SimCity BuildIt, Contest of Mayors
Intro
So far a good chunk of my game design career has been spent working on one title, SimCity BuildIt. It remains the game to which I am very much emotionally attached, probably because I learned so much while working on it. Over the years there have been very many features that we introduced as updates. Many of them were game changing in the most literal sense. In this post, I will take a deep dive into a feature that I am probably the most fond of. We have introduced it relatively early on, and it still remains after all these years one of the most beloved features in the game.
The Game
A game that operates as a service is at its worst at launch, as my good friend and colleague Markus Kiukkonen likes to say.
SimCIty BuildIt is a mobile free-to-play game. It is a part of the venerable old SimCity franchise started off by legendary Will Wright back in the mid ’80s. As a game, SimCity BuildIt is inspired by the other SimCity games with which it shares a lot of its DNA. However, it is a very different game, as adaptations had to be made to accommodate the demands of the mobile platform, the game as a service business model.
As all games in the SimCity family, this game puts the player into the role of a virtual mayor, i.e., a person charged with managing and growing a virtual city. SimCity BuildIt is a sprawling game. Trying to summarize its internal mechanics can be a daunting task. However, at its core this game revolves around two types of puzzles that the player needs to constantly keep solving.
The first layer of the puzzle is about managing the urban layout of the city. The virtual metropolis is divided into residential and non-residential zones. Residential zones need to be covered with various services and amenities, such as police protection, firefighters, sewage, water supply, etc. The placement of city blocks matters. Different kinds of buildings within the city require different types of resources. As the city expands, so do the needs of its virtual inhabitants. Player needs to constantly manage these resources in order to keep his citizens happy. This “smart placement” puzzle is the strongest connection between SimCity BuildIt and other SimCity games.
The second layer of the puzzle is tied to the production of the resources. In order to grow his city, player needs to keep on building new and upgrading existing residential areas. Both of these actions require a certain number of different resources, items such as wood, bricks, nails, even pink donuts. The player can produce these items in crafting facilities in the city. Some basic items can be further combined to produce more advanced ones. Producing any type of items requires time. The player can schedule production of the items by using the production queue of individual crafting buildings. This constitutes the time management puzzle that the player needs to keep on solving over and over again. This type of game mechanic is typical for many games in the farm simulation genre, for example Zynga’s FarmVille, SuperCell’s Hay Day or Township by Playrix.
At the time of its global launch, the game focused almost entirely on these two layers of the gameplay. It offered only rudimentary means of social interaction. It was almost exclusively a single player experience. Players did have a possibility to exchange some of the resources via an inbuilt trade system, a useful, but relatively shallow social mechanic.
The game was envisioned as a service, but the groundwork for most of its live service operations had yet to be laid down.
Due to limitations of the schedule, the game was launched with a limited amount of content. Our live service during the first year consisted of regular client update releases, packed with a certain amount of new content, usually new buildings with the new architectural styles. These updates were always very warmly received by the players and had a positive impact on our KPIs (Key Performance Indicators). Furthermore, players would consume the content in a matter of days or weeks at best, while it took us months to produce it. This content treadmill was clearly unsustainable.
Goals
In order to change things we needed a different kind of feature. We needed something that would be evergreen. Something that would keep our player engaged without a need to constantly manufacture new sets of visual assets.
Our business and design goals could be summarized as follows:
- Stabilize the KPIs,
- Introduce an evergreen feature,
- Breakaway from the content treadmill,
- Introduce PvP (Player vs. Player) gameplay.
The feature that we came up with is still an integral part of SimCity BuildIt. In fact, it has become one of the pillars of its metagame supporting several other important features.
Concept
Let’s be clear, “Contest of Mayors” as a feature is nothing revolutionary. It rather represents an evolution of ideas and features that are common in other games of this genre. Similar features are almost an industrial standard. See for example Derby in Hay Day.
The feature itself is a sort of a mini game, with its own core and the feature metagame.
As the name implies, the feature at its core is a contest. It is a weekly event that runs on the regular schedule, originally starting each Wednesday morning and ending around noon on the next Monday. Players compete with each other in order to achieve the highest score in this allocated time. Rewards are given at the end of each competition round.
Players collect points by completing assignments which are related to other core aspects of the SimCity BuildIt, i.e., the aforementioned area and crafting/time management puzzles.
These weekly repeating contests are organized in a system of leagues, which form a meta structure of the feature.
Assignments
The assignments, the tasks that players need to complete in order to collect points, are the beating heart of this feature. As I mentioned, these assignments are tied to the very core of the SimCity BuildIt. Some of the assignments, for example, require the player to produce a certain number of items of a specific type. Others require him to upgrade residential zones, or fix the damage caused by other players. Yet, other tasks require players to simply collect taxes or earn a specific amount of one of the game’s numerous currencies.
Over the years we have implemented more than one hundred different assignments for this feature. What all of them have in common is that they revolve around the actions that the player is in any way supposed to be doing while playing the game. Furthermore, since the list of different assignment types is so broad, an individual player can focus only on those assignments which suit his playstyle.
Assignments are presented in a form of an endless list. At any moment of the competition, the player can choose from twelve different assignments. If the player manages to complete one assignment, a new assignment is automatically added to the list. This list is generated locally, for each player at the start of each competition round. Each player sees only his own list of assignments. The system takes care so that the lists offer the same opportunities to all players.
Each individual assignment has its own time limit, i.e., the amount of time within which the player needs to complete it. The player needs to accept the assignment by tapping a start button. The player can do only one assignment at a time.
KEY IDEA: By accepting an assignment, the player makes a conscious commitment to try to finish the work!
This conscious commitment is important. It is much stronger than the alternative in which a player could simply claim the points only after casually completing the assignment.
If the time expires before the player manages to complete the assignment, the assignment has failed and the player receives no points. Each player has only a limited number of attempts at assignments during one competition round. If the player manages to complete the assignment, good for him, he receives the points, but the remaining number of attempts is reduced by one. If the player fails or cancels an assignment the number of attempts is again reduced by one.
Different assignments offer different amounts of points. They also require different amounts of time, ranging from minutes to several hours. The effort needed to complete them also differs and depends on the level of the player’s city and his skill.
Gameplay Depth
This set of constraints forces players to strategize. The choice of assignment matters. The order in which the player completes the assignments matters. Yes, the player can commit to only one assignment at a time, but he can always plan forward and prepare for the next assignments. Some assignments are a natural preparation for the others. The Wood produced in order to complete one assignment can be used to produce Chairs needed for the next one, etc.
For many assignment types the player can prepare in advance, before starting them, reducing the time pressure significantly. Others are deliberately designed to be trivial, such as collecting taxes, etc.
These intricate interconnections between different assignments offer an opportunity for players to form tactics. They also offer a possibility to play the contest on different levels of engagement. A player can play it casually doing only a handful of assignments in the order they are presented on the list, or simply picking the easiest ones.
Other players on the other hand can go on to create elaborated playbooks of strategies and play patterns in order to maximize their score while minimizing their effort.
KEY IDEA: The rich system allows the players to choose the level of engagement, i.e., play casually and still get results or go deep and create engaging strategies.
Somewhere on Reddit there is a complex multi-page guide compiled by some of our most dedicated players. I was amazed by its complexity and the depth of knowledge about our own system accumulated by these players. Some of them even tried to reverse engineer the behavior of our randomization algorithm in order to gain a sliver of advantage over others in the competition.
This depth of knowledge is where the true skill of playing the SimCity BuildIt lies. It’s our original time management puzzle on steroids! Time management is a very adult skill. This is something that we grownups have to do every day. Most of our daily routines revolve around time and resource optimization. This is why games such as these can feel so satisfying. We might do a subpar job at running our actual lives but we can ace the Contest of Mayors in the SimCity BuildIt.
Furthermore, this depth of knowledge serves another purpose. Advanced players can transfer this knowledge to other novice players. They can mentor them about the most efficient strategies. They can reinforce their social standing within the circle of other players in this very powerful way.
The assignment list has yet the third purpose. Individual assignments are a powerful guiding tool. By trying to complete assignments and reach the very tangible goal of collecting points, a player builds on and expands his own city. He also learns about new ways to play the game, and how to optimize his own creation. We have been using assignments as a way to drive player’s attention to particular aspects of the game. As a rule, whenever we added a new feature to the game, we would also introduce a new set of Contest of Mayors assignments that would reward players for engaging with this new feature. This proved as a very effective way to drive the engagement of particular new features.
Leaderboards
In order to keep the competition fair and balanced and the gameplay exciting, we divide the participants in the contest in, the so-called buckets of up to one hundred players.
These player buckets are formed all over again at the start of each competition round. We group the players according to their level.
Players are ranked against the other players in their bucket. This ranking is displayed on a leaderboard. Each bucket has its own individual leaderboard. A relatively small number of players on each leaderboard ensures that the competition stays exciting. As the standing on the leaderboard is refreshed constantly, completing each new assignment can lead to the change in the player’s standing.
In the past, I have already written about the best practices of leaderboard design. See Building Better Leaderboards. Much of what I know on this matter is derived from my experience with the Contest of Mayors leaderboards.
Furthermore, we have created a ladder of six leagues. At the end of each competition round, the best performing players from each bucket are promoted to the next league. Likewise, the bottom ranking players are relegated to the lower league. This adds another layer of the game, yet another set of goals on a different scale. Completing an individual assignment requires minutes or hours, winning a contest round is a matter of days, while climbing the league ladder requires weeks.
As only a few players from each bucket are promoted each time, the buckets form a pyramid, with lower leagues having more players and competition buckets than the higher ones. The promotion system is balanced in such a way that there is only one bucket in the highest league. The winner of this bucket is the overall winner of the weekly competition round.
Rewards
Of course, players take part in the Contest of Mayors for an intrinsic reward that success in a competition can offer. However, we have also implemented an elaborate system of extrinsic rewards.
Player’s success is measured against the success of other players in the same competition bracket. Players are ranked in descending order according to their score. At the end of each round they are given the rewards based on their Leaderboard standing.
We have divided the leaderboard in five different sections. The top ranking player gets the Grand Prize. The next two players receive a so-called Big Prize. Next seven receive a Medium Prize. The players ranked at places 20 to 50 receive a Small Prize, while everyone in the bottom half of the leaderboard receives no prize at all.
As I mentioned before, we were determined to stay away from the content treadmill as far as possible with this feature. Therefore, the prizes that we decided to give to participants of the Contest of Mayors consist entirely of resources, a selection of currencies and items.
The players can use these rewards in the core gameplay to further develop their cities. However, for dedicated Contest of Mayors players the best use for these winnings is by reinvesting them into the next round of the contest. In turn this creates a positive feedback loop reinforcing further engagement with the feature!
Effects
What I have described is a pretty elaborate structure. The combination of individual assignments, weekly competition rounds, reward tiers and league ladder produces a very dense selection of goals with different time scales. Immediate goals of completing individual assignments require several minutes to several hours. Moving up the leaderboards to each the next reward tier requires several hours to several days. Completing a competition round is a week-long effort. Reaching the next league can require several weeks, while climbing all the way to the top league requires at least a month and a half of intense gameplay.
This density of goals allows for a very flexible gameplay. Each player can decide which set of goals to pursue in each round of the competition. One player might play a relaxed game and collect only a Medium Prize in one of the lower leagues. Others might go all in in an attempt to win the ultimate prize. Yet others will be satisfied with floating comfortably somewhere in the middle of their current league.
Furthermore, players can adjust their effort level from week to week and even during one competition round based on the stiffness of the competition that they are facing.
This freedom of choice over which goals to go after is complemented with the freedom of choice of which assignments to engage with. In other words, the player is able to choose how to reach his goals. This combination satisfies the player’s inherent need for autonomy.
The skill is hidden in the depth of the gameplay. On the other hand the learning curve of this particular feature is not particularly steep allowing most players to satisfy their need for game play mastery.
Finally, this is essentially a PvP feature. Every single player is pitted against a selection of other players. He can easily see his place in this group by glancing at the leaderboard, satisfying his need for relatedness with other humans. Furthermore, relatedness is reinforced in the scope beyond this feature, on guild chats and web forums where experienced players can exchange their knowledge with other players.
The effect of this feature on our KPIs was amazing. Above all we managed to stabilize both the player engagement and the revenue curve. This was obviously a result of the qualitative change in the player behavior. Our players started to develop their weekly gameplay routines. They would organize their playtime around the schedule of the Contest of Mayors. What is more important, a significant percentage of them started developing their spending habits as well. We had testimonies of people saying that they spend a certain amount of money each week knowing how far this would get them in the Contest of Mayors.
The game truly evolved into a digital hobby.
Another interesting fact is that this feature did not include any direct monetization opportunities. All monetization derived from this feature was created indirectly, by players trying to use their money in a smart way to gain the advantage in the competition. The winners of the top levels of the competition were always either the player prepared to grind the most or the ones ready to spend the most. However, since there is no obvious purchase designed to make the winning easier, we managed to avoid the usual pay to win accusations.
Limitations
However, this system is not perfect. It too has some limitations. The most obvious is the relative short ladder of only six leagues. Simply put, what happens to a player that reaches the top league? Qualifying for a top league is very rewarding in any game. Winning it is even more so. Winning it three times in a row is also an accomplishment, but what about staying in the top league for 20 weeks? Let’s forget about the top league. It is anyhow out of reach for the vast majority of players. How about languishing in the middle leagues for months? This quickly becomes a thing of diminishing returns. As players reach their comfort levels of effort, they start to optimize their gameplay. Which in turn results in a stagnant gameplay.
Seasons
In general, when building SimCity BuildIt we treated all major features as platforms on which to build even newer additional features.
The origins Seasons were designed to alleviate some of the shortcomings of the basic Contest of Mayors concept. The obvious solution for the stagnant gameplay in the leagues is to shuffle things up and make the players reclimb the league ladder. As there are only so many leagues supported by the system, this needs to be done periodically, thus the idea of Seasons.
Each Season is a series of Contest of Mayors competitions. At the start of each season, all participants begin from the lowest league and try to reach the highest league in the allocated time. Our system included six distinct leagues. Getting promoted to the next league required at least one cycle of the competition. We didn’t want to make our Seasons too punishing and decided on the 8 week duration. This meant that players would have a chance to miss promotion in two weeks and still be able to reach the top league.
Unfortunately, Seasons were not part of the initial Contest of Mayors at the time of the feature launch. We simply didn’t have enough development bandwidth to implement everything in one go. In turn this meant that, by the time we got around to implementing Season, some of the people had already comfortably settled in the top leagues. We were fully aware that many of those players would not be happy to be suddenly and unexpectedly relegated to the bottom league losing their prestigious status and lucrative rewards. Therefore, and unlike the Contest of Mayors, the Seasons needed to be an opt-in feature. In other words, players would need to explicitly sign up for the participation in the upcoming season.
Again, we knew that not all players would be up for this proposition. In order to motivate them, we offered a bonus on all existing Contest of Mayors rewards, clearly communicated in the UI. In addition, we decided that we needed something spicier than resources to make the things work as designed. We settled on offering a unique building as a reward whenever a player reaches a new league, six new buildings in total each season, including the one given to players for joining the bottom league, effectively a reward for signing up for the upcoming season.
This of course brought us back into the content treadmill, which was something we set out to avoid as much as possible. The justification we found was that this setup is at least more economical, six new buildings which would be unlocked by players over a period of two months proved to be tolerable production load.
The system worked surprisingly well and we kept running it for more than two years, for 16 seasons in total.
However, it too had some disadvantages, besides the constant demand for the new content. From the player’s perspective, buildings seemed like very attractive rewards. Players used them as sort of a trophy, proudly displaying them in their cities. However, your chance to win a particular building was determined not only by your own skill and engagement but also by the stiffness of the competition in your own bucket.
From the product management perspective, since competition in lower leagues were by definition easier, our best and most engaged players had less incentive to spend for the first couple of weeks each season.
Finally, after two and a half years, we replaced the whole season structure with a better system. We introduced our version of Premium Pass.
Links
- Contest of Mayors Strategies and Tips at Reddit,
- Contest of Mayors: The Primer at Reddit
- Contest of Mayors: Task Planner at Reddit,
- Derby in Hay Day at Fandom Wiki,
- Building Better Leaderboards, my previous text
