The Case for Limiting Complexity and Depth

A lot of times when people talk about strategy games, a discussion about the differences between complexity and depth arise. This is a topic that can give one a lot of insight, but it’s been expounded upon in far better ways than I can offer. For example, here is a great article by Sulla on that very topic. It’s well worth the read!

What I would like to discuss is what realistic limits should be placed on depth and why this should be so. Ultimately, I think it’s important to set limits in order to both avoid overwhelming players or to trivialize the game. Let me elaborate…

There’s a concept in psychology known as the “Rule of Seven”. Although there is a lot of disagreement about how valid it’s specifics are, the concept arises from the common sense notion that there is a finite number of complicated elements that the human brain can juggle at one time. Not exceeding this limit, whatever it may actually be, will be a characteristic of a well-designed game. As a guiding principle, I want to ensure that a player in Remnants has no more than seven of these complicated elements to juggle at any given time.

For starters, what is a “complicated element” in a 4X strategy game like Master of Orion? In my view, it is a functional subsystem of the game that interacts and affects other subsystems of the game, while also requiring specialized knowledge to use that doesn’t apply to those other systems. In effect, these subsystems are like “mini-games” within the game that the player must master in order to win.

Let me clarify by listing what I think are the eight functional subsystems of the original Master of Orion (and therefore Remnants).

1. Scouting and Colonization – this subsystem actually represents two of the 4 X’s on a 4X game: Exploration and Expansion. In MOO, each is simple enough and they are interrelated enough to consider them as part of a single subsystem required to expand throughout the galaxy. Think about the skills you need: optimally target unscouted systems and prioritize which systems you want to colonize first. It’s not always about scouting the nearest planets or colonizing the best planets. Sometimes a player will follow patterns that are less optimal for strategic reasons, like colonizing a well-placed Ultra Poor planet first to ensure that an enemy can never reach an inner region of more valuable planets.

2. Research – deciding what to research and when to rush-research important technologies. A player’s research strategy literally affects every other subsystem on this list. Decisions made here can have cascading effects throughout the game, so a good knowledge of which techs are more valuable for your strategy is essential.

3. Colony Development – while getting your colony up to speed is not as complicated in MOO1 as it is in other games, the player needs to decide when to push for population increase, build ships, build missile bases, or focus on research. These decisions will vary on each colony based upon their maturity, their resources (or lack thereof) and strategic position on the galactic map.

4. Ship Design – one of the most interesting subsystems in MOO involves the design of your ships. Because of interactions with specials, some players try to build lethal combinations of ship stacks. Others focus on brute force damage. There is also the consideration of building dozens of huge ships or literally thousands of small ships. Personal preferences, racial bonuses and your technology base all play into this subsystem.

5. Fleet Deployment – this is the skill of knowing where and when to exert your military power against your opponents. It is possibly the most abstract “mini-game” in MOO and a well-timed surprise invasion (or perhaps the just-in-time arrival of a defensive fleet) can often make the difference between winning and losing. When players come back to win from against long odds, it’s often because of their skill in this subsystem.

6. Ship Combat – maneuvering your ships in tactical combat is another important mini-game. Which fleets should you target first? Should you chase down their fleets or kite them? Is retreat an option? Perhaps a mad but suicidal dash to bomb the enemy planet is the best goal. All of these decisions are unique to this subsystem and important to learn.

7. Diplomacy – this subsystem revolves around the art of manipulating your opponents into doing what you want them to do. Who do you choose as your enemies and friends? Is it worth giving up a valuable technology to get the Alkari to break their alliance with the Bulrathi so that you can take over those bear-infested planets?

8. Spying – the last important subsystem in MOO that I could think of is spying. Espionage or sabotage? Whom to target, what technologies to steal. We don’t really need that computer technology, but it will make future spying a little easier. Can we knock out some missile base on a planet we want to invade? Who should we frame? Although the MOO spying mini-game is abstracted quite a bit, there are still a lot of interesting and meaningful decisions to make. And if you are playing the Darloks, you can control the entire galaxy through effective spying.

Some may have noticed that there are 8 subsystems listed but I have said that a good limit should be 7. It’s important to realize that not all of these systems are active at any given time. For example, by the time that you need to worry about the complexities of “Ship Combat”, the “Scouting and Colonization” phase of the game is almost over.

But it’s not just about the cognitive capacity of the players. After all, as players get more familiar with the game it becomes easier to juggle more subsystems since they are proficient with the others. We see this all of the time in the 4X genre as experienced players ask for more and more features while new players are often overwhelmed with the game.

Another thing to consider is that increasing the number of strategic subsystems in the game has the unappreciated effect of devaluing them all. For example, if you have 7 well-balanced subsystems in a game, then each one should arguably contribute 14% to your chances of winning against another player or a very competent AI. But if a developer adds too many subsystems, they become increasingly irrelevant. Twelve subsystems means that each one contributes only 8% to your chances of winning. This means that a player is free to completely ignore some subsystems and still win against any AI except perhaps the strongest.

Allow me to make a sports analogy to illustrate further…. football vs basketball. What is the relative merit of a scoring event in basketball vs. a scoring event in football? Obviously, the difference is huge. A basketball team may score 50 times in a game while a football team scores 4 to 6 times. Every scoring attempt in football is potentially critical to attaining a win, whether it’s in the 1st quarter or the 4th. In basketball, however, many fans don’t really tune in until the last 5 minutes of the game because that’s when the important baskets count. You never see a sports replay highlighting the big 3-pointer made early in the game (because no one cares).

In the same way, it’s important for a strategy game to have enough subsystems so that there is enough interaction between them (i.e. “depth”) to make the game interesting, but not so many that they become trivialized. A player who can win consistently without even bothering with seemingly important systems in the game will often conclude that the game is too easy.

So what does this mean about incorporating additional subsystems into a post-MOO1 version of Remnants, such as Leaders, Strategic Resources, Colony Improvements, Golden Ages, new Victory Conditions, tactical Ground Combat, etc, etc etc. It means that great care must be taken to ensure that these changes provide enjoyable variety and replayability, but don’t create so much interaction that they increase the complexity and depth of the game. We often forget that too much of a good thing is actually a bad thing. So many developers seem to give into the urge to throw in everything but the kitchen sink that, while players get what they wanted, they nevertheless often walk away from the game with a vague “meh” feeling (see: Stellaris?).

Every 4X game designer has arguably over a dozen well-trodden strategic elements and tropes to choose from when making his game. My position is that the best-designed 4X games are those where that developer picks the 6 to 8 elements he thinks should be in his  game and then makes those work as well as possible.

Those are my opinions on this topic. Thanks for reading!


8 thoughts on “The Case for Limiting Complexity and Depth”

  1. I always thought that most of the additional elements added to MOO2 took away from the game. I hated the planet buildings. Leaders were nice. What I wanted was MOO1 with leaders.


      1. I would say that leaders in MOO2 were more of a flavor item than a subsystem. although they did interact with ship and planet systems; they were, for the most part, a fairly minor influence–they were fun to pick and place, and occasionally move around, but you did not win of lose based on leaders. They provided enough effect to be memorable, useful and interesting, but not so overpowered that any or them unbalanced the game.

        They WERE a huge immersion factor. Everyone remembers being pleased to get Rashiki or Mentar early, and no one wanted the worthless Yota. The fact that most of us can still recall many of these leaders by their individual names is a testament to the power of this idea when well done. You tell me the name of a Stellaris leader or governor, and i will send you a cookie–I can’t offhand name a single one, but Mystic X I still remember.


  2. Interesting read. I have always preferred increased depth and complexity, to the point that few (unmodded) 4X/grand strategy games satiate that need. Having read your article I’m trying to reason why. Perhaps in not being new to the genre, and given the dozen or so tropes you mention, one has already mastered a few mini games before even starting the game.

    An interesting system is seen in Distant Worlds: Universe where the player can turn human involvement in the subsystems on and off to varying degrees. I got most enjoyment playing this on full manual (indeed, I didn’t see what all the fuss was about until I did). Giving players the option, though, and a relatively competent AI, seems like a relatively unexplored direction in dealing with the question of complexity.


    1. Actually, MOO3 dived head-first into the “automated subsystems” idea. It didn’t work. Players could win or lose the game and have no idea why.

      I wrote an article that directly addresses this several years ago if anyone is interested.

      Here’s a quote:

      “If complicated systems were “abstracted” into the interface by simply posting a few simple
      symbols or numbers on the screen, the complexity would mean nothing. The player would
      not understand what was happening under the hood. The complexity of the system would
      be lost on the player, who 1) does not understand all the intricacies that the developers put
      into the system, 2) cannot interact with the system to influence it in any meaningful way,
      and 3) probably doesn’t care anyway. For the player, a few simple rules would have been
      better. Can you still build complicated gameplay mechanics without a simplistic interface to
      hide it? Only if the player can comprehend the rules.”

      Liked by 1 person

      1. That article looks incredibly familiar – I may have read it a few months ago!

        Granted, MoO3 may have got it wrong, but I don’t know if that can necessarily be said about the example I provided. A lot of the design principles outlined in your document are sensible and would probably produce great games, but that’s not taking into account personal preference.

        Your document, when adhered to, would probably help produce a game that appealed to the widest audience. There are those amongst us who do like to try and understand what’s happening under the hood and, indeed, would be willing to spend hours trawling through forum threads, and run experimental campaigns, to get to grips with it and revel in our triumph over the complexity cast before us!

        My preference for increased complexity and depth doesn’t exclude my appreciation for a well-crafted strategy that limits those same qualities. I just don’t think I’ll get as much playtime out of the latter for the very reason that there’s less for me to mentally engage with.


  3. Level of complexity wasn’t the issue with MOO3. 4X games are naturally complex and appeal to people who like to tinker with the options. The problem was that it automated many of these systems. Under the hood it was running a complex simulator, but the player never interacted with that. They just ended up with a glossed-over high level view.

    We also need to distinguish between complexity of rules and complexity of strategy. ‘Go’ is a simple game but has complex strategy. I would rather 4X games lean in that direction as well, but it seems like the current crop of games is all about more and more rules.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s