Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 20, 2014
arrowPress Releases
October 20, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

The "perfect" strategy game
by Simon Naus on 11/26/13 02:23:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Quick introduction: my name is Simon Naus. I am part of a game studio called Monogon Games and we are currently working on an RTS-game with an entirely different approach. Throughout the development of this game we’ve come to know a lot about strategy and strategy games. What I’d like to do with this article is show you what I’ve learned and observed regarding strategy and its implementation in games and how to make the “perfect” strategy game.

First off I’d like to define the kind of strategy I’ll be discussing. Strategy is a very broad term which describes a set of ideas that bridge means and end together. So basically it’s the implementation of a plan to achieve a (long term) goal. A strategy game would be a game that requires strategic thinking to win. Every game in which your decision making rather than chance or motor skills determines the outcome is a strategy game. This describes more than half of the games I know so I’m narrowing it down to strategy regarding combat.

Let’s look at the way strategy and combat are being implemented today and have been in the past. These games are commonly divided into two types: real-time strategy and turn-based strategy. Even though this describes most strategy games, there are exceptions like the text-based strategy game—Travian, OGame--which is neither. It isn’t always possible to classify a game as one of the two. The Total War series, for example, features both real-time and turn-based phases which are equally important. Most strategy games offer low-level decision making (building bases, gathering resources and researching technologies), and high-levels decision making (how, when and where to attack).

So what would, in theory, be the perfect strategy game? Since we already defined that a strategy game relies on decision making instead of chance or motor skills we can now see what we can do to further enhance and exploit these given facts.

Motor skills
Let’s first exclude as much motor skills as possible. Since strategy games are often vacant of things like aiming and direct control of units there isn’t much motor skill required to begin with. Yet in real-time strategy, looking at diamond-league StarCraft players for example, faster and better control of units pays off. If we could remove this aspect of control and enable you to, let’s say, mind-control your armies there’d still be things like reaction time and keeping track of all your units. These aren’t technically motor skills, yet they are in conflict with requiring victory through decision making so we want those out as well. The best way to do this is to make your game turn-based. This allows each player to assess the situation and make informed decisions.

Now for chance, there are a lot of different ways strategy games implement an element of chance. Examples are hit chance, critical hit chance, random buffs and random drops. Most of the time factors like these don’t decide the outcome of a game, but if you are about to achieve a Pyrrhic victory any of these elements can rain on your parade so we shall have none of these.

Then there’s a third thing we want to avoid, which is obscurity. Hidden information interferes with the decision making process. With a lack of knowledge certain decisions are a shot in the dark. Most obscurity in strategy games is implemented deliberately and is called fog of war. Fog of war is the uncertainty in situational awareness experienced by participants in military operations. In a lot of strategy games actual fog is used for fog of war. There is much more to it though. Not knowing where your enemy is, what his numbers are and how much resources he has are also considered fog of war.

Besides fog of war there’s unintentional obscurity. This is everything else that isn’t clear to the player. If I have a huge army of foot soldiers and my enemy has a huge army of foot soldiers it is hard to predict who of us is going to win and how much units will be left on the victorious side. So attacking his army is taking a gamble, thus an element of chance. This is mostly a problem for real-time strategy games, yet there are a lot of turn based strategy games in which the only way to know how much damage your attack will do is through experience from earlier matches.

Those were the don’ts, now for the do’s. Since strategy games are all about decisions, your game should allow for good decision making and an overall strategy. Make sure there are as many options as possible in any given situation, for example, one of the other player’s armies poses a threat to one of your armies. You can attack them first, stand your ground and wait for them to attack, flee, or call in another army to resolve the situation. Each one of these actions has different consequences for your main strategy as well as for the strategy of the other player. This allows you to make your decisions with your strategy in mind. A wider set of choices allows for more variation in strategies.

Decisions should offer different options with equal value. Dilemmas with a clear right or wrong solution are bad since there isn’t actually a decision to make there. A lot of strategy games with rock-paper-scissor elements cope with this problem. If your enemy throws scissors, the only sensible thing to do is throw rock. In this case there is clearly a best action to take regardless of your overall strategy.

In conclusion:

  • Remove as much motor skills and time-related skills as possible.
  • Remove any element of chance.
  • Remove intentional and unintentional obscurity.
  • Have many different options in every situation.
  • Make sure options have equal value.

Following these steps should, theoretically, result in the perfect strategy game. However, this doesn’t mean it’ll be a good game. It might not even be fun to play. There’s a reason there are very few games that check all these boxes. A game is much more than a good implementation of the genre. A lot of games play with elements of chance and obscurity because it makes the game more exciting. To be fair, strategy isn’t the most action-packed genre (especially turn-based strategy). Game developers tend to find ways to market their games to a broader audience who crave action and excitement.   

According to this list our game is far from a perfect strategy game, and that’s ok because we don’t want it to be. We’ve learned that implementing a lot of these “rules” results in a good strategy game. When your design clashes with “correct” implementation of a genre you should adjust the genre to your game, because at the end of the day it’s more important to have a good game than a right game. 

Related Jobs

Nuclear Division
Nuclear Division — Sherman Oaks, California, United States

Senior Game Designer
Rumble Entertainment, Inc.
Rumble Entertainment, Inc. — San Mateo, California, United States

Lead Designer/Creative Director
Nintendo of America Inc.
Nintendo of America Inc. — Redmond, Washington, United States

Visual Development Artist
Filament Games LLC
Filament Games LLC — Madison, Wisconsin, United States

Web & Interaction Designer


Ben Sly
profile image
I'm currently working on a turn-based tactics game with some rather similar design goals. Were I to make a similar list of goals, I'd keep the first three verbatim but I'd fold the last two into "Make sure there is a number of distinct and viable options but keep the number of nonviable options low." The whole point of strategy is that options *don't* have equal value - if they did, it wouldn't matter which one you pick - but have similar enough value that you need to carefully think through each option instead of dismissing it outright. Making sure that each option is distinct from the other options is also important, as it helps the player differentiate the options and provides greater diversity of play; you can, say, give the player twenty different ammunition types for each type of gun each with a minor bonus and drawback to pick from each time he reloads a weapon, the game would have roughly equally valuable choices but the gameplay wouldn't benefit much in depth nor would the complexity added be worth it.

I really wouldn't call this list of goals criteria for a "perfect strategy game" either. It's designing towards a specific game philosophy the likes of which chess, Go and German board games tend to be much closer to than video games do. It tests your ability to think through the game mechanics and both predict and react to your opponent(s) in the purest form possible, but it precludes a number of other elements which can also be strategic as well. Incorporating chance doesn't remove strategy, it shifts the tenor of the game by changing the optimal decisions into probability calculations and allows perfectly played games to still result in losses; planning around chance is its own form of strategy. Similar things can be said about obscurity. I personally don't particularly enjoy those two elements, but claiming that they don't exist in a perfect strategy game implies that they preclude strategy. This game philosophy still is a coherent one and worthy of a name, but I'd much prefer one with less judgmental implications - perfect-prediction strategy games, maybe.

There is also a strong downside associated with the philosophy as well. Having no random nor any hidden information means that memorization becomes very effective, and that's usually a bad thing because the game becomes about regurgitating a textbook of plays instead of thinking through this current game; see the chess's openings for a prime example here. Once the best plays are calculated, then games become stale as they just turn into "can you remember the best choice?"

There are several tools to address this issue. Within the purview of "the perfect strategy game" I can think of variable starting conditions, asymmetric play, and making the opponent's actions affect your own; using these well makes rote memorization much, much harder to pull off. Combatting rote memorization isn't just an issue that affects perfect strategy games, but since three of the other elements useful to doing so - chance, obscurity, reaction time - are marginalized in this philosophy, it becomes significantly harder to achieve.

In my game, I've been using highly variable combat maps for the variable starting conditions, a deep pre-battle point-buy system for customizing the forces a player controls to secure the asymmetric play, and a host of mechanics - terrain destruction and manipulation, positioning, the standard combat actions and plenty more - to ensure that the best choices are strongly contingent on what your opponent does. It still needs a lot of playtesting, but it's looking promising.

Simon Naus
profile image
What I meant with choices with equal value is that there should be a best choice regarding your strategy, not a best choice in general. Sacrificing your units might be a bad idea if you need them for an attack, but if you want to distract your enemy it might be a good plan. It's all about context here.

About perfection, I believe that there can't be any contamination because that would, by definition, make it less than perfect.

A good resolution for rote memorization, besides the solutions you offered, would be to randomly generate the terrain (yet symmetrical). There are more resolutions for this problem but the problem is mainly with the examples you've given. Chess, for example, only allows you to move one piece per turn.

I'm very interested in your game though! Could you send me more about it?

Curtiss Murphy
profile image
Have you read Paradox of Choice? Civilization, which defines the turn-based-strategy genre, deals with the paradox of choice by limiting the number of decisions/options a player deals with at any one time. It starts with a simple - do I want to start my city here, and slowly, one simple decision at a time, evolves organically into a game that is extremely complex - and yet, never feels overwhelming.

Simon Naus
profile image
I haven't actually, thanks for the tip!

Nathan Mates
profile image
My peeve with strategy games is that a bad choice upfront can cripple your game 20-30 minutes later (RTS, turnbased plays out at its own speed). RTS games tend to have a critical build order players MUST follow for optimal economy. If you deviate from that, you're being outproduced later, and you don't know why. It's just opaque to the user why they didn't do well -- until they basically have to read GameFaqs/strategy guides/youtube to figure out what should be done.

Simon Naus
profile image
That's something I really detest as well. I started looking at actual tutorials to become good at this. The biggest problem with this is that it becomes a requirement to actually make a chance in online matches and it's really hard to figure out how the other player does it due to fog of war. We have resolved this problem by abstracting resources to become more clear. We also have one unit responsible for resources, this removes the dilemma of balancing the amount of gatherers with other options to spend resources on. It also helps that our game has really short matches so losing doesn't feel that bad if it takes only 5 minutes or less.

Garret Bright
profile image
Two of your subsections, Chance and Obscurity, reminded me of a book I read recently called Uncertainty in Games. I recommend it to anyone who read those parts and wanted to dive a bit deeper in the theory behind them. It's a quick read and its only $10, and it's really good. On Amazon:

Simon Naus
profile image
Thank you! Will read it.