A Question (Or Three) Of Strategy

I sat down for my first session of Blizzard’s MOBA Heroes of the Storm. I’d watched some Let’s Plays and completed the tutorial. I’d read up on my hero of choice and studied the game mechanics and map objectives. Then, the match started. The gates opened…

I realized I had no idea what to do.

That’s because in any good MOBA (Multiplayer Online Battle Arena), game mechanics are only the tip of the iceberg. What lies under the surface, invisible unless you know to look for it, is far greater depth of gameplay. In Heroes, as in many games, great gameplay lies in making judgment calls about where to invest your most precious resource: your agency as a player. In this way, it is a lot like more traditional strategy games in that it involves, well, strategy.

My strategy still consists of “Oh god oh god oh god WTF am I supposed to do?”

“Strategy” can be a loaded term and hard to define, so I’m going to introduce the word “heuristics” into the conversation. Heuristics, referred to as “game sense” in some circles, refers to the part of the iceberg that’s below sea level. It’s the ability to – with incomplete information – judge a game state and make an appropriate or optimal decision that improves one’s own position. It always, by definition, requires some kind of strategic decision, prioritization, weighing of alternatives, or give-and-take compromise by the player. A player’s heuristics sense is what fires up every time they make a strategic or tactical decision about what to do next. 

All of those critical decisions that we encounter in the course of a game, I believe, can be distilled down into the following questions:


  1. What am I doing right now?
  2. Why am I doing it?
  3. And what am I NOT doing because of that choice?”


“Who are you?” and “What do you want?” rank a close number four and five in importance

To me, how these questions are answered is the mark of a fundamentally solid strategy game. In fact, I’m going to posit that the strength of any strategy game’s mechanics can be measured by the answers they provoke to these prime questions.

But before we dive in, I want to clarify the assumptions I’m working under:

  • Since we’re talking about strategy games, having more meaningful choices is better than having fewer
  • Meaningful or strategic decisions are structured around “hard choices” where you do option A or option B, but not both
  • Although not every decision needs to demand in-depth analysis, I’ll take one meaningful choice over 100 bland ones
  • Choices don’t need to be complicated to be meaningful
  • There’s room for a wide range of 4X and strategy games with a variety of levels of depth 

For example, a gameplay element that usually encourages strategic thinking is combat. It’s easy to see why this succeeds at engaging the player’s heuristics. Any unit can only be in one place at a time. It can only fire so many times, or engage so many (usually one) enemies per turn, and doing so often makes it more vulnerable or causes it to take damage. By pursuing one decision, you are inherently forsaking many other enticing avenues. Deciding how to best use a unit requires making a judgement call that hinges on interpreting the game state.

If every element of 4X were this awesome, I would be a happy man

Unfortunately, other 4X mechanics – like city management or research – often fail to engage the player on the same level. This eXposition will cover build orders, research, diplomacy and player agency. Then, we’ll examine how these hold up under the prime questions, and how tweaking the decision making process could result in better, more strategic, gameplay.

Case Study: Build Orders

One age-old gripe of strategy grognards is the build “treadmill.” Once you puzzle out the “best” order for a particular end result (e.g. a “research” colony) – or even once you get close – you almost never have any reason to deviate from it. At that point, the only thing separating a colony from having every building available is the tedium of clicking through the queue. How does this look through the lens of our prime questions?

What am I doing right now? Why am I doing it?

I’m building X, because it comes next in the optimal build order.

What am I NOT doing because of it?

I’m not building all the sub-optimal buildings.

I think most will agree that this isn’t terribly engaging or strategic. There’s little need for heuristics because the optimal combinations are small in number and static from game to game and colony to colony. Sometimes you’ll put your queue on pause to build other assets – reconnaissance, military, workers, colony ships, etc. – but this still doesn’t really address the core issue.

It’s no gold standard

There are many ways of fixing the problem, but here’s one of the simplest: change the question from “In what order do I get the buildings?” to “What buildings do I get at all?” In other words, a system for forcing trade offs, where a given colony may only be able to construct a subset of all the buildings. This introduces exclusivity while providing compelling, competing reasons for pursuing different options in different situations.

Games like Starbase Orion, Armada 2526, and the upcoming Stars in Shadow improve the colony build queue system by introducing building limits. Under this system, you have to carefully choose what you build and how you specialize your colonies. Galactic Civilizations III features tile-based yields and adjacency bonuses, both increasing variance and capping the number of buildings per planet. Civilization VI’s “unstacking the city” approach limits the number of Districts a city can produce while also requiring Wonders be placed on specific tiles (which then can no longer be worked by citizens). Other potential avenues include simply oversaturating the build queue so no colony can conceivably build every building – like in the popular Civ IV mod Caveman 2 Cosmos – creating designated city types à la Warlock II, or introducing punitive building maintenance that punishes cities that don’t specialize.

The granularity of Caveman 2 Cosmos is not for the feint of heart

As we pointed out in a recent eXposition, deterministic mechanics like solvable build queues have more in common with puzzles than true games. I’m of the opinion that there’s nothing inherently wrong with having puzzle mechanics in strategy titles. However, colony management and build queues are so central to 4X games that making these mechanics more strategic is a good way to add depth to the game – as long as you teach your AI how to play it!. This goes a long ways towards increasingly replayability and keeping players engaged and having fun – they’re actively participating, rather than simply following a fixed, optimal order.

Case Study: Research and Technology

The tech tree – the staple research system within 4X and strategy games – is another area that often falls short of providing engaging, strategic decision making. For example, in Civilization V, there are several tech “paths”, with each tech leading to the next and with many techs requiring multiple prerequisites. We’re oversimplifying for the sake of brevity, but once the player decides which one route to focus on, they’ll move through the tech tree in similar, optimal ways from game to game. This approach is pretty much the industry standard, although in Civ V’s case the dependencies are particularly stringent. So our answers to the prime questions are as follows:

What am I doing? Why am I doing it?

I am researching technology X, because [I want it] OR [it leads to something I want].

What am I NOT doing because of it?

I am not researching another technology (or down another technology path), which I can take later without penalty whenever I feel like it.

The simplicity of this approach makes it appealing, and, as such, it’s widely used. But it’s also weak for the same reasons the generic build queue is weak. Further, Civ V’s heavy technology prerequisites cut down the number of meaningful divergence points. Civ IV’s system, while also lacking in many of the same ways, is fundamentally stronger because, ironically, its relaxed prerequisites lead to more distinct strategies – there are more discrete tech “paths,” which in turn leads to more give-and-take between beelining specific techs vs. backfilling.

It’s not Path to Exile, but it’s something

Master of Orion II (and similarly StarDrive 2) doesn’t have a tech tree in the traditional sense. Instead, it sports eight different disciplines, each of which has ten levels. When you research a tech, you advance its field to the next level. But, the well-known kicker is that each level has up to four technologies, and with the exception of the first few basic techs , you can only research one of them (unless you’re one of the overpowered Creative species. Then all bets are off). Once you make that choice, you’re locked out of the other technologies unless you can trade for or steal it later. So while questions one and two are the same as in Civ V, the final answer is different:

What am I NOT doing because of it?

I am not getting technology Y and Z, possibly ever.

Wow – suddenly the stakes are a lot higher! This is by no means the best tech system that’s ever been tried, but it’s a big step up from generic, ho-hum research trees. Those interested in strong and unusual tech systems should give the original Master of Orion and the original Sword of the Stars a look.

Finally, let’s return to the Civilization franchise, this time with the oft- (and rightly-) maligned Beyond Earth. For all its faults, Civ:BE tried to push 4X research in a fresh, exciting direction. Instead of a tech tree, it features a tech web. Players start in the center and progress outwards in any direction, and can research any tech if they’ve researched one adjacent.

Here, just look at the picture. 1,000 words, and all that.

Terrible UI and awful balance aside, this is a huge step in the right direction

But the real innovation of BE is the branch and petal system. Each step along the tech progression ladder is a “branch” technology (the top item of each category), representing some kind of broader fundamental knowledge. Each branch then unlocks a number of “petal” technologies (the smaller items underneath each branch). Petal techs represent refinements and applications of the parent branch, and as such, do not advance the player further along the tech web. Those interested can read more on the branch/petal system here.

By combining the relatively free-flow nature of the tech web with the branch and petal dynamic, Firaxis breathed a lot of life into their research system. The answer to the second question (“Why am I doing this?”) will depend on whether we’re going for a branch – to progress further through the tech web and open up new technologies – or to further refine those disciplines we already know. And the answer to the third question will reflect the opportunity cost of not progressing somewhere else. Like the build queue example, this is still just a question of “In which order?” But at the same time, with the give-and-take between researching petals vs. branches plus the free-form nature of the web, the strategic divergence between the possible pathways is great enough that the player must rely on their heuristics.

Case Study: Diplomacy

Diplomacy is often the Achilles’ heel of 4X games. A recent eXplorminate poll found that diplomacy is the area our audience most wants to see overhauled in 4X games by a huge margin. And that’s not without reason; standard 4X diplomacy fails the prime question test spectacularly.

It’s hard to select a particular example, since 4X diplomacy has been mixing and matching the same half-baked mechanics across the genre for over two decades. So take your pick: Civilization, Planar Conquest, Age of Wonders, even vaunted titles like MoO II and Master of Magic.

Even Paradox, renowned for their diplomacy, dropped the ball on Stellaris

The principle problem with diplomacy in all of these games is that the limiting factor is simply “What will the AI let me do?” Is it willing to swap techs? Cool, let’s do that. Trade agreement? Sweet! Open borders? Sure, I’ll take better scouting and a free relations boost. Trade maps? Why not? The trend towards revealing all the little numbers that go into the AIs decision process makes the AIs behavior more transparent and predictable, for better or worse.

Let’s examine these games’ diplomatic exchanges under our prime questions.

What am I doing?

Whatever the AI lets me.

Why am I doing it?

Because it’s in my interests and I have no reason NOT to.

What am I NOT doing because of it?


In most of these cases, maximizing diplomatic output is a boring, often exploitable, case of just checking what every faction is willing to do on every turn of the game. Not edge-of-your-seat gameplay by any means. Just drudgery. Often, the best we can expect to see is being forced to pick between enemies or choose sides in an alliance.

Fortunately, there are a few games that break the mold. Recently, StarDrive 2 introduced a diplomatic tolerance system: a renewable resource that depletes with each deal you commit to. This limits how frequently you can bombard the AI with requests (and they you), but the fundamental nature remains the same. A few other games, like Endless Legend, similarly limit deals using Influence as a form of diplomatic currency. EL also features factions that can use diplomacy in a more assertive fashion, like the Drakken who can spend Influence to unilaterally force peace or alliance status. These approaches are perhaps a bit brute-force and not the most interesting, but they’re something.

However, a few games really do spice up the diplomatic game. Star Ruler 2 boasts delightful card-based diplomacy, in which factions can directly (and unilaterally) enact diplomatic actions against other factions by spending resources on diplomatic maneuvers. Crusader Kings II’s spectacular character-based diplomacy makes balancing individual relationships, alliances, marriages, factions, plots, etc. an actual, honest-to-goodness subgame. The classic strategy game King of Dragon Pass features an innovative system, in which receiving aid from another clan means they can show up on your doorstep at any point and demand you return the favor. On the indie front, diplomacy in the upcoming Lord of Rigel looks to be built largely around a Babylon-5-esque cold war between elder species, with the major factions coalescing into alliances on either side – or forsaking both.

Star Ruler 2 – a breath of fresh air in an otherwise suffocating tomb

Another approach is to allow your empire’s populations to react differentially to your diplomatic actions. Maybe you buy out of a war by ceding a few border planets to your enemies. Perhaps you’re a peace-loving empire that’s swapping weapons techs with the warlords down the starlane. Or say you launch a surprise war against your long-time allies. How does the public respond to those actions? While making popular opinion too powerful a force in diplomacy hampers player choice in one regard, it frames diplomacy a bigger trade-off question – which again requires players to refine their heuristics.

Which leads us nicely into…

Case Study: Player Agency

It is the rule much more than the exception that player characters in 4X games are agents of unprecedented power. We can direct the development of entire worlds at the push of a button, martial forces from all corners of creation, explore the furthest reaches of space, make peace, make war, and direct the engines of science – just to name a few. It’s the unwritten, almost unspoken, rule of the genre that the player has control of all of these things, and that their control is limitless.

Having ultimate control over everything sounds like a nice power trip, but unfortunately, it may not always the best for gameplay depth. In fact, complete, absolute power deprives the player of a potential heuristic layer, while also leading to one of the oldest and most frequently bemoaned problems with 4X games: micromanagement hell. When you can optimize every workflow down to the last minutia – and have no reason not to do so – you get the tidal wave of tedium that plagues so many titles. Here’s how we can think of this in terms of the three questions:

What am I doing?

I’m doing as many things as I can before the tedium overwhelms me.

Why am I doing it?

Because tedious micromanagement is what separates mediocre play from optimal play.

What am I NOT doing because of it?

Playing a strategy game – I’m playing an Excel doc.

Micromanagement in 4X isn’t a bad thing per se. But in too many titles, it stands in the way of the game’s primary focus: strategy. While the language of the above example is overdramatic, it illustrates a deep problem shared by many strategy games: at a point, optimization becomes simply making lots of tedious non-decisions (little thinking, lots of clicking) that don’t engage the player’s heuristics.

Speaking of overdramatic…

Most 4X games that try to address micromanagement hell do it either with a slick interface, allowing min-maxing to happen in as few clicks as possible (MoO II, Endless Legend), or by scaling back the mechanics or scope of the game and eliminating some of the optimization puzzles (Polaris Sector, Age of Wonders III). These can work, with varying degrees of success. But I’d put forward that there’s an alternate and more intriguing solution: impose limitations on the player’s agency.

This is pretty experimental territory for 4X, but it’s not new within strategy games by any means. Think about chess: the player’s agency is made all the more precious and vital specifically because you can only move one piece each turn. In fact, limiting player actions is a defining characteristic of board games – and how player actions are limited play a central role in building strategic depth.

We can find plenty of examples within computer gaming other genres, like Heroes of the Storm. That X-factor I talked about – the base of the iceberg that left me gawking at my screen in indecision – was my lack of knowledge of how to spend my limited agency as one little piece on a large board. Those limitations are even more visible in RTS games; actions per minute are essentially the mechanic through which a player’s agency over the game is channeled.

Not that I want my 4X games to turn into this. But god damn!

Within the 4X genre itself, there are a few intrepid games that have attempted to limit player agency. Master of Orion III was meant to have a mechanic called Imperial Focus – but it was scrapped late in development. We covered this in greater depth in our MoO III eXposition, but basically, it came down to government sectors generating Focus Points which players would spend for every action they take. This kind of limitation could transform optimization tasks into a strategic tradeoff. Do you meddle with a planet’s build queue, or do you instead spend those points shifting research priorities? Do you make sure to eliminate each point of production overflow, or do you transfer colonists to a fledgling world?

The Romance of the Three Kingdoms series has used this same mechanic to great effect for years, and the upcoming Alliance of the Sacred Suns (formerly Imperia) looks to feature a similar system as central mechanic. In King of Dragon Pass, you can take only two actions each season (e.g. game turn). Simply letting players do “anything but not everything” creates an exceptional and frankly thrilling strategic layer, while also crushing one of the age-old bugbears of 4X gaming – excessive micromanagement.

As a bonus, this kind of mechanic can also be used to help balance other game systems. For example, wide, sprawling empires may have to carefully ration their actions compared to tighter territories. Large fleets can be reigned in with an action point tax instead of poor band-aid fixes like MoO II’s Command Points or fleet stacking limits.

Another approach for regulating player agency is to better connect the will of the populace to strategic choices and the  game’s internal mechanisms. Typically, going against the population results in happiness-based penalties to settlement productivity, maybe with an (easily crushed and pretty inconsequential) rebellion if things got really bad. That doesn’t exactly require fluid compromising and strategizing – just managing another resource.

But more recently, games like Endless Space 2 and Alliance of the Sacred Suns have added popular opinion and internal politics. (Stellaris may capitalize on this as well, but it needs to improve its faction system first.) While the systems are implemented in different ways, they have the potential to create wonderful emergent gameplay – plus a strong simulationist/roleplay element – by simulating the response of the populace to external and internal events. Even better, these social and political currents can bestow different bonuses and penalties upon the player based on their actions. I’d love to see more games embrace this gameplay layer, as I’d say it’s one of the most underexplored elements in the 4X genre.

Don’t blame me! I voted for Ati Tan

Parting Thoughts

I believe that the three prime questions outlined here (and how they’re answered) can serve as a useful guide for creating meaningful gameplay. We can see examples of this not only in 4X and strategy computer games, but also board games, RTS, and even in MOBAs.

None of the ideas I’ve discussed here are cure-alls, nor should the entire 4X/Strategy genre adopt them wholesale. Further, I recognize that more depth means the game is more challenging to play well, which can cut out some of your potential audience and challenge AI programmers. My aim in writing this is simply to spark discussion and thought within the community and offer some new, fresh angles for consideration by developers and modders.

As a 4X and strategy gamer, I live for those moments where I get to exercise my heuristics. They’re what draw me in, fire up my mind’s reward circuitry, and keep me coming back turn after turn and day after day (and DLC after DLC, wink wink nudge nudge). True, games with more strategic decisions tend to have higher learning curves, but the hunger for that greater depth of strategy is real. There’s a whole legion of diehard strategy gamers champing at the bit to reward designers who satisfy that hunger.

4 thoughts on “A Question (Or Three) Of Strategy

    1. Yeup – I think Ben has a fantastically insightful article here. I’d go further and suggest everyone interested in strategy games read this as a basis for evaluating their own thoughts regarding strategy games. Excellent piece.


  1. FWIW – On the topic of player agency, I think boardgames do a better job focusing their gameplay around real strategic decisions precisely because player actions are limited. The core mechanism of most boardgames hinges exactly on what players are allowed to do on their turns. Worker placement, role selection, action point allowance, area activation, card play, etc.

    The above all mechanics that constrain player action and force players to make categorical trade-offs in what aspects of the game they invest their limited agency into. The games become focused around this central challenge which keeps the gameplay operating at a higher strategic level. Granted, these mechanisms are also used as a means of controlling things like downtime, look ahead, managing game length, and so on – but they have the significant benefit of deepening the gameplay at the same time.

    Eclipse (which we reviewed) is in some ways a realization of what MoO 3 tried to do. In Eclipse, you have a pool of influence, which is spent when taking an action or used permanently to control a star system. The more influence you’ve spent in a round, the greater the upkeep cost in money you have to spend.

    The elegance of Eclipse’s system is not only does it force trade-off decisions between where you invest your agency (i.e. on research or building ships or moving fleets) but it also creates tradeoff’s betwen growing tall vs. growing wide. Tall empires have less influence (and also less upkeep) locked into controlling systems, allowing them to perform relatively more actions at a cheaper price compared to a wide empire. This perfectly captures the notion of a tall/nimble/flexible empire vs. a large/slow/ridged empire. The entire game hinges on managing these two heuristic layers: what actions to invest in, and how big to grow.



Please log in using one of these methods to post your comment:

WordPress.com Logo

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

Google photo

You are commenting using your Google 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 )

Connecting to %s