Avalon

On November 1st, I begin my job as a designer on Magic: the Gathering.

This is, of course, immensely exciting, and I’m looking forward to working with other incredibly talented designers on a game that’s captured my imagination for 15 years. However, this time of joy and new beginnings is also a time of bittersweet parting, as it means that Phantom Knight Games will be ceasing operations indefinitely. I will no longer be publishing games or writing blog posts on my own for as long as I work for Wizards of the Coast, though I hope to keep the blog archives and print-on-demand games up.

There were a number of reasons I began Phantom Knight Games. Perhaps one of the largest was the influence of the game Sentinels of the Multiverse, which convinced me that designing board games was fun and within my abilities. (I write more about Sentinels and its influence on new designers here.) However, another large reason, and the one that kept me going despite the cost and trouble, was a burning internal need to create. I found in tabletop games what I wasn’t getting from playwriting: The ability to give myself fully to a creative pursuit without worrying about how hard it would be to get an audience to see my work. It nourished and fulfilled me throughout some of the hardest times of my life.

Without my work in the trenches at the most independent level of game design, I wouldn’t have developed what I needed to be a Magic designer. While of course I got better at making and playtesting games, I also gained a number of other things from the experience: A design philosophy based around excitement and emotional impact, a sense of confidence from people loving my work so much they’d pay for it, and, most of all, the burning conviction to keep honing my skills until they carried me into the sky and pierced the vault of Heaven.

While I am proud of the games I made, especially Stand Back, Citizen!, I always felt a sense of inadequacy in the back of my head. I’ve never been particularly good at marketing, and I knew that I had to either promote myself more and better – either to pitch to external publishers or launch a Kickstarter – or keep puttering around local cons, selling a few dozen games a year. At times, I felt afraid that this was all my work would ever amount to. Being chosen to work at one of the most prestigious tabletop game companies in the world on the basis of the skills I developed through making these games validates this work I did, and I finally feel without reservation that all the effort I poured into Phantom Knight Games on (k)nights and weekends was worth it.

To everyone who ever bought one of my games, playtested one of my prototypes, or read one of these blog posts: Thank you. I’m afraid I must go now, but I’ll carry your kindness in my heart for as long as I live.

Jeremy Geist

Designer, Wizards of the Coast

The Illusion Of Control, Part 1: The Ellen Degeneres Slot Machine

About a year ago I went to Reno on vacation. While there, I took a look at some of the slot machines that were located in the casino-hotel where I was staying and in the airport as well. One thing I noticed that every modern slot machine had in common was the complexity of their rulesets. Each machine had enough rules to take up three to five pages on a large screen, and used terminology that even I, a certified Brass: Lancashire owner, had trouble parsing due to lack of experience with the genre.

I eventually ended up spending $5 on an Ellen Degeneres Show-themed slot machine that had become something of an ironic fascination among my friends. The experience was quick and bewildering – lots of stuff was going on on the screen at once and I had a hard time figuring out anything beyond “the symbols line up sometimes, by the machine’s definition of ‘line up’.” It was an experience that demonstrated that you needed to spend significant time with gambling master Ellen Degeneres to understand her intricate system of rules.

I found this interesting because slot machines’ clientele skews heavily older, a demographic that’s significantly less willing to learn in-depth rules for a game. (Incidentally, younger gamblers are increasingly drawn to more “skill-based” slot variants; slot game designer Edvard Toth wrote a really interesting article about designing them.) This includes slot machines with broadly appealing themes like Sex And The City and the 2010 Sherlock Holmes movie. So why, if slots are trying to attract a wide range of people at all ages and experience levels, are the rules so dense and complicated?

Part of this is because the mathematical models behind different slots are genuinely different, particularly in how often they pay out and how their payouts are distributed. (This Gamasutra article by Timothy Ryan goes into that in more depth.) But I suspect another component of these complicated rules is providing players with, not actual control, but the illusion of control.

Slot machines can’t genuinely reward player skill because by definition they’re luck-based machines. But what they can do is provide, through rules, the illusion that your greater familiarity with the game will result in better payoffs. All these rules must mean that something you did, something perhaps as minor as choosing how many rows to play with, had a genuine impact on how the slot machine plays out, though of course we know it doesn’t really.

As it turns out, providing players with the illusion of control is a valuable concept in tabletop design as well. Variance is an immensely useful tool for creating unique game states and providing replay value, but players like to feel smart and in control and get frustrated when variance takes that away from them. So, like video slots, designers can find ways to disguise variance to make it more palatable to players. Here’s a few ways you can pull that off.

Rock-Paper-Scissors: Disguise Variance Through Player Input

One of the most tried and true methods of hiding variance is by making the player the random number generator. Take rock-paper-scissors, for example; even though the outcome of a normal game of RPS is mostly random, players feel like they have agency because they get total control of which sign to throw.

There are many, many games that use a simultaneous selection mechanic; this provides a healthy amount of variance, as the game states that result when players don’t know what their opponents will do are much more diverse than those that occur with total knowledge. It also has the beneficial side-effect of making player decisions easier because you have less open information that you need to take into account when making a decision. Despite Go Nuts For Donuts having a lot of moving parts, Gamewright (which mostly publishes games for kids) published it because having most gameplay be reliant on other players’ hidden actions made it almost as beneficial to throw out a number at random as it was to deeply consider the other players’ behaviors.

An interesting component of this illusion is that it becomes closer to real strategy the fewer players there are. Libertalia with three players really does provide a lot of opportunities to figure out what your opponents are going to play based on the pirates they’ve already played and the booty tokens they’ve accumulated; Libertalia with six is a clown fiesta where there’s so many moving parts it’s impossible to make a good guess. This gradient obscures how much of simultaneous selection is strategy and how much is luck.

Cockroach Poker: Disguise Variance As Bluffing

This one’s very clever, because it’s a system of logical strategic moves based off a faulty core assumption: That people can successfully tell if someone else is lying.

A great deal of scientific research has demonstrated this not to be true, but what’s important for us as game designers is that it feels right. When you successfully guess if someone’s lying about being a bad guy in Resistance or smuggle contraband goods in Sheriff of Nottingham, you don’t feel like you got lucky; you feel like you outsmarted the other player.

A special shout-out goes to the delightful Cockroach Poker, a game built almost entirely on the illusion that you know whether your friends are bluffing. (A quick primer: The game has eight suits of cards. In turn, players place a card face-down in front of another player and claims what suit it is. That player has to guess whether the claim is true or false.)

This illusion is built in two ways. The first is that the game allows players to pass cards to each other and restate what the card is. Even though this only serves to flip the coin again, it gives the illusion of more evidence to determine whether a player is bluffing, as you can read the bluffs of multiple players. You can also team up with opponents to confuse a “target” player with a chain of bluffs.

The second is through weighted decisions. When a player incorrectly calls a bluff (or someone bluffing gets caught), that player puts the card face-up in front of them; whoever collects four cards of the same suit loses and the game ends. This means that when you put a card in front of a player with, say, three stink bugs, and claim that it’s a fourth stink bug, that player is taking a much larger risk by doubting the claim, which would make them lose the game.

The thing is, however, this secretly makes Cockroach Poker a press-your-luck game more than a bluffing game. Instead of asking “is my opponent trying to intimidate me?”, subconsciously, Johnny Three-Stinkbugs has to decide “is the off-chance this card really is a stinkbug worth taking the risk of doubting it?” Then, when Johnny doubts it and is proven right, he gets both the thrill of gambling and the feeling that he’s a deductive genius.

To Be Continued…

This article ended up getting really long, and I have more methods of creating the illusion of control, so I’ll be posting a followup in two weeks. Look forward to it!

(Read Part 2)

The One About Expansions

As of now, I have not been lucky enough to experience demand for a game that pushes me to create an expansion. However, I’ve certainly played many games with expansions, and the idea of releasing expansions for successful games is universal enough that I decided to put effort into thinking about what makes some expansions better than others.

Cognitive Load

Although I don’t know the exact statistics, I think it can be mostly assumed that most gaming groups are not the exact same group of people. Many are public meetups, clubs at universities or companies, or events at game stores; even with private groups, if they grow large enough, it’s unlikely that the exact same group of players will play a game more than once.

This means that a substantial percentage of people playing your expansion will have to be taught everything from scratch, which decreases the chances the game will make it to the table. In general, the more rules that have to be explained before a game begins, the less often that game will get played. If the game owner feels like the expansion makes it too cumbersome to explain repeatedly, the game is going to collect dust on their IKEA Kallax. (You could argue that since the owner has already bought the game, it doesn’t matter, but I’m assuming most game designers want people to enjoy their game, not just give them money.)

Thus, the goal for an expansion is to increase the possibilities of play while limiting cognitive load as much as possible. There are a number of ways to accomplish this, and I’ve detailed some of the common ones below.

Limit the number of components that show up at once

Some games have modular components that can be swapped in and out from game to game; expansions for these can just add more components without increasing the number players have to know at a time. As an example, Kingdom Builder‘s board is made of four interlocking mini-boards, each of which adds a special power to the game; the base game came with eight boards and expansions added more. Players only have to know the four powers on the boards they’re currently playing on, no matter how many expansions the game owner has.

It’s also helpful if the expansion includes components that are relatively simple for beginners to grasp but interesting enough for veterans to like. Sentinels of the Multiverse released a number of expansions with extra heroes, villains and environments. Heroes are rated by “complexity” on a scale from 1 to 3, indicating how much work they take to pilot. Greater Than Games was disciplined about including a number of heroes with complexity 1 in their expansions and not succumbing to “designer fatigue” and increasing complexity with each new release. (This is true for heroes; the villains and environments mostly got more complex in later expansions.)

Replace rules instead of adding new ones

An expansion can change the existing rules of a game instead of just adding new rules. This is more often the case for “patch” expansions that are designed to fix problems in the original game, but there’s no reason you can’t do it to limit complexity.

The RPG Monster of the Week introduced new actions in its Tome of Mysteries expansion themed around weird powers. The way these work is that during character generation, instead of the “use magic” action, you can choose a new weird power (just one) that’s appropriate for your character, ranging from telekinesis to receiving ghost whispers to just being preternaturally strong and swift. This limits the number of actions players have to remember, which makes it smoother for them to navigate the system and tell the story.

Work within your game’s vocabulary

Your cognitive load becomes heavier if you include components that are hard to parse with your game’s vocabulary, especially if they introduce new keywords that have to be memorized.

Small World‘s sequel, Small World Underground, was standalone and had to introduce a whole new set of races and special powers. However, the designer was careful to limit the complexity for these: Many of them are remixes of the original Small World‘s powers, and the rest used comfortable mechanics that Small World players could grok easily after seeing four or five other powers. This way, the components from both games could be mixed with reasonable success without increasing cognitive load.

Compare this successful to the game Smash Up. Smash Up was essentially designed out of the gate for expansions: New half-decks that could be mixed with the decks from the base game. However, it also tried to be a “gateway game” with simple text on the cards that was easy to parse for people who hadn’t played a lot of games. This meant that as expansions were released, they had to include difficult-to-understand phrasing like “choose a card that’s played on a minion” instead of employing keywords to make understanding the cards more manageable. It’s a tough balance, and I don’t fault the designer for choosing to make the base game as entry-level as possible, but it resulted in cards from later expansions having a higher cognitive load than they might have otherwise.

A Case For Odin

Let’s look at an expansion that succeeded tremendously – A Feast For Odin: The Norwegians.

The idea of an expansion for A Feast For Odin was tough to begin with. The base game is one of the heftiest Euro-style games ever made, with dozens of potential actions, special abilities, and side rules. Explaining the base game can take up to 30 minutes, and adding more content would just make teaching it harder. The Norwegians applied some of the principles described in this article in an incredibly intelligent way that resulted in lots more content for game owners and very little added cognitive load for new players.

  • The original Feast for Odin had four island boards that players could claim and fill with goods. Norwegians introduced an extra four, but advised that they be switched randomly one-for-one with the originals. This way, players still only have four boards to work into their strategies.
  • Norwegians also introduced a new action board, with fewer options (so it was actually easier to teach than the original), more balanced strategies, and an extra column that could be explained in a sentence or two.
  • With the addition of more irregularly-shaped “treasure” tiles, the treasure board from the original (which had individual spaces for each unique tile) couldn’t be used anymore; Norwegians split up these tiles into low-cost and high-cost tiles to make eyeballing which tile you want easier. (There was also a new action that only let you acquire tiles under a certain cost, and these tiles got their own board so you could tell at a glance what was available.)

Despite beginning with a daunting cognitive load already, The Norwegians managed to add a lot of interesting content to A Feast For Odin while only slightly increasing complexity. It’s a masterclass in expansion design and something to draw lessons from, no matter what kind of game you’re working on.

Conclusion

Even if you’re not in the business of designing expansions, playing games with expansions and thinking about their designs are great ways to understand the balance between content and complexity.

The next time you encounter an expansion, whether it’s for a game you’re very familiar with or one you’ve just been invited to play for the first time, think about what it adds and what it costs. Are you more likely to play the game with the expansion added, or less? Is the fun you’re having worth the length of the teaching period? Does the expansion material feel integrated or like a sideshow? You’ll be surprised at what insights you can glean.

Politics & Poker

Territory control games like Risk or Game of Thrones and take-that games like Red Dragon Inn face a common issue. At various times in the game, especially right at the start, all the players are roughly equal, barring starting powers and whatnot. But the game encourages you to be mean to people, whether it be by invading their territory or playing a card on them. So what do you do when the decision of who to target is arbitrary?

(No actual poker discussion here; this article is named after a song from 1960 Tony Award-winning musical Fiorello! I didn’t expect you to get it, but if you didn’t want dense references to things that only eight people are familiar with, you wouldn’t read this blog!!!)

With many of these types of games, what you do is pick someone at random, and frankly, this sucks. Picking someone to target for no reason can feel mean, like you singled them out because you like them less as a person. But making it clear that you’re being arbitrary, by rolling a die or whatnot, can also be just as bad – though your opponent may care less, choosing at random removes what agency you had from this decision and turns your turn into a miniature bean machine.

While this issue is common, it certainly isn’t unsolved. There’s a number of games in both genres that implemented ways to balance targeting decisions between “random” and “obvious”. Let’s check some of them out!

Always Incentivize Decisions

Some games solve this issue by making decisions of who to target have meaningful incentives from the first turn and for the rest of the game. The trick is that the “right” decision can’t be too obvious – otherwise we end up in the opposite problem where choices are so obvious it’s boring.

The take-that game Epic Spell Wars (which I’ve written about in more detail) does this by giving the players benefit for matching cards from the same “school” of magic. Each card targets in a different way – some a specific player to your left or right, some the player with the most HP, and so on. But the benefit from casting a higher quality spell generally outweighs aiming your spells at a particular player. This also adds a chaotic feeling to the game, making spell resolution wilder and less strategic (a good thing in this case).

Some war/territory control games resolve this issue by making the players’ locations asymmetric. Diplomacy puts each player in control of a major power during World War I, forcing them to negotiate and betray each other in order to gain control of a majority of Europe. Because the map of the game is literally a map of Europe, with some players starting the game farther apart from others, decisions of who to target and coöperate with are thus influenced from the start of the game by your neighbors. A player controlling Turkey is going to more likely target its neighbors, Austria-Hungary and Russia, while it’s more likely to make alliances with the player controlling France in an effort to catch Austria-Hungary in a pincer attack. (Caveat: I’m terrible at Diplomacy and haven’t played in 10 years, so forgive me if I got the strategy wrong.)

Compare this to a game like Eclipse where each player has exactly two neighbors at an equal distance to each other. This symmetry makes the decision of which sector to expand into and which to broker a truce with largely arbitrary until players start developing different ship technology, meaning the political interplay stays static and largely uninteresting for most of the game. (That said, Eclipse primarily focuses on Master of Orion-style empire management and technology, so not being great at something it’s not trying to do isn’t the end of the world.)

Don’t Let Players Choose

The territory control game that does the best job of handling arbitrary targeting decisions is Cosmic Encounter. That’s right, fool, I’m praising Cosmic in another article and you’re going to have to read it!!

Cosmic‘s solution to the issues of players not knowing who to target cuts the Gordian knot by simply not letting players choose who to target. Instead, on each player’s turn, they draw a random color from the “destiny deck” and they have to deal with them, for better or for worse.

This accomplishes several things. First, it makes the politics in the game much more fluid. If red is targeting purple, even though they worked together in the previous round to gain colonies, their previous actions mean nothing now and they have to re-establish relations from scratch. This may not appeal to people who enjoy building coalitions long-term, but it keeps player dynamics fluid for the entire game.

Second, assigning random opponents takes the blame away from the players. Choosing a target frequently carries hurt feelings with it – if you chose to attack me instead of my opponent, and both of us offer functionally identical options, you’re essentially just picking on me. It’s much healthier to have a deck of cards to blame for

Third, the destiny deck allows matchups to be uneven. Some alien powers are much, much better than others in combat, so if people had their say they would just attack aliens who weren’t useful in combat. This would make it less fun for people assigned these powers as they would only be able to use them on their turn, while other people with diplomatic or resource-generating powers could see benefits throughout the whole game. This way, combat-power players get their time in the sun without requiring their opponents to make decisions vestedly not in their self-interest.

Other games use different systems to randomize targeting; for example, Tournament at Camelot uses the trick-taking genre as a randomness mechanism. Each trick results in the player who played the lowest weapon card taking the combined weapon cards played as damage. Since you can’t control what cards are in your opponents’ hands, the best you can do is make sure that player isn’t you.

Small World: Hard to Categorize

Days of Wonder’s classic territory control game Small World addresses this issue in an interesting way that doesn’t really fall in the previous two categories.

At the start of the game, and roughly 2 to 3 more times during the game, you select a new fantasy race and send it rampaging onto the board, conquering your opponents and “nameless” NPC tiles. You can choose to launch your attack from anywhere, so who do you pick on?

Small World does the opposite of many territory control games by not encouraging players to fight each other; instead, it incentivizes fighting as little as possible and requires fighting to happen due to the size of the board. Each space you occupy with your civilization earns you 1 point at the end of the game, and spaces containing your opponents’ civilization require significantly more forces to conquer. Thus, in order to maximize your points per turn, you’re encouraged to find empty space on the board and only attack if you have to.

You will have to. The game’s called Small World for a reason; after the first turn or two you’ll have to attack someone in order to expand. Fortunately, the game offers lots of things to balance besides just playing kingmaker and attacking whoever’s in the lead. There’s a large number of different race and class powers, some of which benefit the attacker or defender; additionally, players can send their faction into “decline,” allowing them to pick a new race/class combination while earning passive income. This makes timing just as important as player selection, as you may want to wait to attack your opponent if you think they’ll go into decline on their next turn.

This article isn’t a game analysis of Small World so I’ll leave it at that for there, but the point is that it spends a lot of its complexity budget (especially considering the publisher primarily aims at a casual audience) making “who to target” an interesting question with many different answers.

Conclusion

Every genre has baked-in flaws: For example, deckbuilding games often have very little player interaction, and social deduction games reward being loud and pushy. “Who to target” is one of the biggest issues with take-that and territory control games, but some of the most stellar games in the genre are able to take on this issue and partially fix it in an interesting way.

In a sense, a good game design lesson to learn from this is to not take anything for granted. It’s great to work on a game in a genre you’re passionate about, because you’ll have a better understanding of the game’s appeals to an enfranchised audience, but by questioning the flaws that seemingly come part and parcel with the genre’s strengths, you may be able to create a game that nobody’s seen before.

In Defense of the Meta

Keith Burgun is a game designer, and like most game designers, he has some very strong opinions. One of his most famous articles discusses asymmetric gameplay, especially gameplay where the asymmetric powers are chosen before the beginning of the game (like choosing a character in Street Fighter). More specifically, he intensely dislikes these systems, claiming that an asymmetric game can never be truly balanced and that it limits strategic options on the part of the players.

I read this article a couple of years ago and I tried extremely hard to disagree with it. If you’re reading this, you most likely know me from my work on Magic: The Gathering (or you’re a personal friend who saw me share this on Facebook, thank you!), and Cosmic Encounter is my favorite game, so obviously I’m fond of asymmetry. But the more I thought about it, the more I realized that he was right to an extent. Asymmetry, and the concept of a “metagame” where you pick a character or build a deck before beginning your games, bring very little to the table strategically and can often make games that are supposed to have high stakes feel flat.

But that doesn’t mean that asymmetry is useless – in fact, outside of the realm of strategy it serves many valuable purposes.

Onboarding

A common response to Burgun’s article is that using asymmetric player powers makes an easier learning experience because it sections off game elements so new players only have to master part of it to play competently; for example, in fighting games, a player can choose to practice a single character over and over again instead of learning the ins and outs of the whole roster. Burgun’s response to this argument is that achieving true understanding of the game system requires players to know what all the characters can do in detail anyway. Though this counterargument is sound, it doesn’t take into account the amount of information a player needs to retain at the same time, which asymmetry can significantly alleviate.

Even in games with low asymmetry like Chess, the sheer number of possible moves is so high that players beyond absolute beginners parcel them into typical opening, midgame, and endgame tactics. High-level gameplay becomes less about making the right decision at the right time and more about memorizing as many possible “decisions” as you can, upon which the “right decision” will frequently be obvious depending on player skill. 

What asymmetry provides in games is the ability to only have to concentrate on a tenth or a twentieth of all of the possible moves, which, in a system with an already large enough degree of complexity, could already pose a substantial cognitive load to the average player. We’ve seen that games where all moves are available to players in every game have a high skill ceiling, but they also have a high skill floor. Not every game is going to want that, whether for commercial viability or for the sake of a more casual play community.

Relatability

People wear shirts with their favorite Magic color, favorite Street Fighter character, or favorite Pokémon…uh, pokémon. Despite Go being a game with more strategic depth than any of the three, nobody wears shirts with their favorite Go opening move.

This is because of a combination of two things: First, separating game elements into discrete packages, especially ones represented by a character or faction, allows players to pick a side they like. Second, people LOVE picking sides. I don’t know whether tribalism is something baked into human psychology or if it’s the product of a competitive society, but if you give players the option to decide between multiple distinct factions, they eat it up. I’m pretty sure a huge part of the appeal of the Harry Potter franchise is the Sorting Hat.

Games are frequently more appealing, more impactful, and better able to retain players for longer when the player can fold an element of the game into their identity. Discrete asymmetric elements, either in the form of a defined style of play or in a character or faction with a personality and visual aesthetic, go a long way into a game appealing to the heart as much as it appeals to the head.

Experiential Vigor

A while ago I wrote a post separating games into “strategic” games and “experiential” games. Strategic games are evaluated among the lines that you most commonly hear from reviewers: Is this challenging? Does it offer a lot of different lines of strategy? Can you play it repeatedly without finding a “best” strategy that makes it boring?

Experiential games, on the other hand, are judged based on whether they produced an experienceBetrayal at House on the Hillis the game I most frequently use as an example of a successful experiential game that’s fairly poor on a strategic axis: It doesn’t matter if the game is mostly about luck if it creates an engaging story about a motley crew trying to uncover the terrible secret about the house on the hill. Asymmetric character stats help with this, making the elderly professor feel like an elderly professor and the dimwitted track star feel like a dimwitted track star.

Asymmetry is very useful for experiential games because it defines a “side” or “character” in a clean way that prevents confusion. If you have every option in the game available to you at all times, it’s more likely you’ll make inconsistent choices in the name of strategy that muddy the narrative. That said, there are ways that a “side” can be developed by choices supporting like choices – for example, in Terraforming Mars, if you build projects that give you titanium, you’ll probably want to spend it on more projects that want titanium, cementing you as the “titanium strategy” for the rest of the game. But start-of-game asymmetry is a seamless way to make the player feel like they had a real identity for the entire play experience. 

Conclusion

If super-deep gameplay with a huge number of options was all it took to make a successful game, we’d all be in a room playing Tigris & Euphratesover and over again until the end of time. Games have more to offer than being able to scratch your chin and go “ah, good move,” and asymmetry is a very useful way for your game to reach these alternate goals. But be careful that you don’t create a system that’s so hard to balance, the whole thing falls flat, and consider other options than just “deal each player a power at the start of the game.” Don’t take anything for granted and discover what works best for your game!

The Tyranny of Numbers

A few months ago, I read an excellent article by Kim Foale about how victory points are often uselessly abstract within the narrative of a game, sometimes promoting a colonialist viewpoint by unconditionally rewarding morally fraught actions. I enjoyed the article, but it also made me think about the use of numbers in general – not just victory points – in tabletop games. The medium of sophisticated hobby games is still in its infancy, and one of the most telling signs is its limited vocabulary. As 3D animation could once do no more than move a few spheres and cubes, board games today are only capable of conveying systems if they can do it quantitatively. 

Math Geeks

Most of the designers that brought tabletop games into today’s renaissance had a mathematical background, most famously Dr. Reiner Knizia. Some early hobby games are explicitly throwbacks to famous mathematical problems; for example, 1998 Spiel des Jahres winner Elfenland is a gussied-up version of the traveling salesman problem.

This is all well and good, because a mathematical background gives a designer a lot of ways to make games tense, balanced, and engaging. But as theme and aesthetic gradually became more important as the board game hobby picked up in the early 2010s, holes started to appear in the seemingly impenetrable mathematical landscape of design. 

The Head And The Heart

Board games are very good at representing systems and relationships that can be easily measured. Many of the themes and mechanics of the best board games are essentially economic in measure, and track things that can be quantified. In Concordia, everything important has concrete numbers – the geographic borders of each region, the quantities of grain and cloth in the player’s possession, the amount of money required to install a trading post in Londinium. 

However, when it comes to determining the unquantifiable, board games often stumble. How can you use a pile of cardboard to measure a concept like “beauty” that each player may view differently, or a complex philosophical idea like “happiness”? Games usually either do the best they can with the tools they have (so you get games like The Pursuit of Happinesswhere “long-term happiness” is a numerical, objective thing that works the same for all people) or try to convert these concepts into something that can be measured (like Modern Art’s artist popularity). Neither really captures the concepts the theme tries to encompass.

These problems are not inherent ones to tabletop games; they exist because designers as a community haven’t yet found ways to metaphorically convey these thorny concepts to the players. In much the same way, video games are just now beginning to figure out how to tell stories that don’t look similar to movies, and Internet media is slowly exploring the space available to it with works like Jon Bois’ 17776.

Measuring the Unmeasurable

Just because unquantifiable systems in board games are rare and in their infancy doesn’t mean that they don’t exist. Let’s look at a few of them.

The most common qualitative system in board games is the impress-the-judge system popularized by Apples to ApplesCards Against Humanity, and Stand Back, Citizen!. You can’t objectively measure “humor” or “suitability” for fuzzy concepts such as words, so a player is chosen to arbitrate with the understanding that their judgment will be biased. This is pretty narrow design space to explore, and many of the innovations within the genre are more based on how you construct your answers instead of who decides the victor. A few games have tried getting most of the table to vote, like Quiplash and The Champion of the Wild, but having all players simultaneously perform a fiddly action that often requires point markers is cumbersome without digital intervention. 

One of the most interesting forms of non-number based mechanical resolution is the simple horror RPG Dread, where all skills are measured via Jenga tower. Whenever you want to do something that’s not assured, simply pull and place a brick from the tower – if it falls, your character is eliminated, whether they got torn apart by werewolves or realized they had to go teach a class. This is a way of expressing a feeling in a way that is neither numerical or based on player opinion, and it does a great job of conveying the tension of a horror movie – you know something awful is going to happen but you don’t know exactly when. 

For a less thematic and more strategic example, Treasure Island elevates the math conversation from arithmetic to geometry. The win condition of the game is delightfully simple – draw a circle around your miniature, and if you nail the exact location of the treasure, you’ve outwitted Long John Silver! By mostly eschewing numbers and relying more on physical locations and shapes, Treasure Island possesses a rare viscerality and impact that makes a huge impression on first-time players.

Conclusion

In order for tabletop games to take their next step artistically, they need to explore the mechanical and thematic options available in a craft format where not everything has to be determined numerically. I am certainly not going to swear off numbers completely – they’re too valuable to cut away just because I want to be a contrarian – but as I continue my design career I will be keeping an eye out for ways for people to win that don’t necessarily involve accounting. 

Scaling Food Chain

About a year and a half ago I became acquainted with Korean reality television show The Genius, which quickly became my favorite TV show of all time. I had already enjoyed elimination-based shows like The Amazing Race (which I’ve written about before) as a guilty pleasure, but The Genius cut out all the unnecessary gristle – the drama, the discomfort, the weirdly personal grudges – and replaced them with juicy strategy.

The games on each episode varied in quality, some of them trivially solvable and some of them incredibly deep. (One of them was literally just the game No Thanks). My favorite of them all was a social deduction game called “Food Chain,” where players draw the roles of different animals, each with their own powers, habitats, and win conditions, and do their best to make it to the end of the game without dying.

I’ve wanted to run Food Chain myself, but getting exactly 13 people to play it would be prohibitive. So why not scale it down?

Well, “why not” is because Food Chain’s interlocking roles make it almost impossible to scale it down without unbalancing it. This article explores this issue in depth.

If you aren’t familiar with Food Chain, I’ve prepared a companion article that goes over the rules. You may want to keep it open in a separate window.

The Problem

Much like a real ecosystem, if you remove one species from the ecosystem of Food Chain, the whole thing collapses. Some of the roles, like the Crow, seem extraneous to the main puzzle of predators and prey each trying to survive, but secretly represent key resources to other roles. Here’s some of the issues lurking beneath the surface of Food Chain that make it so hard to rescale:

  • For predator roles like the Lion, player count is a resource. If there are fewer players, it will be less likely for a predator to be able to meet their feeding requirements, so life will be significantly harder for them.
  • The Sky represents an asymmetric benefit on behalf of the Eagle and similar bird roles. If a bird is removed, it makes life significantly harder for the Eagle, who now has to leave the Sky more often and dodge the Lion and Crocodile.
  • Every role cares about every other role, both for predator/prey reasons and for more nuanced ones like habitat or opposing victory conditions.

The Approaches

Here’s some approaches to scaling Food Chain down to a manageable 8 or 9 players, in roughly the order I thought of them.

Trim extraneous roles

As mentioned, some of the roles just sort of…exist on the fringes. The Crow doesn’t really have much to do with the game, the Mouse and Plover are “nice to haves” instead of “have to haves”, and the Chameleon’s only purpose, on its face, is to mix things up a bit.

But as soon as I thought about a Food Chain without these roles, I realized: there are no extraneous roles! Cutting the Plover would make life significantly harder for both the Crocodile and the Eagle (who can eat the Plover if it goes to the Sky), removing the Chameleon would make it trivial for predators to avoid the Snake, and removing any prey animal makes the Four Stooges (otter/deer/rabbit/mallard) more likely to die, making starvation at the end of the round due to indestructible prey animals exponentially more of a concern.

Remove predator/prey roles symmetrically

This approach focuses on maintaining the predator:prey ratio by taking out both roles at roughly the same rate. The current ratio is 4:9, so if I were to remove, say, the hyena, I’d also want to remove two prey, perhaps the chameleon and the crow. 

The issue with this is that every predator except the Lion also counts as prey to animals further up on the food chain, so if we remove the Eagle and some other roles, the Crocodile and Lion are significantly more disadvantaged than the prey animals, who know have fewer adversaries and a higher chance to make them starve to death. 

Remove prey and make predators less hongry

You can avoid some of the issues of the game becoming too difficult for predators if you change their win condition requirements while removing some of the extraneous prey animals like the Crow. For example, the Lion only has to eat 3 out of 4 rounds instead of every round.

While this is more likely to create a lower player count game with roughly the same balance than simply chopping away roles, the “knobs” that you have to adjust are pretty meager. You can change predators’ eating requirements, but getting a whole extra round where you don’t need to eat can make the game less stressful and therefore less fun on the whole, and there’s no granularity to this adjustment. (The game’s complex enough so that’s fine.)

Change the roles entirely

 What if the food pyramid was entirely different? Different roles would be used in different player counts, and maybe even different habitats. Instead of the current Lion-Mouse-Hyena triangle we have now, we could have a different apex predator with different needs, powers, and adversaries. We could also eliminate the Sky entirely, or make it so that animals can access it more easily (by jumping really high?). We could even change the number of rounds.

But at what point does this game stop being Food Chain and start being Food Chain 2? If the attack mechanic is the only thing that stays the same from player count to player count, Food Chain with 8 players and Food Chain with 13 are as much the same game as Clueand Parcheesi.

The Solution

Here is my galaxy brain solution to solving the scalability issues of Food Chain: Maybe there isn’t a perfect way. Does a game need to have the same balance and gameplay experience for every player count?

At some level, you want to know that when you bring a game to the table, you’re going to roughly know what you’re getting into. Some games, like Gloomhavenand Sentinels of the Multiverse, put a lot of effort into making the lowest player count feel as challenging as the highest player count.

But with others, you’re going to have a wildly different time depending on how many people you’re playing with. Libertaliaand 6 Nimmt!are tense mechanisms of risk and bluffing at 3-4 players and gradually become more of a lunatic circus where nothing matters as more people arrive to complicate things. Roll for the Galaxy at 2 players involves hard-reading your opponent’s plans and jamming your hand into them, while at 5 players it mostly involves doing your own thing and occasionally benefitting from your opponent choosing to develop technology.

And frankly, social deduction games are both the most likely to feel different at different counts and the most likely to have a huge range of player counts, topping off at 8-10 players. With some of these, it can become trivial for one side or the other to win – in A Fake Artist Goes to New York, for example, the fake artist is extremely likely to win at high player counts, but players still get a lot of fun out of the game’s base mechanic of one-stroke-at-a-time drawing.

Put Down That Controller And Go Outside And Go Back Inside And Pick Up That “Stand Back, Citizen!” Deck!

Why would you ever see a play? Movies are more affordable, have higher budgets, can be seen anywhere, and let you eat snacks in the theater. 

Most arguments for the productivity of seeing Our Town over Die Hard are based on intangibles – that visual metaphor is more possible in theatre, that you’re experiencing the magic with real actors in real time, that certain abstractions of time and character are available. How accurate this is, and how much weight it carries, depends on the individual, but these are the most common arguments. 

The competition between board games and video games has a similar dynamic. Video games are more portable than board games, and (depending on your buying habits) easier to purchase cheaply. They have animation and sound design, and the computer behind the game allows for complex calculations and radical mechanics that humans can’t do by hand. I’ve certainly heard arguments, with some validity to them, that any scoring-track based game would be better accomplished on a tablet. So in this, too, we must turn to intangibles*.

One of the most common intangible arguments for the benefit of board games over video games is that they create a more social atmosphere than sitting in front of a TV. I’ve found that to be true myself, but I didn’t want to state a claim without exploring why. In this article, I will detail a few reasons why I believe board games are often more social than video games.

*For a designer, board games have many concrete advantages. For example, it is much cheaper to produce a board game, and you don’t have to know how to code, animate, or compose music. But this is about players.

I Demand Your Attention!

The first reason is more a quirk of design than something inherent to the two media: The vast majority of multiplayer video games are played in real time, while the vast majority of board games are turn-based or otherwise broken up by pauses.

In a real-time game, your attention is inevitably drawn to the game instead of the people playing it. You can’t afford to not think about the game for too long, so while there may be some conversation, much of your mental effort goes to winning the game. 

When astonishing moments happen in a real-time game, you don’t get a chance to celebrate it. If you’re playing, say, Street Fighter, and you parry your opponent’s entire Chun-Li super, the game is still going aftewards unless that cool moove instantly ended the match. Not even “ended the game”; unless you can take a break to appreciate what a good move you just did, Street Fighter’s merciless announcer will make you go on to round 2.

Pace of Play

This theory doesn’t fully hold up under scrutiny, because there are board games in real time (Magic Maze) and turn-based multiplayer video games (Pokémon). The underlying mechanism that makes video games less accomodating to social interaction is a bit subtler: The pace of a tabletop game is determined by the group as a whole, while the pace of a video game is determined by the program.

As an example, let’s take the game Quiplash, an impress-the-judge game played on phones. Quiplash is a very well-constructed game, and its questions are more consistently funny than any other of the numerous party games I’ve tried. However, the results of these funny questions don’t necessarily generate as much impact from the players as they could, because even if an answer is so funny the whole room is laughing for a minute straight, the game skips merrily onward to the next question, cutting off the reaction that the previous one might have deserved.

When an exciting or funny moment happens in a game where progress is determined manually, the players can stop for as long as they want until they mutually feel it’s time to move on. When progress is determined by a computer arbiter, it can diminish a moment of excitement that might have been developing for most of the game. 

Communication

For the third reason, we can look at the exception that proves the rule: Video games that are successful in engendering communication among a group of players. The best example is the infamous co-op game Overcooked. In Overcooked, players are chefs attempting to fill a rapid series of food orders by chopping ingredients, assembling meals, and washing dishes; levels include gimmicks such as cooking on two moving trucks, in a haunted castle, or in the hot lava level that is legally obligated to appear in every video game.

The secret of Overcooked’s success as a social game is that it is impossible to win without in-person communication. Orders fill up fast enough that the only way to keep up is for players to work together to expedite the cooking, serving, and dishwashing process, and the only way for everyone to coordinate effectively enough is to talk. 

Normally, the demands of video games outweigh the ability of players to talk to each other, but because role division is such a critical part of Overcooked, the rapid pace of play actually encourages the players to interact. And this is the main difference between it and many other games, even party games: Talking to other people is part of the flow of play, whereas in many other cases it’s a distraction from winning.

In board games, interacting with other players is a distraction, but unless they’re deep in thought and you interrupted them, distractions are forgiveable. It’s mostly when arbitrated by a computer that taking your mind off the game for even a short period of time can lead to big problems.

Conclusion

Despite the ubiquity of digital games, board games have experienced a renaissance over the last decade, growing significantly both in number of hobbyists and in the number of quality games. There must be a reason for this, even if that reason is an intangible, and this post was my way of trying to puzzle out reasons people choose to manually adjust VP totals instead of letting the Nintendo box do it for them.

Guardrail Safety With Professor Geist

Since the 1940s, the humble guardrail has protected drivers all over the world from themselves and each other. This critical piece of highway infrastructure helps prevent deadly head-on crashes, and comes in a variety of styles depending on the needs of your transport system. For the rest of this article, I will go into detail on the history and uses of guardrails.

Just kidding. But did you know that games have guardrails too, and they’re just as important? Only instead of preventing you from dying in a fiery car crash, they prevent you from being frustrated for 90 minutes.

The Perils of Leonardo da Vinci’s Advertising Budget

The game that inspired me to write this article, ironically, is one that has a significant lack of guardrails: The Princes of Florence.

Princes of Florence is an auction game with some tile-laying elements. Players bid on various buildings, landscape features, and clowns to get better work from the artists in their colony, which they can then sell for more money. What’s significant about the game, however, is that when you sell a work, you can choose any ratio of victory points and cash that you want, and that this is by far the most common way to get VP.

So although this means there are a lot of ways to play the game, it’s also very easy to shoot yourself in the foot. If you don’t grant yourself enough money after selling a work, you won’t be able to catch up with the other players or the slowly increasing point minimum for works. If you aren’t VP-greedy enough, you will never be able to catch up to your opponent. The fact that this game was made in 2000, when games were significantly less user-friendly, is apparent in the lack of mechanics that constrain players’ choices in favor of dangerous liberty.

The Goal of Guardrails

A game’s “guardrails” are its mechanics that allow players to remain in the game with the ability to take meaningful actions until the end. What this actually means depends on the type of game.

An example of a game with good guardrails: In Concordia, a slow start can price you out from building trading houses. Fortunately, the Prefect card (which allows you to get any good available on the board) and Mercator card (which lets you exchange money for goods and vice versa) mean you’re never in a position where it’s impossible for you to participate in the game’s economy. Concordia especially demonstrates the nuance of a good guardrail mechanic, because the Prefect card rewards players who built all over the board while allowing someone who needs two cloth to get back into the game their out.

Strategy games construct their guardrails in many different ways. Here’s just a few of them:

  • Tzolk’in: You need corn (currency) to place workers. If you don’t have enough corn on your turn, you can lose VP to beg for enough corn to continue. This normally accrues a penalty, but if you’ve gotten into such a bad situation that you can literally take no moves without begging for corn, the gods (and designer) take pity on you and you can get the corn for free.
  • Galaxy Trucker: The “Ship Insurance” mechanic caps the amount of credits you lose at the end of a round for your ship’s components getting blown up.
  • Ra: You always have four opportunities to win auctions each round. Having no high-numbered Sun tiles means you’re less likely to win large pots, not that you can’t belly up to the table.

Some very common guardrails are making actions more valuable in later rounds (Modern Art) and keeping rounds entirely separate so that an early mistake isn’t a disaster (Libertalia and the aforementioned Galaxy Trucker). 

Designing Guardrails

Guardrails have a lot in common with catchup mechanics, but the big difference is that a guardrail doesn’t care if you can’t win the game anymore: In the games mentioned above, you can easily put yourself in a situation where victory is wildly unlikely. A guardrail mostly cares that you can’t play the game anymore. In a sense, it’s more about preventing pseudo-player elimination.

If your game has any kind of economy, try to notice in playtests when players get locked into non-action. Is there a situation where a player has no equity that can be turned into resources somehow? In that case, consider adding an emergency action (like discarding cards for money in Terraforming Mars) that can allow someone to claw their way back into empire-building.

If you’re building a combat game, can someone never fight back against a vastly superior opponent, but at the same time be able to keep playing the game in a futile struggle? You may consider making the game more gentle, so this player still has a chance, or more brutal, so the game’s over more quickly. As a case study, Nexus Ops’ guardrail is that the method of gaining victory points is through individual achievements instead of board dominance, so whether you’ve conquered the board is only sometimes relevant to whether you’re able to win.

Essentially, assume your players will make the worst decisions available to them and see what situation that gets them in. Can they still laugh and have fun, or is it a frustrating grind? Try to skew your game more towards the former.

Conclusion

 This article was mostly written as a historical examination, as guardrails are so common in modern board games it’s the exceptions (Food Chain Magnate, for example) that are notable. However, situations involving guardrails or lack thereof appear frequently in early builds, and recognizing when you need to create a floor for how badly a player can do is an important step in the development process. If this article helps make those situations more visible, it’s a success.

The Most Beautiful Cards, Part 1

It’s the little things that make a difference.

The beauty of a game’s design can sometimes be condensed into something as small as a single card, a tiny packet of information that revolutionizes the way it – and games in general – are played.

In tribute to Mike Selinker’s blog The Most Beautiful Things, I am writing a series about the most beautiful cards in tabletop games. Here’s the first installment, with three cards that made an outsized difference in the games they represent.

Libertalia: Brute

The mechanic at the heart of Libertalia– that everyone gets the same randomized hand of pirates – is genius and elegant, and would stand alone as a brilliant design innovation no matter what pirate cards it ended up using. But without the Brute, Libertaliawould be a lot more easy to parse, and a lot more polite.

Most of the other player-versus-player interaction in Libertalia is various shades of passive-aggressive. The Beggar steals money from the person with the highest-ranked pirate, the Monkey dumps cursed Incan gold on the player to your left, Granny Wata cancels other Grannies Wata, etc. Not the Brute. When you play the Brute, you are actively choosing to punch another player in the face.

The best part about the Brute is that it’s a fairly low number in the booty-acquiring order and does nothing else of value. So you only play it when you really feel like saying “screw you” to whomever you suspect is going to be greedy and go for the treasure.

No, wait, the best part about the Brute is when multiple players play them at the same time and it looks like this:

Wait, wait, wait, the best part about the Brute is guessing when your opponents are playing theirs and playing a low card to counter it – or better yet, a Parrot so you can play the exact right card to get first pick of the treasure without getting punched overboard.

Hold on, the bestpart about the Brute is holding onto yours for two rounds, until everyone’s forgotten that there’s a Brute left in play, until you punch out the opponent’s Captain they were relying on to win the game and smugly help yourself to a 1-point barrel.

I could probably find more things that are the best about the Brute, they just never stop! A cracking good card.

Dead of Winter: Sparky the Stunt Dog

Dead of Winter is a grim co-op game about the last vestiges of humanity desperately clinging to life as they’re threatened by both the walking dead and the bitter cold. If Plaid Hat had thrown it out there without any levity to it, the experience would be so heavy and personal that only the hardiest players could stomach multiple plays. But varying the tone on the “crossroads” cards could backfire, creating goofy moments when they aren’t called for.

But what if all of the crossroads cards could be goofy in the right context? Here to the rescue is Sparky the Stunt Dog.

Sparky questions the unspoken assumption of crossroads cards: That they are all performed by humans. Sparky erects a barricade against the walking dead! Sparky researches the undead plague! Sparky plays some Johnny Cash on the guitar! Good dog, Sparky!!!

Through just one card, the tone of the game can dramatically shift. And the brilliant thing is, for those groups who want a consistent tone, it’s as easy as putting Sparky back in the box during setup.

Cosmic Encounter: Negotiate

Brute and Sparky are cards that elevate their respective games from “decent” to “great”, but the humble Negotiate card is what changed Cosmic Encounter from a run-of-the-mill space combat game into one of the most influential – and most impactful – games of all time.

The Negotiate card does several vital things. First, it increases the variance of each player’s hand strength in combat. This seems like a bad thing off the cuff, as more variance means more instances where player skill can’t affect the outcome of a round, but Cosmic’s strength is, as described by the box, that it’s a game of infinite possibilities. It’s the times when you have a grip of 5 Negotiates that make you appreciate the times you have three double-digit Attacks.

Second, it provides an element of player interaction that goes beyond “shoot each other with pew pew guns”. Cosmic’s emphasis on alliances, going so far as to allow multiple players to win the game at the same time, is enhanced by the ability to create mutually beneficial deals.

Finally, and most importantly, it creates gameplay that isn’t strictly fettered by rules. What players do or say during negotiations is pretty much anything goes – for example, you can claim you have an attack 40, offer to trade it, then grimace and decide against it when you really never had an attack 40 in the first place.