Arcs in Strategy Games

It is common to hear players talk about “tactics” and “strategy” in games. In this case, the colloquial understanding of these terms happens to be pretty useful, in that it maps well to something that actually goes on in playing strategy games. With that said, it’s worth taking a moment to clarify these terms:

Tactics” usually refers to “short-term decision-making”. Questions like “should I move this character two steps forward, or three steps forward” are questions of tactics. Tactics are micro-level decisions in strategy game play.

Strategy” usually refers to “longer-term decision-making”. Questions like “should I be aggressive early, or be defensive now and attack later on” are longer-scale choices about a game that players make. Strategies are macro-level decisions in strategy game play.

In both cases, we are talking about a grouping of gamestate information over time and how it changes. I refer to this grouping as an “arc”. Continue reading “Arcs in Strategy Games”

CGD Podcast Episode 17: Rogue-likes and Other “Bar Games”


I don’t mean “bar” as in “pub”, I mean it as in like a resource bar. In this episode, I talk about Rogue-like games in detail, why it isn’t really a genre, and what the future of these games are.

What can we do with single-player strategy games? Must they all be “managing highly random resources”? I think we should question our reliance on output randomness and heavily variant input randomness (such as map generation in Civilization) to make single-player strategy games work.

It turns out that my two articles I wrote on score in the past were really outdated and they’re in the shop to be worked on. In the meantime, I recommend reading these for more thoughts on why the traditional high score system is a problem (which I claim in the episode but don’t really back up).

“Why I Hate The Term Permadeath”

Episode 12 of this podcast, which talks about good goals

Auro’s Single Player Elo System

A Roguelike Radio episode that I’m featured on that discusses score

A reddit “Weekly Game Design” that touches on it

Thanks for listening!


Support the show here on Patreon!

CGD Podcast Episode 16: A Conversation About Rogue-likes with Black Shell Games


I had a conversation with the main developers at Black Shell Games – Daniel Doan, Raghav Mathur, and Thomas Espinoza – on the topic of Rogue-likes, a design pattern that has appealed to both of us. In the episode, we cover topics like the relationship between Rogue-likes and gambling, grinding, difficulty, replay value, and other related concepts.

Check out Black Shell Games on Steam. Here’s Black Shell Games’ – which also acts as a publisher – complete list of games.

(Side note: apologies about the audio. There were four people on the conversation total and there’s a bit of noise, but I think it’s easy enough to hear.)


Videogames Are Broken Toys


“Whoops, I have the antidote for the Kragle, how did that happen?”

– Lord Business

At best, my design theory work tends to get mixed reviews. It’s my belief that if any creative work fails to connect with people, it’s the fault of the communicator. As a writer, you’re setting out to connect with people, so if you fail to do that – regardless of your reasons/explanations – it’s your fault. For the past few years, one of the most significant factors in this disconnect I’ve experienced has – to some extent, at least consciously – eluded me.

I’ve always known that many videogames are primarily toys (obvious examples being Minecraft or Grand Theft Auto), but what I failed to realize is that nearly every videogame ever made is essentially functioning at the “toy level” (which I’ll explain in a moment). I failed to realize that even supposed “strategy games” like Starcraft, Civilization, X-Com or Street Fighter are all primarily toys, despite the fact that they have many markedly game-like features and even market themselves as strategy games.

This caused me to raise concerns that failed to connect with people, because I was looking at them as games, whereas everyone else was looking at them as something much more like toys.

In addition, my criticism was actually backwards. I was complaining about the elements that damaged the game, when I should have been complaining about the elements that damaged the toy.

(In case you’re confused, here’s a quick primer on my four prescriptive interactive forms (read more here): the base interactive system with no goals is a “toy“. Add an objective/solution and you get a puzzle. Add measurement and you get a contest. Obfuscate game information (allowing for decisions to exist), and you get a game (a contest of decision-making).)


The Arguments

A huge point of contention has been my position on traditional videogame-style asymmetry. In 2013 I wrote an article called Debunking Asymmetry, which explained the problems I have with asymmetrical forces in games (i.e. selectable races, characters, classes, etc).

The article got a lot of attention from some well-known designers like Jon Schafer, Greg Costikyan, Raph Koster, and David Sirlin (who wrote a dismissive and mean-spirited comment which got upvoted like crazy, despite the fact that it didn’t really address my article’s points).

Overall, though, the consensus seemed to be more or less something along the lines of “you can’t say anything is good or bad”, something that I hear all the time and always found really irritating, and attributed to people being perhaps “anti-progress”. This comment from Shay Peirce somewhat embodies the sentiment as well. Here are a couple of excerpts from his comment:

“There’s no such thing as “correct” or “ideal” game design; there’s just rules of thumb, which may or may not apply to solving the problems of the specific game you’re designing, i.e. the specific experience you’re trying to give people.”

“Finally I’d like to say that your dismissal of the value of self-expression through gameplay style is, to me, breathtaking in its audacity and obtuseness.”

When I read these comments, I was really kind of confused. One of my concerns I raised in the article was what I called “playing designer”. Since the objective in, say, Street Fighter, is to win the fight, then I should choose the best character (or at least, the character I am best with) every single time. In either case, the actual “choice” of who to pick is therefore non-existent/solved.

Only if you decide to sort of step back and ask, “well, would it maybe be more fun for me to choose a character other than the best / that I am best with?” do any of the other choices become viable. In that situation, you are making a design choice – what are optimal rules for this coming match in terms of my experience, which is markedly different from strategic choice in a game, which is always about finding more optimal moves.

I brought this up, but it didn’t connect. Why not? Well, because everyone else was already assuming that of course we all make design choices. We make design choices all of the time in videogames – we’ve grown up making design choices. Early popular videogames like Super Mario Bros., The Legend of Zelda and Final Fantasy all trained us to constantly make design choices – to make sub-optimal strategic choices for the sake of keeping ourselves entertained. We’re used to making sub-par moves for the sake of “fun”. Sometimes we refer to this as “going for style points”.

Using four fighters in FF1 is incredibly strong, and incredibly boring.
Using four fighters in FF1 is both incredibly strong, and incredibly boring.


Further, we even have a derisive term for people who actually try to play optimally: “min-maxing“. Min-maxing refers to the act of picking the best stuff, especially in a role-playing game such as Dungeons & Dragons. D&D is inherently a role-playing toy, so if you’re “playing to win”, you’re playing wrong.


The “Toy Level”

Most game players and game designers are not thinking about “games” – even “strategy games” – the way I do. They are thinking about them fundamentally as toys, and since toys are so loose and formless, it’s much harder to make clear statements about what works or doesn’t work. I only know of one sure-fire way to fundamentally break toys, which I will get into in a moment.

Another example is randomness, and my positions on that. If you are thinking of something as a strategy game, hard output randomness isn’t desirable. But in a toy? Sure, why not! Well-known designer and theorist DanC once told me that I was (paraphrasing here) “obsessed with attributing value to tools”. His point was that randomness is a tool, asymmetry is a tool; everything is a tool and, echoing Shay Pierce’s sentiment, you might want any of these things in your system.

If you are operating on “the toy level” – which basically every designer on earth with the possible exception of some Eurogame / abstract designers – this is true.


The Problem

eheheI really enjoyed The Lego Movie. The theme, or controlling idea of the movie was pretty powerful: that the spirit of Legos was creativity, exploration, doing things your own way. The primary force of antagonism in the film was essentially instructions, or the super-glue that holds Lego bricks in place once set. In the first act, the main character lives his life by a strict set of rules, and throughout the story he learns to embrace his creativity. Wild, loose creativity is, after all, at the core of the spirit of legos, and of toys in general.

As I said earlier, I only know one sure-fire way to break a toy. Toys are very low on the ladder of forms, very simple, and thus, very durable. Just about anything you can imagine can work in a toy, except for one thing.

Toys break when you give them a goal.

When you give a toy a goal, it technically becomes some other form – a puzzle, a contest, or a game. The problem is that the loose, scattered bits of rules everywhere that could have made for a good toy is not strong enough to support any of those other higher forms, all of which require way more systemic support to operate properly.

Things end up hobbling along and working, but only because of the efforts of the player to ignore your goal entirely (and simply play with it like a toy), or worse, partially ignore the goal, which requires a lot of mental effort.

This might be OK for the opening introduction to your toy – i.e. the instructions that come with legos – but you should quickly abandon any sense of a real goal as soon as possible.

Some systems are very obviously toys with some goals slapped onto them. Probably the most obvious examples of this would be things like Grand Theft Auto and The Legend of Zelda. In these systems, we have a great big open world with all kinds of stuff to explore and things to do, which is then flattened out into a linear, rote chore by the missions and fetch quests. In GTA, I can go off huge ramps and drive around on rooftops and explode and jump out and parachute into the water – all kinds of fun “play” activities. Nothing takes the fun out of those like a big yellow arrow telling you exactly what to do.

RPGs, too, also have this quality. I always remember being frustrated, even as a kid, that so many weird, interesting, almost tactical spells that technically existed in Final Fantasy took a direct back-seat to something like FIR2, which just nukes them with damage. I mean, the point of Final Fantasy is to get to the end, right? To “beat the game”? We certainly don’t want to die and be “set back” further from the goal, so we should just use the damn fire spell, not muck around with other stuff that might be more interesting.

At the other end of the spectrum, though, you have supposed “strategy games” like Civilization. But now I have to do a weird mental gymnastics routine where I am trying to win, but I’m not trying to win too hard. In Civilization, I should probably look up which of the six victory conditions is the most consistently easy to win and with which nationality, and then just pick that nationality every single time and do nothing else, ever.

It’s funny – that sounds like such an alien concept, doesn’t it? I’ve played thousands of hours of various Civilization games and I did nothing of the sort. I always wanted to “try out” all the different factions and “go for” all the different victory conditions to see. What I was doing was not “trying to win”. I was “exploring the edges”, as one does with a ball or with Legos.

Even with highly competitive games like League of Legends, the reason they have so many characters and so many items is so that you can constantly “explore edges”. It only hurts a strategy game to have that much content, in that actual balance goes way out the window. Someday, when Riot stops issuing weekly balance patches (which makes solution a moving target), the community will quickly settle on the top few team builds, and soon after that, the game will die off, with some other “less-explored toy” taking its place.



When I say that videogames are broken toys, of course they aren’t so broken that they can’t be enjoyed (at least, not the popular/successful ones). Players inherently do a lot of the lifting themselves with toys, so going the extra mile or two to ignore / partially ignore goals isn’t that hard to do, and basically everyone is doing it right now and will continue doing it for some time. That doesn’t mean that we shouldn’t improve the situation, though.

I think it behooves us all – myself in particular – to understand that everyone else is thinking about things on the toy level, even for highly competitive games. So much of my criticism has kind of been invalid. When I complained about CCGs, or too many characters in Street Fighter, I was wrong. What I should have said was, “these things shouldn’t have goals.”

We also need to get over our anti-toy bias. Players, particularly adults, are a bit ashamed of playing with toys, and I think that needs to change. There’s nothing wrong with playing with toys.

Some might be wondering “how would you do something like Magic: The Gathering without a goal?” There’s a lot of answers to that. I believe there’s a variant of Magic: The Gathering that automates the actual “match” aspect. This might be a start; basically it’s a “deck-building toy” – you build a deck, and instantly get a number or some other feedback that allows you to explore the edges of “all Magic cards”.

The immediate response to that idea is likely, “well, that wouldn’t be fun for very long”. I agree – the problem is that Magic isn’t really that great a toy. The designers themselves weren’t entirely thinking of it as one; if they had, it’d probably have way more edges to explore. Forget completely about balance or depth or dominant strategies or any of the things that one worries about when building a strategy game.

Because toys are a primitive and simple form, it’s not hard to create a functional toy. However, creating a great toy – such as Legos – is pretty hard. Like anything else, you need some “core idea” that ties the entire thing together. Toys can – and should – be elegant, just like any other human creation.

The modern player has been raised in a world of toys. Pure strategy games, like some designer boardgames and my own Auro, can come off to today’s player as unforgiving, difficult, strange or even “feeling like work”. In time, when we have more examples of pure games, I expect this problem to diminish, but for now, I think it’s an issue.

The next thing I make with my development team, Dinofarm Games may end up being a pure, unconflicted toy. Toys are the most common language people speak these days, and part of being a good communicator is speaking to people in a language that they understand.

If we want to make toys, we need to really focus on that. Or alternatively, focus on the other forms. Portal and Professor Layton do a great job of focusing on being puzzles. Guitar Hero and Dance Dance Revolution are great, focused contests. My favorite form by far are games, so I would love it if we could start producing some unconflicted, pure strategy games as well. Outwitters is probably the closest thing to a pure, unconflicted digital game, but even it has asymmetrical forces.

I hope that in the future, developers take the point of The Lego Movie to heart. Stop spraying glue all over your toys.



Thanks to my Patreon patrons for their support! Click here to become a supporter.

EMPIRE is out!

Yesterday EMPIRE, a game I’ve been designing for Crazy Monkey Studios for the past 8 months or so, was released.  It was a pretty quick turnaround – after working on 100 Rogues, which took about a year and a half, and AURO, which so far has taken nearly two years, it’s nice to design a game, make it, and put it out there.

It was a pretty different experience, process-wise.  Unlike AURO, I really knew what EMPIRE was going to be at the get-go.  The game that it ultimately was released as was very close to what I had in mind from the start – especially combat, which I feel is very strong in version 1.0.


Despite the fact that I’m really happy with where EMPIRE is now, I feel that version 1.0 is just the beginning.  I have a lot of other plans for the game in future updates.  For instance, I’d like to change the way that monsters work on the overmap.  It would probably be good if monsters had the bases that they have now, but also sent out troops which milled about randomly until they came in contact with the player’s city.  That way, there’s a bit more life/emergence to the monsters, and it also makes the whole “I target you, you target me” thing – which is kind of strange at the moment – less of a problem. Monster cities would never “attack” you, only launch wandering monsters.  Monsters themselves would attack, but simply by walking onto your city.  Since you and the monsters are already very asymmetrical, it makes sense that the way they attack would be different than the way you do.

Another thing that I think the game might need is some third resource – perhaps “gems”, or perhaps “settlements” – that you can see through the fog at different locations on the map.  These would be required for certain tech things (such as perhaps Shaman’s Huts), but also finite, and could be wiped out by wandering monsters.  This would give exploration a much-needed boost in its coherence as a mechanism.

Anyway, overall I’m so excited about having another game I designed out there.  I can’t wait to hear what people think of the game.  If you know anyone who wants to review the game, send me an email and I can probably get you a promo code.

Introducing: EMPIRE

StyleSketch1Today I wanted to introduce people to a new game that I’m designing called EMPIRE.  In order to do this, I think it makes sense to first talk about what already exists, and then talk about what I’m doing that’s different.

EMPIRE is my take on the so-called “4X Strategy” genre of digital games.  I’ve always been a fan of games like Civilization, and even more so of Master of Magic.  I do have a number of problems with the genre, problems which have not been getting better.  For instance, Civilization V, the latest game in the Civilization series, did not correct most of the games worst problems.  You can read about my problems with that game, which are fairly similar to my problems with just about every game in the genre, here.

Suffice it to say that with EMPIRE, I have an opportunity to do what I did for 4X games what AURO does for roguelikes:  namely, find some kernel of an actual core gameplay mechanism, and build a carefully constructed system around that.  So unlike most videogames, this game will be system-based, not component-based.

Why does that actually matter?  Well, because it means that we can have an elegant design, which in turn means that we can have a system that’s both extremely easy to learn, and equally difficult to master.

In short, EMPIRE is a modern, elegant solution to the problems of 4X strategy games.



EMPIRE is a game centered around the concept of maintaining a growing set of resources.

I often start with some thematic metaphor to help me in designing a game, and with EMPIRE, that metaphor was one of “the rise and fall of an empire”.  I think it’s very interesting and dynamic how a real life empire can grow more and more powerful, but sort of break under the pressure of its own weight after awhile.

I’ve also been playing a lot of Puzzle Strike, and before that, Dominion, and I feel that the “deck-building” mechanism is a fantastic way to express that.  So, the “set of resources” that you’re maintaining in EMPIRE are digital “cards” that you use in battle and win from victories.

So, in a sense, EMPIRE is the world’s first Deck-Building 4X Strategy game!

Right now, the game is in an early alpha stage, so you should expect some of what’s written here to change in the coming months as more playtesting begins.  Also, keep in mind that all screenshots and such are very early – excuse the temporary buttons and such!


Working Alpha prototype.  Forgive the temporary art and menus!
Working Alpha prototype. Forgive the temporary art and menus!


EMPIRE In Detail

EMPIRE is not only built to avoid the pitfalls of traditional 4X strategy games, but it’s also built primarily for mobile, and the game is being designed around this.  I’ll explain some of the rules to show you how it works.

Essentially, EMPIRE is a war-game.  This puts it in stark contrast with most other 4X games which have a more toy-like “do whatever you want” feeling to them much of the time.  In this game, you are trying to conquer other civilizations in a constant need to take new territory.  Eventually, your civilization will fall, so it is a matter of surviving for as long as you can and winning as many battles as you can to achieve the highest score possible.  To return to the metaphor, you could say that this reflects entering your empire into the textbooks of history as one of the world’s greatest.

When you start the game, you have enough resources to found one city.  When you do create a city, that city starts sucking up resources from the surrounding tiles each turn.  Eventually – and this is one of the most different things about the game – those tiles will produce fewer and fewer resources, until they finally become “desolation” tiles:  scars on the earth that not only produce no resources at all, but actually spawn dangerous monsters.

So, this means that you must stay on the move to keep a steady flow of resources coming in.  And if you don’t keep that steady flow of resources coming in, and a nearby Empire does, well, then you can guess that he’ll likely overpower you.  So, there’s a natural struggle for new, un-desolated territory.



I started with the question, “what are cities, really, in a game like this?”  If we can identify that at its core, this is a war game, then cities are a stepping stone towards creating your army.  With this understanding, we can realize that the system for cities is not central, and should be limited in its complexity.

The system for how cities work is extremely simple, yet still has enough resolution to support expressive gameplay choices.  A city is taking in “food” from nearby tiles, and when it reaches a certain threshold, it “levels up”.  When it does this, you can choose between a choice of 2 buildings.  Once you choose one, that choice is permanent.  You can’t go build the other building now.  Eventually, you’ll level up again, and now you get another choice of 2 different buildings.  This is “Tier 2”, and there are 3 such tiers.

So, you basically have 3 rounds of choices to make, which leads to somewhere around a dozen or so possibilities for the city’s configuration (someone else can do the math for this and let me know the exact number!).

Of course, you can also have more than one city.  We’re currently working with a system where the maximum number of cities is 3, but even to have 3 is difficult.  So, having 1 city is tax-free, but when you have a 2nd city, there’s a decently harsh tax on all income.  So, if you’d normally be sucking up 10 food a turn, now it’s reduced to 8 food a turn, or something.  Which might be totally fine while the surrounding resources are good and healthy, but makes the desolation tiles even more of a threat.  3 cities is almost never sustainable for very long due to a significant tax that’s imposed.  If you have 3 cities, you need to either be constantly winning battles (winning some battles can yield some resources) or just moving quickly to new areas (this would probably require winning battles anyway!).

The primary role of cities is to suck up resources from the land, produce new military units, and produce new Action Cards (which I’ll get to in a second).


Ignore the text and HUD stuff, it's all temporary!
Ignore the text and HUD stuff, it’s all temporary!



This is probably the biggest area that “it being a mobile game” helped influence the design, but honestly, mobile design is good design, in a way.  What I mean is, you never want a game to be super fiddly UI-madness;  you always want interacting with the game to be as simple as it possibly can.

With armies and units, one thing I wanted to do away with was the concept of “moving units around from city to city”.  It’s extremely fiddley, and even when you have a mouse it’s just annoying.  Grouping units together, waiting for that last swordsman you just produced to walk allll the way over the map to get to the rest of the group, etc.  I didn’t want to deal with any of that.

So in this game, your army is ever-present.  It’s like a resource.  If you attack something, you have your whole army.  If you’re attacked at any of your cities, your whole army defends.  Making an attack on something takes time, by the way – if you want to attack an enemy city, that city is alerted to it, and it takes a certain amount of time (this amount calculated by distance, the terrain covered, and how many Mounted type units are in your army).

This way, we can avoid any fiddling.  It should make for a really pleasing, easy to use, yet still super strategic experience.



It may be surprising to know that cities are not central to this game.  Armies are also not central.  Armies, too, are merely a resource that is used in combat.  So what is central to EMPIRE?

The EMPIRE Action Deck.  In the game, you start with a deck of about 10 Action Cards.  These are used in combat to give your troops commands.   One of them might say “all archers advance”, one might say “soldiers fortify”, or “all units retreat”.  Some of them have special effects, like making one unit invulnerable for a turn, or even summoning monsters.

When a battle begins, the game draws a number of these that you may use on your turn. First, your troops advance on their own, and then you may use an Action Card from your “hand”.  Then, any combat that is possible happens and is resolved.  It’s a really simple system that’s still highly tactical and interesting.  I made a paper mockup of the combat which worked really well.

Here’s where it gets really interesting, though.  Winning fights is of course, the objective of the game.  And when you do, you gain points.  But here’s the rub:  those points are given to you in the form of a Victory Point card, which goes into your action deck.  Players of Dominion are very familiar with what I’m talking about right now – what this means is that your deck is slowly getting clogged up with more and more useless cards. 

Late game, you may have added lots of fantastic, magical, powerful cards to your deck… but you’ve also likely won a good number of battles, and have a good number of Victory Point cards.  This means that on a given turn, it’s increasingly likely that you might just draw 2 or 3 VP cards, severely limiting your combat options!

Old empires can get Action Cards that re-draw hands to help mitigate this, build buildings to increase hand size, and even sometimes trash some old unneeded cards, but these all come at a cost.

Combat is fought until 3 units are killed on either side, OR the base-line of a side takes 5 damage (it can be attacked).


Obviously, this is just a conceptual mockup - not artwork at all!
Obviously, this is just a conceptual mockup – not artwork at all!


Throughout the game, not only will young new Empires spring up to try to take power away from your old, mighty empire, but Monsters are also spawning with increasing frequency.  It’s always a dangerous, unsafe world in Empire.  When your last city is destroyed, the game ends, and your score is tallied.



I should mention that EMPIRE is not a Dinofarm Games game.  Instead, I’m working with a new team, as lead designer.  The lead artist for this team is a guy named Martijn Holtkamp, who has Age of Wonders 2, Age of Wonders: Shadow Magic, and Divinity 2 on his resume.

Concept art by Martijn!
Concept art by Martijn!

The theme for the game is loosely inspired by Conan the Barbarian illustrations by Frank Frazetta and other similar artists, but also blended with a touch of stylized cartoonyness.  This is all really Martijn’s domain, which maybe I’ll get him to write more about in a future post, but for now suffice it to say that I think the game will look unique.

We expect the game to be out Fall of 2013.  As for platforms, the game is being created in Unity, so whereever Unity can exist, EMPIRE can exist.  Certainly iOS, Android, Windows, OSX, just to name a few.

So that’s EMPIRE!  What do you think?  I’d love some feedback on the idea, and I’ve created a forum over at Dinofarm Games to talk about it.  Eventually, we’ll likely start a beta phase, and if you subscribe over at Dinofarm, you’ll be the first to know about it.  Thanks for reading!