while simultaneously reinforcing the theme.

The advertising marker is now a torch marker and is located on the side of the board (thus recalling an archaic, yet fun take on how the planets "shine" during the night). I made each planet double-sided, with a light side and dark side. Now players have to choose which of their planets aren't eligible for scoring if they don't have enough torch light to keep all of their planets "lit" throughout the night, thus creating an additional and consequential decision point.

I took out the previous runaway leader fix and – thanks to playtesting with other BGDG members – received feedback on thematic additions, such as having points for the biggest constellation. The sum of all of these changes resulted in this being the new game board:

Meeting Jay Again

When Jay came back to Utah in 2008, I presented him with a very different game that had undergone fundamental changes. The theme was interesting, the game mechanisms more streamlined, and everything that happened in the game made sense within the theme.

My game won, and he accepted it for publication.

Heavens then entered the development stage, with some numbers being adjusted and tweaks from Rio Grande improving the overall game. I also added in a new runaway leader fix that was more intuitive, simple, and fair with respect to the torches dropping. Mirko Suzuki, Claus Stephan, and Martin Hoffman – the guys behind the artwork for Race for the Galaxy – were enlisted by Jay to do the artwork and I was blown away by what they came up with:

Keeping the Good

Now The Heavens of Olympus is approaching its release and in spite of the many revisions throughout its history, a number of early concepts I implemented have remained in place in some way throughout the iterations of the game:

• Players buy pieces from a supply and first place them on an individual player board. Later, these pieces are eligible to be placed on the main game board.

• Putting pieces on the board awards the players money, which they can use to buy more pieces, which can earn them more money...

• Points are awarded for having the most pieces in an area and for "connections" of pieces of the same color when scoring occurs.

• The game uses a "go first" piece for the round (similar to the Governor card in Puerto Rico) and a "go first" piece for the phase, which rotates through some of the players before the round ends.

• The diminishing marker stayed in place in some form or another throughout the various prototypes.

• Players decide to take actions during a phase via simultaneous action selection.

Though I kept some things along the way, the most important thing I learned in this process was figuring out how to let go of or change those things that weren't working, even if I was emotionally or intellectually attached to them and even if they represented one of the key motivations behind designing the game in the first place.

My hope is that players will enjoy the game that has emerged.

Mike Compton