How to Play Magic

Adam Bignell, 2021

The Gathering

In late 2016, I met with three people to assess their viability as roommates. One year into my computing science degree, I was looking for someone who I could peacefully cohabitate with for the foreseeable three (more like four, actually make that five) years. My criteria were simple: Relatively quiet, reasonably clean, unlikely to throw parties that would disturb my obsessive studying. Friendly, one hopes. I met with three people. Two of these meetings lasted for sub 30 minutes. One was, at least according to my now rosy memory, painted with the time-dilating brush of nostalgia, supra two hours.

That longer meeting, the last, was how I met Stefan. Stefan and I have a lot in common it turns out. We both, with our physicality, manner of speech, tendency to underdress etc. announce ourselves as academics. We both transitioned from arts-leaning backgrounds to computing science. We have both experienced the domestic strife that is troubling and worldview-shaping but short of traumatic. Our families are distributed, non-traditional. Our mothers are teachers and our fathers are named Randy. We each have multiple sisters. We were both born on the 18th day, and we saw some special significance that we lived in Suite 188. We both once believed in God. We both, ultimately, love people.

The first potential roommate spoke interminably about the near miraculous economic potential of Pakistan, without noticing my awkward chuckles and “Anyways wanna see the bathroom?”s. He suggested I buy Pakistani real-estate within ten minutes of meeting. He was a business major, and chemistry was nowhere to be found.

The second was better, though he was noticeably older than I was. He didn’t stoop to coldness, but spoke in a matter-of-fact, utility-minded way that didn’t bode well for cohabitant relations. He didn’t for instance smile or attempt small talk. He commented on the cleanliness of the stove.

When I showed Stefan Suite 188, he viewed his future room with a once-over. He didn’t check for outlets in the bathroom, or inquire about where to do laundry, or test out the sketchy blind-runners. He viewed the physical space with an expediency that indicated that the suite was sufficient; it had a kitchen, a living room, and room for a bed. What more could he need?

And then we sat on the dual second-hand couches that filled the living room and spoke at length. The couches, 5 years later, show visible wear.

Stefan has had more Magic Cards stolen from him than I’ve owned. I have watched him reach and sustain 120 actions per minute in a 3v1 Starcraft game that he won. He knows the difference between satire and parody. One summer he worked the night shift at a local grocery store as a way to inspire discipline, and it worked. Recognizing the deficiencies in the official rules of D&D, he designed his own system from scratch, an endeavour that took multiple hundreds of hours and countless low-tech solutions for brainstorming. He has an old blackberry whose note-taking application is a treasure-trove of role playing insight and world building afflatus. More than once, maps of invented countries apparated in our dining room, having leapt from felt-tipped wands onto reusable, rollable canvases. He has resisted the gravitation of social media better than any other human I’ve met. He got his first smartphone multiple years into us living together. Multiple times he has walked home two hours in the snow and rain because the buses have stopped running, arriving soaking wet. He is, officially, a Hearthstone Legend. He reads citations. He quotes Shakespeare. He did drag.

In that first conversation I received only shards of the above, shimmering gems that read as proof of a great quarry somewhere over the horizon. That evening, I messaged options one and two to tell them sorry, no dice. Stefan, I think, recognized that I too had gems of the sort he was interested in. We have spent the last 5 years mining.

Midnight Duelists

When I first meet someone, a countdown begins before I begin extolling the virtues and vices of adopting a postmodern perspective. This perspective, for better or worse, is my idle state. There is of course another essay to be written about postmodernism at large, but let us boil postmodernism down to “contextual awareness” at multiple levels of abstraction.

What does what you said mean? What does it mean that it was said? What does it mean that you said it? What does it mean in the context of this culture? How does your notion of what it means differ from my notion of what it means? As I move through the world, these questions are ever-present, mutatis mutandis for, “What does that scene/advertisement/relationship/word/body language mean?”.

If I say “I love you” and you say “thanks”, the exchange is heartbreaking, not polite, despite virtually all the component words of the exchange being positive. What other implicit semantics are bubbling on the periphery? 

The most meaningful conversations and relationships are those that leverage this perspective. In an attempt to open the floodgates and signal “hey, it’s okay and even good if we communicate this way” it is only a matter of time before I get meta with the people closest to me. This is me sharing what I love.

Stefan loves systems. Yes, he loves video games, board games, puzzles, conversational duels, media analysis, and nerd culture generally, but it is systems, the architecture of all the prior, that he loves. To Stefan the highest form of beauty is that of interlocking mechanics, mental cogs and pulleys that interact and form crystalline structures. A revealed attack in chess (which he handily beats me at). A clever proof. Entendres. Multiply-applicable components, multiply-achievable results.

Stefan never humoured me, or played-along, or patted me on the back when I began talking about how syntax was an underrated linguistic subfield. And neither did I him, when he explained the finer points of his epistemic building blocks. In fact, it turns out we loved the same thing.

The conversations I have shared with Stefan have been life-changing. There have been countless instances where I have badly wished for a recording of the prior hour (or two, or five) of conversation we shared. We once momentarily discussed recording ourselves, but decided that it would corrupt the magic. There were times that we felt that we had unlocked some idea-generating potential energy that was inexhaustible: I know we both felt this way because we each said so. A common outcome when we sat around our shitty and horrendously scratched and burned dining table was that our bodies would eventually fail us before our minds did, and we would be forced to go to bed with something like a cognitive contact high.

I realize this sounds melodramatic and perhaps a little conceited, but the abject aliveness I felt during these conversations deserves capturing: these moments are what justify getting out of bed and brushing your teeth yet again.

We had two modes of engagement. The first of these was conversation proper, conversation as is typical. We discussed things that happened, things we saw. Games we played, movies we’d watched. It was normal content, but the flow was quick-paced, sustaining tempo. Because both of us saw that conversation was about more than simply sharing. At its best it was about engagement.

When we talked, it was as if each of us was asking ourselves “What is the most interesting thing I can say in response to what was just said to me?”. A single party doing this is often sufficient for a conversation to be functional, with potential to be interesting. But when two parties do this, the result is an industrialized engine of connection and intrigue.

A frequent pattern was this:

1. One of us brings up an interesting situation from a piece of media.

2. The other notices how this situation applies to other media they have engaged with.

3. The first person generalizes to a broader principle that applies to the world at large.

Up and out, in other words. But we’d just as frequently go down and in:

1. One of us brings up an interesting situation from a piece of media.

2. The other suggests a slight alteration to affairs that changes the situation.

3. The other changes the situation further, inventing an extension to the story.

4. New components are added wildly and explored by both parties, until 1am (at the earliest).

The first of these bore fruits of insights and wisdom. The second, hilarity and creativity.

Our kitchen has, I estimate, no more than 30 square feet of standing room. With two people it is cramped. With three you must engage in a form of dance and mumble ‘sorry’ or ‘excuse me’ every 15 seconds. The upper corners sometimes bear drops of a mysterious orange slime — mold, we suspect — that must be due to poor ventilation and a barely (now not) functioning stove fan. You can’t even see a window from the kitchen let alone look out one. Despite these shortcomings, Stefan and I have spent hundreds and probably thousands of hours standing in the kitchen, our minds in flight far away from Inlet Drive. We would meet at the fridge accidentally, during our irregular excursions for ramen and burritos, and conversation began so quickly that we never even thought to move to the living room. It’s not that we dropped what we were doing: it’s that comparatively, we were doing nothing before.

Recently, Stefan compared me to an Alka-Seltzer tablet that produces an effervescence in him. “I realize, ‘Oh yeah I DO have things to say’” were his words. I reciprocate this sentiment completely. Where typically I have to carefully vet the things I bring up and the way I bring them up to maximize the chances of people engaging, virtually everything Stefan and I discuss is a launching point. Eventually, we stopped needing to ask “What is the most interesting thing I can say?”. It was not the content but the methodology that produced the magic, and we have practiced the method so wholly now that when one of us doesn’t do it, it’s indicative of poor health.

Which leads to the second mode of engagement: meta-conversation. We exchanged analyses of conversations we’d had with others, hypothesized about the best ways to get people to test the water with both feet. How to parry and how to step in for a winning point in duels of the mind. Verbal balestras, social slide-steps. When to let someone score a point against you. In our more cynical moods, of how odd it was that people seemed to want to end conversation as quickly as possible. How frustrated we were when it felt like people were looking for excuses to tap out.

The most common point we doubled back on was how we might encourage other people to engage with us in the same way we engaged with each other. We had already independently established that there was something fantastic(al) crackling in Suite 188, and we desperately wanted to take it out into the world. Systematicians both, we decided to assign names and phrases to our methodology for easy reference.

KeyForge

We referred to our style of conversation, naturally enough, as ‘rabbit-holing’. We established our shared conviction that all recreational conversation should be exploratory. We bemoaned people’s inclination to say that they “didn’t have anything to say". As we had tested for ourselves, there was always a connection to be made. The world was simply too dense with information. We saw it as impossible that human beings, rich with interior experience, could make it to adulthood without encountering so many things as to be able to relate with, well, just about anyone. There were of course limitations to people’s time and energy, but the issue was not differences of interior richness. The issue was with methodology.

We likened the space of all possible conversations to an infinitely large apartment building, where each apartment represented a single topic. Once you gain some familiarity with a topic, you are granted a key to that apartment. So if you are a chef by trade, you can probably converse freely about — I am not a chef — favourite knives, preferred spices, nuances and quirks of the industry, relationships with the waitstaff etc. These are your meat and potatoes (if you will). You get a key, and now you can hangout in your cheffing-themed apartment, with culinary posters on the walls and exotic utensils on the counter.

Another person is an avid skateboarder. They have deep knowledge on their favourite griptape, preferred foot stance, nuances and quirks of the sport, relationships with the BMX kids etc. They have their own apartment, replete with half-pipe, streetwear, and slacker jargon. Superficially it might look like neither has anything to say to the other. But by paying attention to e.g. the language both parties are using (“favourite [tool of interest], relationships with [adjacent though dissimilar party]”) it is possible to bridge the gap. Analogies can, we believe, always be found. This isn’t just what makes conversation good: It’s why we converse in the first place. If it were about data, we’d read textbooks.

In discussing this, Stefan had remarked: “Everyone is obsessed with rooms. I want to make keys.” This is the thrust of what we valued. It was never about finding pleasing rooms, where the furniture was arranged in your favourite way and the drapes were your favourite colour and you wanted to set up shop forever. It was always about finding doors between these rooms, into rooms that could be equally — and, I dare to suggest, more — interesting. Hybrid rooms that met in the middle. Novel rooms, with chairs of improbable geometry. Rooms no one had seen but us. We wanted to sprint through the apartment building, occasionally stopping to smell what the chef was cooking, but rarely sticking around for dessert. The deep connection I share with Stefan is due to the fact that we each left our rooms independently and met in the hall.

And began, with the greatest haste, to forge keys.

Memory Jar

I have debated with myself if I would include some examples of my conversations with Stefan. There are a few reasons I almost omitted them. For one, capturing the dialogues directly, which would certainly be the highest fidelity and most effective humour-wise, is totally impossible; my memory just isn’t reliable enough. Second, the most profound and moving conversations I have shared with Stefan took place over the course of hours, and relied on deep personal context that is hard and even inappropriate to capture here. Third, writing out humorous, situational moments is never funny. “You had to be there”. Finally, and related to the prior, there is some feeling of ‘dissecting the frog’, of blaspheming the conversations themselves. ‘Kissing and telling’.

But the voice telling me to include some descriptive examples prevailed: I think they do a good job of summarizing the playful tone that was omnipresent in our conversations, and show how the most innocuous of source material could go supernova. Plus, my unreliable memory is actually all the more reason to capture examples while I still can, lest these conversations which are so precious to me disappear completely (I occasionally feel a pang of loss when I consider how many of these dialogues have been lost already: It took us months to remember “Never judge a cook by his brother”).

Journey to Nowhere

One night in particular does a good job of capturing our shared curiosity, and willingness to entertain (superficially) utility-less ideas. We had just finished James Byrkit’s Coherence, which is a psychological thriller that tracks a dinner party among friends. The movie features (spoiler alert) an interesting mechanic: if one of the guests leaves the host’s house, walks to the street, and returns, the house will be one of potentially infinite houses in infinite parallel universes. With potential to e.g. meet a parallel version of yourself.

We discussed this at length. The premise wasn’t particularly complicated, it was just so interesting. What are the limits of this system? What are the risks? Most importantly, how could we game it?

The bit about meeting yourself was most salient. Because there’s a catch: You can only meet a version of yourself if one of you decides to walk to the street, and one of you decides to stay. If all versions of you leave the house, when you come back you may find that each version of you has simply traded houses with another version of yourself. Yes, there might be some collisions, but very many of you will simply return to a parallel universe, and not necessarily know it (the movie is pretty tame when it comes to parallels: A hand of cards might change, but the world is still recognizable).

So eventually we asked: Could you theoretically create an army composed entirely of clones of yourself? And if so, what would be the best way to do so? We hypothesized that the infinite versions of you behave differently based on some internal ‘likelihood’. Like, you might decide to have seconds after dinner, but it’s unlikely that you’d murder the dinner-party-goers. Maybe one in a million do, but forget about that. We restricted our thought experiment to a finite but militia-ish number. Say, 1000 clones or so.

We played around with random assignments of leaders. For instance, when the infinite versions of you realize this is happening, maybe you all immediately get a deck of cards, shuffle it, and the first one of you to deal yourself a sufficiently unlikely hand immediately runs outside. The rest stay put. Then it’s incredibly likely that they will return and find themselves paired up. You can repeat this, the pair travelling as a group on the next round.

The issue with this approach is that at the end, when there are just two huge groups, it will take a monumentally long time for them to meet up: They have to first draw the improbably hand, go outside, and then luck out by being assigned to the one house where the other half-army is. If this happens, you have to start dealing hands again.

Eventually we decided that a simple coin flip was best (please email me if you think otherwise!). This perfectly balanced the likelihood of leaving with the likelihood of staying. Any other strategy, we reasoned, made it too likely that parties would be moving past each other by all leaving the house or just staying put too long.

We discussed the above and many other parallel house possibilities that night, but the Army of Me scenario held our attention longest. And after we worked out the rough idea, we realized it was reasonably easy to whip up a Python script simulating the scenario, which we did the same night (I emailed myself a copy at 1:45am, surely more hours after we stopped watching than the runtime of the film). Here’s the script:

A “blunder” by our definition is any time a party leaves the house and returns to an empty house. Theat group wanted to meet up with another group, but wasted a journey to nowhere. You can see below that for one run of the 32 houses version (hardly even a hunting party if we’re being honest) there were 125 such blunders:

I wish I had some statistical analysis on what the average blunders are as a function of n (and also more confidence in the correctness of our code) but alas, eventually we needed to sleep.

I hope this example captures the sort of supernova of ideas I alluded to before. What began with “I heard about this cool movie” turned into a puzzle that we never would have thought of otherwise. The movie itself became basically insignificant to the night: the entire fun of the evening came from our exploration of the film’s ideas after the fact. Which, kudos to the movie for being able to achieve, but my point is that the relationship between film and explosion of intrigue after the fact is not necessary. It is entirely possible to watch the movie, remark “Wow, total mindbender. Anyways…” and totally forgo all of the richness waiting to be discovered.

I understand that asking people to stay up for hours recreationally coding and solving problems that cannot be instantiated in reality is a big ask, but it explains my internal wince when I try to bring something to the table, and it is summarily tossed aside before conversation shifts to more important things (like e.g. something that bothered you for fifteen minutes like last week). The opportunity cost is palpable. I have thankfully done a reasonably good job surrounding myself with people who, like Stefan, create places to explore.

“...Are your precious things so different?” - Izoni

The second example is far less technical and far more silly. It is the story of Ratlar.

One night, after an evening of playing Apex Legends, we were discussing a character named Gibraltar. For one reason or another, we decided to reverse the word, and came up with Ratlarbig. We further decided — though ‘decisions’ in these contexts are more like word-salad sand-boxing — that Ratlargib was funnier than Ratlarbig, and we started playing with the former: Gib was reasonably close to a mispronunciation of Give, and in fact this had been a long running joke among my sisters and I, that is requesting something by simply saying gib. Don’t ask.

From these simple facts was born Ratlar! We imagine Ratlar as a lesser character in a fantasy world, perfect for e.g. D&D or a whimsical children’s adventure, emerging from a hole in the wall improbably far from home. Ratlar wants only to receive gifts, and will communicate this using a single word (you guessed it). It was late in the night, and all of this was side-splitting (it’s okay, you don’t have to laugh).

Our favourite bit, however, was that we both agreed that Ratlar, despite his name, didn’t have to be a rat. Just as someone named Matthew does not have to be a mat, Ratlar was not necessarily going to be a rat. Which is to say his ratness was contingent. We arrived at something like axioms for Ratlar’s existence:

1. Ratlar wants you to gib him things.

2. Ratlar didn’t have to be a rat, but is.

3. Ratlar could not have been a monkey.

This last bullet may come as a surprise, but something of how Ratlar came to be just made it really clear that, while he needn’t have been a rat, he could NOT have been a monkey. A lizard, or a small wild-cat perhaps, but not a monkey. And also just to be clear here he is indeed a rat (but he didn’t HAVE to be a rat is all we’re saying, given the understanding that never in a million years could he have been a monkey). Also you ought to hand over whatever you’re holding. Gib, in other word.

The above is cute, and fun, and may seem totally pointless. It isn’t however, and the exact way it isn’t is really the reason I am writing this essay/tribute to one of the best friends I’ve ever had.

In December of 2020, I made the decision to not visit my family over the holidays. Alberta was still in the tight grip of the pandemic, and travelling felt somewhat irresponsible. I normally look forward to the 2ish weeks I get to spend in my frozen-over hometown as it’s one of maybe 2 times a year I get to see my aging parents, and the only time I get to see old high-school era friends which each year scatter farther to the wind. Not to mention getting myself out of my apartment which was feeling smaller by the day.

I was bummed. But my family and I made the best of it, skyping for most of Christmas day, and my girlfriend’s parents were extremely hospitable in inviting me to their own family traditions. I was grateful to be able to still feel connected to people despite the distance, and to feel accepted by my partner’s family for what is often the most intimate time of year.

And to top it off, there was also a special Christmas visitor on the morning of December 25…

Don’t ask Stefan about the immense effort and kilometerage it took to actually find a stuffed rat: Turns out they are not a popular species among stuffed animals. And oh yeah, the Ratlar label is sewn by hand.

On top of being just incredibly thoughtful, and way beyond the call of duty for making Christmas special, Ratlar also represents the sheer value that can be produced from something as simple as the word ‘Gibraltar’. “What is the point?” Seeing Ratlar on Christmas morning brought me pure joy. It showed both Stefan’s incredible kindness and his commitment to the bit. Ratlar transcended the joke, and has become a permanent fixture on the far end of one of the aforementioned couches. The point is that these conversations, while superficially frivolous, are literally productive. ‘Gibraltar’has humorous connotations now. Rats, a special significance. From these supposedly frivolous conversations spring joie de vivre, opportunities for connection. Laughter. Why else are we alive, if not for moments like these?

The Wizard of the Coast

I hope now that you have a feel for what I mean. The above two examples are among countless others. Every week we would find ourselves miles away from worldly considerations. It was never planned, but it never failed. And with consistency came momentum: because we spent so long analyzing, talking about talking, and critically engaging, we ended up developing a short of shared language (maybe more like a dialect, or accent even).

To be sure, much of this was due to our shared computing science education. In talking about real world — that is to say, not programming-related — phenomena, we’d often employ the language of our trade: Weighted sums, Big O runtime, A* Search, Binary Trees, Priority Queues, the Pigeonhole Principle, Back-propagation. There are surely others I am forgetting. With these words we would tackle the problems of the world (or our worlds) as though they were word puzzles.

Here’s a simple example: The Isoperimetric Theorem for Rectangles states that for all rectangles of a given perimeter a square maximizes the area. We both knew this, applied it dozens of times as we discussed things. Our analogies are probably a stretch, but the idea is roughly “If you have two quantities for whom a tradeoff exists, setting them equal to each other is often the way to maximize the output value”. This isn’t strictly true everywhere, and is far from a mathematical proof, but it’s a pretty good heuristic. All of the above ideas from CS are this: heuristics we picked up, tools in the toolbelt.

The great part of these heuristics is that they provided us a way forward. Sometimes they aided in solving actual problems — no doubt a boon to our first Software Engineering Interviews, which praise be to Richard Stallman are in the past now — but they also aided in the casual, fun, puzzle-ish sorts of conversations like the examples given above. For instance, “Try it backwards” gives you Ratlarbig. “Try setting the quantities to the same value” gives you the strategy for creating the army of clones by making it equally likely that you stay or leave. It’s not like we’d be sitting there saying “according to the Isoperimetric Theorem…” as we pushed our spectacles up our noses, but in a eureka moment it would be common for one of us to exclaim “Oh dude it’s the square area thing!”. Whenever we found something machine-like (a word, a mechanically interesting film, a social dynamic at a dinner party, any system, really), we’d try hitting it with all our tools to see if anything interesting happened.

Stefan, master of systems, was constantly coming to me with novel heuristics. I relied primarily on my computing science education, but Stefan has been learning how to optimize games of all sorts since he was a child. Magic: the Gathering was an obvious lightning rod for his attention. D&D another. And through them, he has generated broadly applicable insights that to me feel like a sage coming down from the mountain with a list of maxims for life.

Magic is reasonably complicated, so I’ll give a brief overview. You build a deck of 60 cards. These cards are broken down into mana cards, creature cards, and sorcery cards. There are others, but for the layperson thinking of the game as composed of resourcey things, monstery things, and magicky things is sufficient. By using the mana to cast creatures and sorceries (&c.), you endeavour to deplete your opponent’s 20 starting health before they deplete yours.

With about 20,000 unique Magic Cards, the possibilities for a 60 card deck are, for a mortal human, literally endless. And while the above gameplay sounds relatively straightforward, it is anything but. The magic of Magic is in the synergy. In interlocking gears. Cast one spell and two spells happen. Every time x happens, y happens, which makes z happen twice, which makes x happen again. And so on. Can you attack by being defensive? Can you twiddle a knob over here, and cause a mountain to fall over there? The answer lies entirely in your creativity as a deck-builder.

The upshot of Magic’s complexity is that it is far from a solved game. Unlike tic-tac-toe, there is not one clear strategy. As a result, being good at the game requires having good heuristics, and given our shared interest in systems, I was always happy to listen as Stefan detailed some specific use-case of x card, or some winning strategy he had employed. I ended up playing a lot of Magic with Stefan.

Where my previous examples of our conversations have been of the ‘Down and In’ variety, the heuristics Stefan ported over from Magic — and D&D to a lesser extent — were more of the ‘Up and Out’ variety: I still think of Stefan’s maxims frequently in my daily life.

When we’d discuss these maxims, we’d strive to generalize their underlying principles. And even if it felt like a stretch, every time we got to proclaim “Oh dude it’s the _____ thing!”, we felt like we were pulling back the curtain on the Wizard of Oz, finding a pattern that had so far gone unnoticed.

For your reading pleasure, and at least partially for posterity, here are some of my faves:

1. Don’t be Afraid to Spend Health

Stefan once pointed this out as something new Magic players often get wrong. Since depleting your health is typically how your opponent will beat you, new players over-prioritize keeping their health high. But there is no benefit in Magic to winning with 17 health instead of 1. Both count equivalently as wins. Further, there are many cards that let you spend health as a resource: you can use it to toughen up your creatures, cast more powerful sorceries, draw cards, and otherwise interact with any of the game’s other mechanics.

Which is to say there are often opportunities to turn the tide of a game by getting yourself dangerously close to death. Let’s suppose your opponent threatens to win the next turn, by dealing you 15 damage. You have 10 health left. Do you spend 3 of your health drawing cards in the hope of drawing the card that will win you the game? Of course! This example is unequivocal, but you can imagine less clear situations. Perhaps mid-game, you can spend a few units of life to make your opponent discard some cards, achieving tempo for yourself. Do you do it? The point of Stefan’s advice is that you shouldn’t immediately answer no, but a lot of new players do. Health should be treated like a resource, because it is a resource. If you’re willing to spend it to get ahead, and your opponent is too afraid to, you probably have the advantage. This isn’t to say you will necessarily win, but you have more options at your disposal, so you will win more on average (see maxim 2 below).

And broadening up and out, Stefan’s advice made me question: Am I fully exploiting the resources at hand? Am I even recognizing them as resources? There is a nice parallel here with entrepreneurs: Many of us fear debt as life-ruining. But if you’re willing to incur some debt, you might be able to pull ahead long term. Ditto with being willing to incur nervousness and the potential for rejection in asking someone out on a date. If you can exchange it for value, it is a resource.

The Pixar movie Onward summarized this nicely as “...on a quest, you have to use what you've got.” Figure out what you’ve got, and spend it. If you lose with surplus value, you probably weren’t spending your resources aggressively enough. A win with an empty hand and one health is a win.

2. The Best Deck in the World has a 60% Win Rate

Right now, the best competitive deck has a win rate of 58.6%. Years ago, Stefan had pointed out this interesting anomaly: You’d expect that the best decks in the world would have close to a 100% win rate. Surely that’s the mark of “goodness”.

This is not the case. People will find answers to extremely powerful decks and learn to deal with them. If a deck becomes too popular, a counterculture of decks will emerge, specifically designed to be able to answer the strong deck. The best decks just have extremely strong synergy, excellent high-value (and cost) cards, and are perhaps less error-prone to play. But they are not unstoppable.

Generalizing this advice: To be the best, or even just competitive, you do not need to win every time. You just need to win on average. It is enough to make good bets that mostly — that is to say with higher frequency than a coin flip — pay off. If you repeatedly make these good bets, and try not to get too disappointed when they fail, you will achieve exceptional value for yourself.

I thought about this fact a lot when I was applying for jobs during my final year of University. I had studied very hard for my degree, and was pursuing my dream job at Google. I had great grades, a strong resume, and I had spent hundreds of hours preparing for the notoriously challenging tech interviews. Things looked promising. But despite all this, there was the underlying worry: what if I don’t win this time?

This worry was unfounded. It didn’t matter whether I got the job or not. I had already designed a good deck for myself that on average worked. If I didn’t get the specific position I wanted, this was absolutely not evidence that I wouldn’t be hired at some other excellent work place. Of course studying, working hard, preparing, and practicing are good bets. Of course I could only do my best. As it turns out, I was rejected by a handful of companies before being hired at Google. Was this evidence that my methodology was wrong, that I had built a poor deck? No, clearly.

This applies more broadly, as well. Consider from the perspective of a company deciding who to hire, or a government deciding which legislation to pass, or a couple deciding whether to stay together. In each case, we may incorrectly believe that there is a room of super-geniuses who are carefully planning a scheme that is certain to work. This, I am confident asserting, is not the case. As long as the company tends to hire candidates who provide more value than they cost, they will be profitable. As long as the government tends to pass legislation that improves the nation, the nation will improve. As long as the couple tends to make each other feel happy and secure, the relationship will flourish.

The best deck in the world has a 60% win rate. You don’t need to win every time. If you move forward 60% of the time, you are still moving forward. You just need to make good bets.

3. Don’t Have More than 60 Cards

Technically, Magic has no limit to the number of cards you can include in a deck. Despite this, virtually everyone who plays limits themselves to 60 cards (for the Magic-playing readers, forget about Commander for a moment).

60 cards is sort of a magic (hah) number. If you balance your mana correctly, you will have enough resources to afford the things you want to do and, critically, a small enough deck to be able to know it well.

Some of the most satisfying moments I have seen in Magic have been when one player is deeply behind, and they spend everything they’ve got (see maxim 1) trying to yank some miracle out of their deck. Sometimes they pull it off, and sometimes they don’t, but if you make this bet well (see maxim 2) you can often achieve from-behind victories.

Every additional card makes it hard to predict these moments. Let me give you an example: Suppose it’s late in the game and you have 5 cards remaining in your deck. There is one card — let’s say Day of Judgment — that will turn the tide in your favour. And let’s suppose that you find a way to draw 3 cards this turn. In this case, you are more likely than not to get the card you need. But if you had another 20 cards in your deck, the probability is much lower. Yes, maybe those other 20 cards are also big fancy resets that you want at that exact moment, but what about the rest of the time? You’ll find yourself sitting with an armful of bazookas when perhaps you need a bandage.

Yes, I can imagine a universe in which the format of Magic calls for 70 cards, and everyone plays 70 cards and just needs to learn that number instead of 60. But the point of this maxim is that you don’t want your deck to be less predictable than anyone else;s. When I think of this maxim, I tend to think of it as a call for minimalism: A complex life with a lot of moving parts might make you think you can do more, but you will spend so much time juggling the complexity that you end up doing less. Having a lot of cool surplus cards might, perversely, mean you never get to do the cool combos you want to do because it’s harder to get all the component parts in your hand.

From my own life, I have a few analogues. First is from my childhood, when I first began playing World of Warcraft. When I played games as a kid, I often gravitated towards high complexity characters because they provided an interesting cognitive challenge to play.  With a lot of moving parts and things to balance, playing the characters well took focus and even creativity. In World of Warcraft, I was thus drawn to the Druid class.

The problem with this was that it was hard for me to actually progress in the game. I couldn’t really wrap my head around the playstyle, and all the cool parts of being a Druid were locked behind actually knowing and executing the tricks of the class. I remember watching beefier, sword-and-shield-wielding characters flying by me and stomping sections of the game that I couldn’t pass seemingly effortlessly. It was frustrating, because my own personal tenets (see maxim 7) informed me that I should do what I want, not play to win.

However I quickly realized that what I wanted more than being a Druid was to actually explore the world and play the game. I made a new Orc Warrior character — basically the most dumbass run-in-and-hit-it-with-an-axe character I could fathom — and restarted my progress. And had way more fun.

I followed an almost identical line of thought when I decided against pursuing a PhD. I had the grades, had published research in an international conference, and had even written the GRE. I had told my Google recruiter after my internship that I would not be returning for full time work despite my eligibility because I was going to be pursuing postgraduate studies.

But when I returned home from my summer in California, and began researching all the labs I might be joining, I considered the Druid. I considered the 60 card decks. A PhD incurs massive complexity, and at considerable cost of your ability to actually do things in the world like e.g. buy a new car, work sensible hours, predict which coast you will live on more than a year in advance. Would I really want to be reading thousands of pages of research papers while my friends were flying by me with epic gear, exploring parts of the map I wouldn’t reach for a decade?

I realized that I was romanticising complexity: Doing the hard, complicated, sophisticated thing would fulfill me in some narrow way, and deplete me in others. I emailed my recruiter asking for an interview with the utmost haste.

The final instance comes from a piece of advice I received as an intern. The conventional wisdom I had heard about commenting code was that if your code did anything sophisticated or un-intuitive, you should leave an explanatory comment. As an intern my manager told me “If your code is simple enough, you won’t need comments”. By applying this I produce code that is easier to read, and thus easier to maintain, and thus easier to extend.

The generalized point here is that paradoxically, simplicity lets you do more.

4. Math is for Blockers

This one’s a little more tongue-in-cheek than the rest. In Magic, there is a turn-taking structure to combat. The active player announces they will be attacking with some number of creatures. The opposing player then announces if they would like to “block” some number of these attacking creatures with some number of their creatures. Any damage that hits a creature will harm and potentially kill it, but the damage will not hit your health. Any unblocked damage will hit you directly, decreasing your health total.

“Math is for blockers” refers to the fact that the non-active player has to decide how to block the incoming attack. It’s a bit of a joke in magic when the active player finds themselves attacking with a complicated mess of powerful creatures, weaker creatures, armies of tiny creatures etc. The attacking player gets to just say ‘I’m attacking with everything”. The defending player has to carefully calculate how best to distribute the damage between creatures and themselves — why not block everything? See maxim 1 — to maximize the likelihood of them staying in the game. If you have enough damage hurtling towards your opponent, you force them into a situation where they need to be very mindful in how they respond, which obviously can be error prone.

Here’s a fun, related quote from the official Magic blog:

I hope the analogues to reality are obvious: avoid getting yourself into situations where you lose control and need to fiddle with small algebra to make it out alive. Fate favours the bold. Attack each turn if able.

5. Couple Flavour and Function

Magic would be nothing without its flavour. The gameplay itself is fun, but what the game absolutely nails is the linking of the aesthetics of the game with the mechanics. This is synergy of another kind. Where — horizontally, let’s say — the game encourages synergy by allowing players to chain together card effects, it also strives for vertical synergy: cards whose art looks a certain way should also behave a certain way.

For instance, with no knowledge of the game, what kind of effects do you think a big wooly mammoth creature would have? Probably nothing subtle, probably nothing magical (unless it also e.g. had a huge glowing crystal around its neck). Probably it’s got strong defense, decent attack, and maybe a resistance to cold or something. And in fact this is approximately what we see. I hadn’t looked up this card ahead of time, but there does indeed exist a card named Woolly Mammoths that gains a special ability if you control a snow resource. The attack turns out to be better than the defense, but whatever. The point is that you probably have an intuition for what a card does by its art alone. Guess what a Vampire does.

Most good video games will do this: Bigger characters are slower, poison causes damage over time etc. But Magic takes this to the extreme. It does not just apply flavour where convenient, but actually seeks it out. Magic cards occasionally include explicit flavour text on the card itself: this text has no bearing on the actual gameplay, but seeks to create a sort of atmosphere or narrative for the cards. Some of the best versions of the vertical cohesion I’m talking about comes from these.

Trading cards do not have the sound and movement afforded by movies, nor the word count afforded by books. They have to tell stories in the limited way they can: Not using the art, or the card’s name, or the flavour text to tell a story is akin to not spending their resources (see maxim 1). But of course they do. Check out Progenitus:

To play Progenitus, you must pay two colours of each of the 5 colours of mana in Magic. This is a ridiculous expense, and typically impossible as most Magic players build their decks with one or two colours (something of an extension of maxim 3: more colours are harder).

But further to this Progenitus is a hydra with 5 heads. They could have gone with more or less heads and the card’s rules wouldn’t have changed at all, but of course 5 is the best: It has one head for each colour of magic.

Then let’s consider the rules: it has protection from everything (meaning it can’t be blocked, targetted, or dealt damage), and if somehow the opponent manages to get rid of the damn thing, it is instead shuffled into its owner’s library. Much like a hydra, remaining alive after it’s superficially killed.

And the text itself further implies this near indestructible intangibility with “The Soul of the World has returned”. It isn’t merely a polycephalic lizard, but some intrinsic part of the world itself, which helps justify its protection from everything. And, like a Hydra, “returns” (probably from the deck you just shuffled it into).

Further, check out its type: It is a “Legendary Creature”, where ‘legendary’ means that your deck can only have one copy of the card: sensible for the soul of the world. It is also a hydra avatar, not merely a hydra. Again, sensible if it is the embodiment of the world itself, and not just a big reptile.

Finally, check out its strength and toughness, in the bottom right corner: It has ten strength and ten toughness, further synergy with the 10 mana you paid to play it.

For another fun example, consider Feldon of the Third Path. The card has an ability that lets you copy a creature from your graveyard. When you do so, it is imbued, on top of its other types, with the type “artifact”. Feldon’s flavour text reads “She will come back to me”. These six words, in conjunction with the rules and art, make it pretty clear what Feldon is trying to achieve with his artifactual copies. Observe also that the story would be incomplete if any of the three was removed. One could imagine a tortured asylum-dwelling creature with identical text and rules, in which case the narrative would be sinister and deluded rather than melancholic.

What’s great about this type of card whose narrative is intermingled with the actual gameplay is that it encourages ludonarratives: Stefan introduced this concept to me and it is to me one of the most fun forms of storytelling.

Rather than being told explicitly, a ludonarrative is told implicitly via gameplay. For instance, the flavor of Feldon pretty strongly suggests that Feldon is pining after a lost love. But nothing in the rules states that you can only copy a (fictional) card called Feldon’s Wife or something.

This means there is space for some very fun narratives to arise out of typical gameplay: Consider if you created a copy of Vindictive Mob. As the mob enters the battlefield, you might be forced to sacrifice Feldon per Vindictive Mob’s rules. You end up with a tight little narrative of Feldon copying a whole mob of people in pursuit of creating a synthetic version of his wife, and the mob that he copies is none too pleased about being copied: they’re Vindictive, even. The result is that they kill Feldon himself (as implied by the ‘sacrifice’ action). This is of course just some plausible example I found by looking at a handful of Magic Cards. But consider the space of all possible creatures you might copy, and all their attendant stories. Then consider all possible ludonarratives the rules of Magic imply with the rest of their 20,000 cards and you can see the incredible depth of the game. 

When playing Magic it’s probable that you will bring about scenarios that have never happened on earth. Part of the joy of Magic is finding these stories as you play: sometimes, you power up your weakling goblin until he can defend against the most powerful of dragon attackers. These wins become more than “just wins” because they imbue the cards with a narrative that transcends even the flavour that was explicitly embedded in the cards. Even a a valiant loss here would be satisfying (see maxim 6).

There are two important takeaways here. One, as creators, we should strive to tell stories using everything at our disposal. Words are good, but the tighter the form matches the function, the more likely it is that our story lands in the heads of our audience. Second, as the audience, we should strive to read stories into everything. By doing so, we can enrich even the most pedestrian of situations. Consider the difference between waiting in an airport for 5 hours because your flight was cancelled, and spending 5 hours devising and then executing a scavenger hunt in the airport’s myriad shops and kiosks.

6. The Rule of Cool / Meta Decks are Boring

I first encountered this advice via Warhammer 40k (could we get any nerdier?). The Rule of Cool states that in the choice between a winning strategy and a cool strategy, one should always lean for the cool one. Often, these are one and the same — hey, we all like to win — but critically to Warhammer and Magic, they aren’t always.

In Warhammer 40k, I am a Tau player. The Tau are an alien race, unofficially dubbed ‘Space Communists’, who are known for their prowess with firearms and gundam-ish mecha suits. Unfortunately according to the actual rules of Warhammer, Tau are not particularly effective at shooting (note here the violation of maxim 5), and in fact are an all round not-strong army to play with. A viable army list for Tau includes 3 huge models that are, under the current rules, the most powerful among the Tau options. This is a good way to win, but not a good way to have fun modelling a variety of units with a flavourful military structure and unique gameplay.

Further, there is the option for one battlesuit to choose between an extremely sleek, building-sized railgun, or chunky missile-pod fists that to my eye look encumbering and kind of ogreish in a bad way. I was disappointed to discover that the missile-pods were strictly better for gameplay. But then someone referred to the loadout with the railgun as the ‘sniper boss’ loadout and I realized, “Rule of cool! What am I thinking? Railgun forever!”. So if you want to win, you probably won’t choose Tau, and if you do choose Tau, you probably won’t choose the units I’ve chosen. 

Stefan has faced similar conundrums. In Magic and Hearthstone, a big part of the game is creating your own decks. But in order to be competitive, many people just Google the best decks (‘Meta Decks’), buy these cards, and then play them until they are no longer competitive. For a lot of people winning is the point.

But for Stefan and I (both Johnnys through and through), the game is about self-expression. If we manage to pull off something interesting and lose, that’s still a “victory” in that it was more fun than winning with a deck we don’t care about.

This maxim isn’t meant to convince you to be a Johnny. It’s intended to tell you that it’s okay to lose. A win doesn’t need to come explicitly. After all, why do we play games if not to have fun? Personally, I have no memory of the specific wins I have had against Stefan. I do however have memory of the most fascinating, mind-bending combos that arose while we played. Once we were both gaining so much health each turn that we ended up putting down our hands and calculating the winner based on the possible outcomes of each of our decks. I have watched Stefan keep a losing player alive in three person Magic because they were about to do something interesting that would have made the game more exciting. Is this the “wrong” way to play? If it maximizes enjoyment, my sense is that it’s the better way to play.

And if you are a Timmy (that is, a player that values winning above expression), I encourage you to at least recognize that other players are not.

This perspective bleeds into Stefan’s social encounters, and once again, contributed to our deep bond. Certain computer scientists, trained in the art of attention to extreme, semicolon-level detail, have a bad habit of publicly correcting others. This, I believe, is a form of “winning”: Too often I have watched other CS students breach all social conventions to point out that “actually this passing detail of something you just said is wrong in some subtle and insignificant way”. I’m not making the claim that all Timmys are like this, but I think there is probably a lot of overlap between people who value winning over self-expression and people who value factual accuracy over social graces.

Stefan on the other hand understands that, like games, conversation is not about scoring points. It doesn’t matter if you look the smartest or the funniest. What matters is that everyone extracts as much meaning out of the conversation as their respective decks allow. Allow your interlocutors to perform the cool combo. If suspending disbelief is better for the flow of conversation, suspend disbelief.

Stefan and I frequently bring up horoscopes as an example. As pretty strict materialists, neither of us believe in horoscopes. But here is a dialogue we hope to avoid:

“What’s your sign?”

“I don’t believe in horoscopes”

“Oh”.

Very fun. Consider an alternative:

“What’s your sign?”

“Gemini. You?”

...At which point they say their piece and get to show off what they know about horoscopes, which is surely their reason for bringing up horoscopes, to which at the appropriate time you get to follow up with:

“If you could have a different sign, what sign would you be?”

or
“What do you think are the best and worst sign pairing for romantic partners?”

or

“Do you have canonical traits or people you associate with each sign?”

etc.

In other words, who gives a shit if you don’t believe in horoscopes? Unless you’re bringing it up to really engage in a way that’s unlike the other dozen times you’ve brought it up, it basically just kills the other person’s fun. When there are so many opportunities for truly learning about someone and connecting and recognizing their humanity despite your differences, and simply experiencing the joy of shared exploration of a theme, why opt for the obvious?

Here’s one more instance, from Dungeons & Dragons. D&D allows for things called “quirks” and “weaknesses”. In traditional D&D, these mostly impact the result of dice rolls. For instance if your character has the ‘Dishonest’, you have a +2 bonus to Deception checks but disadvantage on Persuasion checks.

Stefan takes this a step further, and asks his players to bake more critical flaws into their characters. This might be an extreme fear of heights, or a severe allergy to some common spice, or a blinding hatred of a specific faction. Strictly speaking, this makes all the characters weaker: They are less likely to achieve their goals. But the value of these constraints is that they force the players to be more creative, and to roleplay more honestly. They have to get in the heads of their characters: If your character is acrophobic, crossing that mountain suspension bridge might pose a serious problem. Now the party has to come up with a novel solution where previously they might have just walked across. Further, they start to consider psychological solutions. The narrative becomes more flavorful because the solution is suddenly personal to the character (see maxim 5).

This maxim might seem to conflict with previous maxims that suggest strategies for winning: If the point isn’t winning, why bother with 60 cards? Why bother with using what you’ve got? What this maxim proposes isn’t throwing every game, or being so totally phatic that you never say what you think, or handicapping yourself so wholly that it’s impossible for you to win. This maxim suggests that you broaden your definition of what counts as a win. 

7. Let People Foresee Their Deaths

Stefan, as a Dungeon Master, often finds himself in control of the fates of the characters that his friends role play. A hard decision in D&D is when to kill a character, since a campaign can last months or even years. When a player has role played a character for that long, they understandably become quite attached. Still, there should be stakes. If it’s impossible for the characters to die, why bother sneaking around? Why bother with armour?

This is Stefan’s guiding principle for killing characters: The players who control them must be able to reasonably foresee their own deaths, and choose to take the risk anyway. So if they are walking on a mountain pass, no random boulders that unexpectedly come crashing down and crush a party member.

His canonical example was from a session where a horde of enemies were swarming the party. One of the party members had an ability that, with a tiny fractional chance, might have been able to eliminate the hoard all at once (maxim 1!!). If it failed he would be totally overwhelmed, but even so he would probably slow the onslaught enough to allow the party to escape. The character went through with it, rolled wrong, and was promptly killed.

The player here should have no reason to be upset: They feel massive loss because they took a massive risk (see maxim 5). And in fact, this little story adds to the experience of D&D (see maxim 6). This very real feeling of having tried something heroic and paid dearly for it is a testament to the fantastical power of role playing. Emotions need not be fictionalized.

The maxim, stated alternatively, says you should let people own their own deaths. This makes the deaths meaningful, and charged with opportunity for learning.

Broadly, I strive to treat my loved ones this way. Not killing them, to be sure, but making sure I am only angry with them when they could have foreseen that what they were doing was harmful. I try to give people the benefit of the doubt: More often than not people wrong each other due to lack of focus, not malice, and who among us is not deeply familiar with the noise of the world? Give people the opportunity to make a better play, and they often do.

If the players are never punished for mistakes, the game becomes meaningless. Anything goes, and no decision is better than any other. But if the players are punished when they play well, then they become frustrated and won’t want to play at all. In other words, there have got to be stakes, but they should be fair.

When you apply this maxim as broadly as you possibly can, you find that the complaining grinds to a halt, the blaming evaporates, the mood lightens and the connection strengthens. You see that some people have different dice than you, and are worse at estimating outcomes. You forgive what you previously couldn’t. You cease to sweat the small stuff.

Not once have Stefan and I sincerely argued. Not once.

Roll for Charisma

Stefan and I love Magic because Magic is fun: It has infinite opportunity for stories, for intrigue, for conversation starters, and for algorithmic combos that we’ve never even imagined, let alone seen instantiated. The game features artisanal art that coheres with its unique rules that cohere with what Stefan and I value far beyond the confines of Magic.

Every conversation is a system. Each day a game. Each situation an opportunity for narrative, bursting into reality out of thin air. Each relationship a push and pull of values and dice rolls. And if you know how to play, these too can be endlessly fun.

The hard part is trying. It’s in the motivation. It’s in averting your eyes from whatever lights and sound are drawing your attention away from the depth of beauty that exists inside each and every human mind. Neither Stefan nor I can keep it up without rest, but we have found that having a partner on the road is easier. Neither of us have spent the last 5 years in unceasing happiness: occasionally, it has felt like the opposite. But we have found that the maxims have helped. We have found that we could drop an Alka-Seltzer in the glass of the other’s mind, and suddenly cause them to come alive, remembering that yes, there are things to say, there is more to life than slowly incrementing societal measures of self-worth.

We have neither washing machine, nor dishwasher. We live far from downtown, in a rundown apartment building. We’ve spent the last year and a half — years longer, if we’re being honest — locked away, watching the world succumb yet farther into a miasmic fog around us. But our deteriorating apartment and the chaos around it disappeared completely as we stood near the sink, pondering puzzles, conjuring the best medicine.

Stefan once quoted “And Alexander wept, seeing as he had no more worlds to conquer” and “Born too late to explore earth, born too early to explore the galaxy” and remarked, “Haven’t these people ever heard about depth?”

Use what you’ve got.

Infinite Mana

Two Wednesdays ago, Stefan brought the (fine I’ll admit it, very good and exciting) news that he had accepted a job in a city other than this one. In January 2022, I myself will be moving to California. Although the pandemic afforded us another year of laughter, the news that he would be leaving hit me harder than I imagined. It is not often that I cry, but here we are. Like a ship in a bottle, I can only approximate the wonder of what once was at sea, the vessel that we sailed between undiscovered islands and untouched shores, collecting gemstones and precious metals. I hope this fragment makes you long for the open expanse. I hope that it captures the message that, for so many years, Stefan and I have worked tirelessly to develop: there is life beyond the veil, hiding in the words, dodging around the corner while you’re distracted by the news, in the kitchen, down the hall. It’s not where you think, but it’s not out of reach. Quick, before it gets away.

Stefan, I will miss you forever.

But how lucky we are to have something we’ll miss.