You Have to Weigh It — Ryan Condon

Context for Patreon: This is a post I made on my personal blog last year - It does not contain significant MTG strategy advice, and is much more personal in content. I wanted to share it here because it does touch on Magic, & I'm also just quite proud of it as a piece. When I say things like "I think this draft pick is close enough to not be worth thinking about too hard", this post is what I mean. Also going to give a content warning for suicide & transphobia, which comes up about 20% of the way in

There's been a revolution over the past few years in the Magic: The Gathering Limited community. This revolution has come in the form of data. Riveting stuff, I know. I'm sure half the people reading this couldn't care less, the other half already are familiar with everything I'm about to explain. Stick around though -- I promise I'll try to make it worth it.

First a bit of background for the non-nerds out there, which rest assured, will be brief and simple. Magic: The Gathering is a strategy card game typically played between two players. The fun part about this game is that the pieces (cards) can be different each time you play. Each player chooses which cards they put into their deck, which adds an entire extra layer of strategy. There are two major ways to play Magic - "Constructed", in which players are allowed to plan ahead and bring a deck using any cards they want, and "Limited", where players need to improvise and build their deck by selecting from a random set of cards "on the spot".

My favorite way to play Magic is Limited. I love the improvisational nature of limited deckbuilding -- You can prepare, but you also have to react. It's a bit like being an astronaut going on a mission to explore a new planet. You can collect information about your destination, know the broad details -- climate, gravity, geography, hazards, life forms -- But you also must train your general physical and mental fitness, be prepared to react and adapt to whatever might come your way. When a crewmember gets sick with a mysterious alien disease lightyears away from home (or you encounter a powerful rare card that doesn't fit the deck you'd previously been drafting), the playbook gets thrown out. All that's left is you and the situation. Every new list of cards to play with is like a new planet, ready to be explored and settled.

Magic used to be mainly a "tabletop" game, played literally across a table between two fleshy humans. Nowadays, as with a lot of games, the majority of Magic I play takes place in data packets sent between my laptop and someone else's. I miss the days of meeting lovely, smart and kind folks at in-person tournaments (which I still get to do occasionally), but there are benefits to the digital side of things. Today, when I play a game of digital Magic, there is an invisible spectator watching everything I and my opponent do. It goes by the name of 17lands.com (a reference to an old convention of Limited deckbuilding), and with my consent, my games are streamed into a pool with hundreds of thousands of others. In exchange, I have access to the results of that pool.

I can know things like:

I have played 2,857 games of digital Magic since September 2021

I've won 61.5% of those games

I play fewer than 2 "rare" cards per deck

There is a slight negative correlation between win rate and rares per draft overall (R^2 = 0.0092)

Still, 17lands thinks I "might be taking it too far and want to draft more rares"

17lands, which came out in early 2021, has taken the Magic limited community by storm. There was plenty of resistance. As silly as it sounds, we care about our little cardgame, to a level some might describe as spiritual. Just as you can't calculate the price of a human life or rate the quality of a piece of music on a linear scale, neither can you quantify the "sweetness" of a limited deck (Magic slang that means roughly some combination of "good", "interesting", and "fun", and will have a slightly different definition whoever you ask). Some things are just more art than science.

And yet...

Life insurance has a specific price. Music reviews with scores out of 10 are some of the most popular content on the internet. And today you'd be hard-pressed to find a piece of strategy content about limited that doesn't reference 17lands data (There's even an entire podcast devoted to Limited data analysis called Magic Numbers).

Why is that? Does the success of all this quantification mean that these complicated, nuanced things can actually be captured by numbers?

By far the most popular tool on 17lands is the Card Performance Table. This table shows, for all possible cards you could see in a certain list, different metrics about when they are picked and played, and most importantly, the game results they generate. This information can be incredibly useful when deciding on which cards to pick for your deck.

Still, it's not trivial to know how to use such information. At first glance, you might think finding the best cards would be obvious... which cards have the highest success rate when you play them in the game? This games played winrate metric is certainly useful! Sort the table by it and you'll find some of the most powerful cards at the top of the list. Easy! Just build your deck out of those cards.

What you'll find when you implement such a strategy is your deck will be full of highly powerful effects, but these effects will be expensive and slow. Players running faster, cheaper cards will outmaneuver you easily. You'll get crushed. So much for brute force.

Okay, let's regroup. How about instead of the cards that win most when played in the game, we look at the cards that win most when put into a deck in the first place? Now we're talking games in deck winrate. This will factor in both how powerful the cards are, and how easy they are to play. And look at that -- Again, we're seeing some of the strongest cards at the top of the list, but this time, we've also got some cheap, easy-to-play all-stars in there. We'll build with these cards instead.

Hmmm, that didn't quite go right either. Sure, we took the strongest cards when we saw them, but we didn't end up seeing many. Seems like other players also knew those cards were the best, and so we were fighting for the same few cards with everyone else. Plus, there were some cards in our deck that really didn't feel like they pulled their own weight. Perhaps those cards simply have a high games in deck ****winrate not because they are good on their own, but because they're often seen in decks alongside other powerful cards, ones that we didn't happen to have.

Clearly, we need to add a little more nuance to our strategy. We can't just blindly draft the "best" cards -- We need to make sure our cards are all working together effectively to win the game. Here's what we'll do: Start by picking the most powerful cards, but once we have a clear direction, we can filter the table down to cards that go well with the ones we already have. Since our deck won't be chock full of the most powerful cards though, we need to make sure we'll still be able to win when we don't draw our best cards. Our cards should stand on their own, but also support each other well. Let's use a compromise metric: games in hand winrate. This will measure how powerful cards are when we have access to them, but it will also factor in how fast & simple they are to use in the game.

I'll stop there, because the last approach, a heavy emphasis on games in hand winrate, is a pretty decent approximation of the way most Limited players utilize 17lands data. The point I want to drive home however, is that this spiral of increasing nuance goes on, and on, and on towards infinity. Here are a few examples of the recurring issues you will run into no matter what procedure you use for "following the numbers":

The most valuable cards are, obviously, the ones most valued by other players, and the ones you will have to compete for

Different cards have different strengths and weaknesses in a game context, and the situations in which they are most commonly played will influence their metrics, sometimes drastically

Each deck has unique strengths and weaknesses, and will often need a particular type of card far more than any other, regardless of objective "card quality"

The way a card is played can be influenced by factors completely unrelated to its power in the game (How fun is it to play? How difficult is it to play correctly? How smart does it make you feel to use it? Is there a lot of "hype" around a particular strategy this card is used for?)

Other players are also acting on cards' data, and the results they achieve through this will directly influence the data itself!*

For those who have read this book, I believe this counts as a "strange loop", where the meaning of a particular symbol is blended between multiple layers, and "higher-level" layers can cause substantial changes in lower ones, much like how DNA acts in a cell. Safe to say, these systems can become quite dynamic and chaotic, even from very simple building blocks.

When confronted with this endless list of pitfalls, it's easy to return to the question of whether this data is truly useful at all in capturing such a deep and chaotic system. I and many other seasoned Magic players started from this skeptical stance. And yet, over the past few years, I've slowly come to realize not just that data is valuable, but also why and how.Moreover, this process has helped me synthesize quite a few lessons I've learned in completely different contexts, into something that is starting to feel like an overarching principle that I can't quite put to words. (Here is my attempt.)

So first, how is data useful for Magic players?

To put it bluntly, data might not provide correct and unambiguous strategies for success, but it does a pretty decent job of highlighting the flaws in the ones we currently have. The process that top players go through when learning how to play a new "set" (list of cards, unexplored planet) looks something like this:

Collect as much information as you reasonably can about the new cards before you play with them - Think about how they are similar to cards from the past, and in what important ways they might differ.

Begin to recognize pattens, themes, and interactions between the cards, and speculate about what these will mean for how to play the set

Compare your thoughts to the thoughts of others who you respect and trust. Modify and adapt your interpretations accordingly.

Play! Use your starting assumptions as a guide, and pay close attention to when you encounter situations that make you question those assumptions. Take note of what's working, and what's not.

Revise & refine your theories. Were you on the right track to start, or have you been hopelessly lost? Are there any new approaches that you could try?

Begin to experiment. Test your alternate approaches and give yourself more chances to be proven wrong or stumble into something useful.

Integrate and synthesize. What has been the most successful strategy so far? What avenues showed promise, and which ones became obvious traps when you attempted to implement them? Use this information to formulate your first "big-picture" view of the set.

Now comes the part where data is so useful. Not before we have been able to develop an intuitive, holistic view of things, but after. Without data, we'd have to jump right back into the cycle of theorize-implement-revise. Now we have a shortcut. This iteration has already been done by thousands of other people playing alongside us!

We have an overarching theory of the case. We formulate questions and propositions about what the data would show if our theory of the case were correct. And we check!!!

We will almost certainly encounter places where the numbers contradict parts of our theory. These discrepancies serve not as a replacement for our theory, but as a catalyst for exploration and further refinement.

Now we can jump back into playing, but with a list of points where we think our theory might be missing some part of the truth. When we encounter evidence along the way, we can compare those experiences not just to our theories, but to how the data conflicts with them.

We know that there are gaps between our theories and the true nature of this world we're exploring, and there will also be gaps between what the data indicates and the underlying truth. Instead of a single point of comparison (theories-experiences), we have multiple (theories-experiences, theories-data, experiences-data). This allows us to "triangulate" our way towards the truth as we gather more and more information.

Describing this process reminds me a lot of the parable of blind people who are trying to identify an elephant from each of their subjective experiences. Each is interacting with a different part of the animal, and is having a wildly different sensory experience. From these experiences, each draw a different conclusion. None are right. No one comes close to describing the true thing.

Having access to quality data about a particular complex system is kind of like having functioning ears in this situation. You are exposed to the full array of experiences each person is having, and can rule out a large swathe of possibilities -- But you still don't have the picture itself. If you know what an elephant is, you might be able to come to the right conclusion, but it won't be easy. The information is useless unless you have the proper knowledge and know how to apply it. Heck, even if you do, you still might end up concluding that what we have here is a scene containing a wall with a fan sitting on top of it, a spear leaning up against it, and a tree in front with a snake and some rope hanging from it.

We're going to get away from Magic soon, but before applying this to other topics, I want to really drive home what I consider to be the important takeaways. With Magic, we're talking about using data in a pretty straightforward and simple context -- You are picking from a small subset of cards (small possibility space), and you can only pick one at a a time (binary choices). A card is either in your deck or it isn't. You either win the game or lose the game (binary outcomes).

Even in this extremely narrow, rigid context, data is only, can only be, useful when you have deep, experientially-informed, nuanced intuition to accompany it. If you let the numbers take the wheel, you WILL end up with catastrophic results. There is simply no substitute for being an instinctual, vulnerable, flawed, biased and bold decision-maker if you want to succeed. You can use your experiences and the cumulative experiences of others as input to help drive this fleshy, messy, awkward and unsure machine, but you can't give up the controls -- You've got your hands on the wheel, and you don't know what you're doing, but you're turning and bobbing and weaving anyways. That is really the best option we have access to when we're trying to win games of Magic: the Gathering -- Even with the full set of information at our fingertips. To put it as simply as possible, when faced with a decision, you just have to weigh it, and allow yourself to learn later if you were wrong.

[I'm about to make an abrupt transition - CW here for suicide & transphobia]

Let's look at another elephant.

Data indicate that 82% of transgender individuals have considered killing themselves and 40% have attempted suicide, with suicidality highest among transgender youth

Pubmed study

What are we to make of this information? Before we go there, it's important to acknowledge that we are now in a vastly different context than a game of Magic: The Gathering. The possibility space is no longer small, our choices and outcomes are no longer binary. Most importantly, the stakes are significantly higher than the outcome of a children's card game. Also crucially, what was previously a mildly fascinating intellectual observation (people's interpretations of the data influence the data itself) now becomes a vivid, tangible reality that it would be foolish to ignore.

I took so much time explaining in detail the relationship between 17lands data and successful Magic: the Gathering players because I want it to be immediately obvious that this statistic, alone, is vastly impotent when it comes to helping us understand the reality of trans experience in modern life. If using data to decide which card you should pick for your deck is impossible to do without first developing an intuition through iterative and intuitive experience with the game, it obviously follows that using statistics like this one to inform our social, political, and legal treatment of trans people without a human-centered intuition informed by the lived experiences of trans people is not just morally irresponsible -- It is simply not rational. If you are interested in truly understanding the world, you cannot start with this data and build your view of the trans experience from there -- You need to start by getting to know the fuzzy, messy, noisy, varied (and beautiful) community of people who actually have enough information at their disposal to develop a theory of reality that we can compare this data to.

I'm using this statistic as an example because it's famously one that is used to argue that being trans is a painful and dangerous way to be -- and therefore we need to prevent people from living this way. I actually believe this is a view one can come to in good faith. If you are starting from a place of ignorance, as many people who don't know an (out) trans person are, beginning to formulate your views on this issue by starting with this piece of information and looking for an argument that fits the data into a coherent story about the world will leave you vulnerable to bad-faith actors who intend to harm trans people by convincing the public that their existence is a miserable and unfortunate one through such "objective" evidence.

To inoculate ourselves against this type of error, it is crucial to recognize that no such "objective" standpoint exists. Everything is ideology, we are all children stumbling in the dark, and no one can see the elephant. Moreover, there is no simple way to see the elephant in full, no shortcut to the right answer. The only shortcuts that exist are shortcuts to finding out that we currently have the wrong answer. If we're not careful, numbers can deceive us into thinking that they are a fundamentally different type of information than our experiences, but they are no more "objective" than experiences. They, like we, are contextual, relational, and limited.

The more I have recognized this phenomenon happening, the harder it is to not see so much of our online communication taking this form. People come across a statement which is like the tip of an iceberg, implying an underlying worldview substantially different from theirs. They then attempt to change the original speaker's (and onlookers') minds by using debate-club style reasoning and evidence that is far too limited to construct a truthful-feeling picture of reality, instead of using evidence and argument only as a gateway into a potentially different reality -- One that could only be understood and epistemically validated by interacting with humans who have experienced it in rich, subconscious detail.

I think this style of discourse is unproductive because it is, fundamentally, epistemically dishonest. As a quick example, you (and I) probably don't believe in the dangerous effects of climate change because we have scrupulously studied all available evidence and "come to our own conclusions". I think it's pretty revealing that most who claim to have gone through such a process are usually sitting on the other side of the debate. Most reasonable, rational people believe in the conclusions of the scientific community through some combination of complex social, philosophical, and institutional trust that has been shaped by an innumerable number of our small experiences in the world and our careful thoughts about it. If we're using a poor facsimile of our true beliefs and their underlying justifications to attempt to change someone's mind, it's not very surprising that we aren't usually successful.

Of course, it's not possible to give someone a course in the philosophy of science in 240 characters, even if you multiply that 240 by "Thread (1/X)". To actually be able to have a productive exchange with someone that results in changed views on either side, both parties must be working from reasonably similar worldviews and yes, sets of experiences (this line of thinking has influenced my new take that some people shouldn't talk to each other). True large-scale persuasion only takes place over mediums that have a lot of room for exposition, with parties that have patience for the process and, usually, already have a strong pre-existing relationship with each other.

One area that I'm fearful we will run (are running) headfirst into the trap of over-indexing on rigid analytical processing is our increased reliance on algorithmic decision-making and artificial intelligence. Much has been written about this, but I'd like to humbly submit my perspective here as an "interested amateur" in the topic. To attempt a super high-level summary of the problem...

Humanity is becoming better at building complicated tools that are (in some ways) extremely effective at making decisions for us in complex, dynamic environments

However, at the same time, we are losing ground in understanding how these tools make the decisions they make, almost in proportion to how quickly they are being developed and used

Many smart people who work on these tools are concerned about the possibility that humanity will soon create an intelligence whose abilities exceed our own, but whose goals are different from ours, and what that might mean for the future of humanity and the universe

I should start by saying I am rather skeptical of the last view, but I agree that something dangerous is going on here. The reasons why I am skeptical are related to the danger, so I will start by doing my best to explain them.

In short, I think that humanity is far further from creating true intelligence / consciousness than a majority of folks who work in "AI" believe. This is partly because I think we are somewhat misguided in our understanding of what intelligence is. If I had to give what I think is a pretty widely accepted definition as we understand it today, I would say intelligence is collectively understood to be the ability to acquire and implement knowledge and skills to accomplish goals across a wide variety of contexts. In a lot of ways, the large language models (think ChatGPT) taking the tech world by storm today fit this description. I think they fit it less than a lot of their proponents do, but it would be hard to argue that it's a complete miss.

However, I believe this definition is missing a crucial component. We don't have very useful language to describe this component right now (if we did, it would probably be because we had a better collective understanding of intelligence), but the best way I can put it is that true intelligence/consciousness also includes the ability to abstract, rearrange, and synthesize structures of knowledge to fit and interact together in arbitrarily new and higher-level contexts.

I'm going to try and use a few other approaches to point to what I'm trying to say here, since my language is limited, but I want to give you as clear of a feeling about what I am trying to say as possible. One way to phrase it is that true intelligence is what allows us to use our previous knowledge about the world to come to a better and better understanding of the elephants we encounter, without ever having encountered an elephant or even the language to describe one. What I believe the current "artificial intelligence" tools are becoming more powerful at is identifying elephants that are already known and fully understood (or producing copies of them) -- Even doing so from incomplete evidence. It's impressive and useful, but I really do think it is a fundamentally different thing.

Another way to put it would be that intelligence is precisely the difference between a specific and predefined strategy of using data to make deckbuilding decisions in Magic: the Gathering, and the intuitive feedback-loop-driven process of top players I described in the first part of this piece. The strategy, or algorithm, can be quite powerful, quite complex, quite nuanced and quite effective. It may contain its own "feedback loops" that allow it to change strategies as it "learns" more about the environment. It may be applicable to other games. It may even achieve better results than the best players (a feat that I'm nonetheless proud has not been accomplished so far in Magic like it has been in chess)). But I believe there is a meaningful and functional difference between what such an algorithm is doing and what the best Magic playing humans are doing when they use data and their experiences to improve their play. I can't prove it, but that's what my intuition says at least.

As we develop more powerful, more complex, and more opaque tools to make decisions for us in areas of reality that contain depth and nuance (read: almost all of them), we are at significant risk of mistaking the first type of thing for the second. ChatGPT is a great example of a recent tool that I think has largely tricked otherwise smart people into believing it is intelligent, even though it is actually only doing a particular (very difficult, complex, and nuanced) thing. It doesn't help that many intelligent and well-meaning people are having reactions like this:

https://twitter.com/michalkosinski/status/1636683810631974912?s=20

Why is all of this so worrying? As I tried to argue with the previous topic, when we confuse numbers and logic, or in this case, sufficiently powerful/complex computation as a substitute for intuition and human intelligence, we are going to end up causing a lot of harm. Is it worth worrying about building a highly capable intelligence that is good at finding truth in the ways we are, or even in more powerful ways? In theory, certainly. A lot of things are worth worrying about. But I also think it is worth worrying about believing we have built such a thing, even when we haven't. It's arguably more dangerous, and in my view, much more probable.

Thinking about this whole topic has been helpful for me not only on the level of understanding some complicated social issues, but also on the level of understanding myself and my own human experience. I have always had a powerful desire for understanding the world correctly, and that desire often feels infinite and unsatisfiable. When I discovered mathematical proofs in school, in many ways they felt like home. They are beautiful because they are like crystals of truth: once you understand one, it is completely solid, immovable... and also completely trivial. There is a common tendency among Math professors to use a theorem while they are lecturing, and claim when asked why the theorem is true that the proof is obvious (common code for this: "left as an exercise to the reader"). But obviousness is a completely subjective feeling -- To a sufficiently computationally powerful being, all proofs are trivial and obvious*, and to a sufficiently simple one, none are.

At least once the proof is discovered, which I also think hints at the type of intelligence required to *understand a proof vs. the type of intelligence required to produce** one from scratch*.

Despite this feeling of obvious truth being subjective and fuzzy, my brain craves it. If I have learned anything about myself in 26 years on Earth, it is this. The more of this feeling I obtain, the more I feel like I understand the Big Elephant. If I am able to poke any kinds of holes in such a feeling, it's very hard to feel satisfied enough to stop myself from pursuing ways to plug them up. The trick, and my salvation, was where this craving led me.

In the early 20th century, the field of Mathematics was zeroing in on a cornerstone result. In some sense, the metaphor I'm using here is a literal one. David Hilbert had proposed a program to find a "grounding" for all of mathematics -- Specifically, a finite set of transparently true assumptions or "axioms" that would serve as a logical foundation to reconstruct the entirety of math from the ground up (a "cornerstone" from which to build the logical structure of math). By doing this, something truly incredible and yet also... seemingly... obvious would be achieved -- A formalization of Math into a logically consistent system whose scaffolding is built on a simple and obviously true set of beliefs (think "2+2 = 4", only much simpler).

Most mathematicians' brains work similarly to mine. We crave logical consistency, and this is what draws us to Math. I think every human can access this feeling in some way -- Math just Feels True in some inexplicable way that causes people to trust it (this is part of humans' struggle with understanding the limitations of information conveyed in the form of numbers). So it came as quite an existential shock to the mathematical community (and my college self) when the quest to establish Math's objective truth and consistency ran us straight into the Looney Tunes tunnel-painting of incompleteness.

Pictured: a mathematician attempting to prove that math is objectively correct

It's quite difficult to explain why Kurt Gödel's incompleteness theorems are so powerful and convincing. I personally would not claim that I fully understand them, but the first ~300 pages of Gödel, Escher, Bach, whose "strange loops" I referenced before, do the best job I imagine is possible guiding someone with little prior math training through it in a convincing way, and I highly recommend them for anyone who is interested and has a lot of time. However, in a way it's quite simple to state what they say. They say that what Hilbert desired is simply, obviously, impossible. Any logical system that has the ability to say meaningful things about The Elephant is absolutely, immutably guaranteed to be unable to describe the whole thing.

What do you do when your desire for logical certainty, possibly the deepest and most fundamental emotion you have ever felt, leads you to an obvious and logical conclusion that this desire will never, can never be enough to understand the full truth of the world?

This was not a rhetorical question for me. It was how I truly felt when I first learned about, and really digested, incompleteness. This may seem like a childish, naive, and frankly, arrogant reaction to have. The only way it makes sense to be deeply shaken by the revelation that omniscience is impossible is to have truly believed you were, at least theoretically, capable of achieving it.

The view I eventually came to, and am still coming to, is that the only way to remain honest to myself and what feels like my fundamental nature was to gracefully and honorably accept defeat. I had to accept that my craving for objectivity and logical structure was merely a feeling -- a feeling that still remains very dear to me, but one that in no way makes me more able to access Truth than any other human or being. The knowledge that we are not capable of understanding everything is something that I think seems clear to almost everyone from an early age; my only excuse for not knowing it sooner is that my threshold for knowing something is that it must feel obvious to me, and this truth hadn't before.

This is a lesson I keep learning over and over in different contexts. One of those contexts is in my romantic life.

"If we want the rewards of being loved we have to submit to the mortifying ordeal of being known."

Tim Kreider

I've always had a lot of social anxiety, and nowhere was this more true than in the context of dating. Of course, some of this was just the extremely normal insecurities that many teenagers have, especially nerdy ones. But I also think a lot of my hesitation in "putting myself out there" was not coming from a place of insecurity, but instead one of perfectionism.

Of course, these two things are related. Being a teenager is HARD, in large part I think because you are simultaneously becoming aware of your values and the qualities you aspire to have as an adult, and painfully being confronted with all the ways in which you are not quite yet ready to live up to them. In my romantic life, this discrepancy was quite vivid: I wanted to be the type of person who made the people I was interested in (largely women) feel safe and comfortable even while I made it clear that I had romantic feelings for them. I also wanted to be a socially conscious person who participated in the struggle to overcome patriarchal norms and power structures that lead to violence/harm against women and gender minorities (or at the very least, not help perpetuate them). As a male-presenting person interested in dating women, it frankly seemed like taking any kind of step towards accomplishing this goal was too risky. Being confident and aggressive risked making someone feel unsafe or uncomfortable, beyond not really feeling true to myself. But being subtle and careful was also bad - Not only is it not how most people actually want someone to "pursue" them, it's fundamentally poor communication and, when done really poorly, straight-up dishonest and manipulative. Is it possible to carefully thread the line and approach someone directly, yet respectfully? Of course it is! Did I trust myself to do that effectively in high school (as well as handle whatever the result might be with grace, emotional intelligence, and tact)? FUCK no.

It was incredibly difficult for me to reconcile my feelings about this with the fact that all around me, my peers were seemingly having no trouble getting into relationships. Did they just not care about doing things the right way? Were they just better at it than I was?

During college, I was able to get over some of my social anxiety. Having a strong network of friends to talk to about dating and tag along with to social events made it easier to meet people organically, and also gave me some of the confidence I needed to initiate romantic connections. Still, I was pretty guarded in my approach. It was hard for me to approach a new relationship with any level of nuance -- Either this person was romantically interested in me, or they weren't. If they were clearly drawn to me as a person, but slow to confirm whether their feelings were platonic or romantic, I saw that as either poor communication or disinterest. I knew I needed practice in this kind of thing to really become better able to navigate it with tact and care, but it was hard for me to dive headfirst the grey areas when the whole point was that I was bad at them.

Slowly, I've begun to realize what I'd been missing the whole time. Believe it or not, incompleteness helped. Fumbling my way into more romantic experiences helped. Growing older and becoming more sure of myself definitely helped. However, I think the most "flashbulb" moment I ever had about this was towards the end of college. I was in a training session for a student group I was a part of, named somewhat cringily, but somewhat admirably (and now I'm realizing, somewhat ironically in relation to this piece), Teaching Racial Understanding Through Honesty, aka, T.R.U.T.H. As is usual for a college student group, our goals were lofty but somewhat unclear. As I remember them, they involved running trainings for other student groups about issues of race, implicit bias and equity, but also generally bringing these issues to the forefront of campus dialogue, whatever that means.

In order to be able to do this work effectively, we all had to go through extensive training & education about the history of racial justice in America, as well as generally learning the skills of how to facilitate large group discussions about difficult and emotional topics. You can probably imagine why joining such a group appealed to me -- less doing, more learning. However, a large part of this training involved learning how to become more comfortable making mistakes in front of other people. If you're a discussion leader, it's important to be confident speaking in front of a group of strangers. You should know how to get your message across, but it's also important to realize that doing so is difficult, and will never be done perfectly. A misspoken phrase, an accidental grammatical error, or even an unintentionally offensive remark can't affect you so much that it prevents you from continuing the discussion (while of course, acknowledging and addressing any harm you've caused when and if appropriate).

This lesson was especially important for us white folks in the group to hear. As well-meaning, liberal/progressive college students in the mid-2010s, we were hyper-aware of how deep and insidious the racism in our country is. We knew that we still held implicit biases, despite our best intentions, and would inevitably fumble our way into hurting people even as we were striving to become better. But as our mentors in this group continually emphasized, we were wanted in the fight anyways*.* Flawed, messy, clumsy action was better than inaction -- as long as we were able to learn and grow from our mistakes.

It's important to give some caveats to this. Intentions aren't the only thing that matters, but they do matter. Being able to gracefully realize when you've unintentionally hurt someone and knowing how to repair that harm (or at least, knowing when you're not capable of repairing it and removing yourself from the situation) is an important skill, and it's one that only works if your intentions are genuinely centered around wanting to treat other people well and thoughtfully. If you're in it because you want to feel like a good person, or get treated like a good person, without being able to constantly recenter on whether you're actually acting like a good person, you're going to be susceptible to some pretty thorny pitfalls. But the ultimate point remains -- It is better to try, and be seen trying, than to sit on the sidelines and plan a perfect strategy.

None of us are the coach, the announcer, or the journalist writing about the outcome of the game. We're playing, and we're needed in the game. You've got the ball and your team is trying to score! Will you run with it, pass it left, pass it right, or shoot? (Maybe you should call a timeout, but that's simply another option on the list). There simply isn't enough information nor time to logically and consciously process your way to the right answer. Someone shouting "Players who shoot in this situation score 45% of the time!" at you isn't gonna help. You've just gotta weigh it (quickly!), live with, and learn from the consequences.

I hope it's quite obvious what the connection here is, and to be honest I'm amazed it took me so long to make it. But at some point towards the end of college, I realized that I was essentially being the romantic equivalent of a white person who had zero Black friends because they were worried they might say or do something offensive. Admirable intentions, terrible impact. Much like this hypothetical (and yet I'm sure very real) white person was ending up treating Black folks differently (and worse) by not allowing them to experience their honest, true, and yes, messy and flawed self, I was not treating women like full people who were completely capable of dealing with, or even being hurt by, my true, awkward, well-intentioned but sometimes shitty self. I think some part of what I truly feared, and still fear, about romantic love is that the idea is to be loved not because of, not despite, but with your flaws -- And the first step of pursuing that is living such that other people will be able to encounter them.

Part of why I'm writing this is for myself to have something to read in the future when I forget this lesson. As we've seen, I'm particularly prone to missing the point in the direction of literal and narrow thinking, so a lot of this is written as a defense of intuition, but I think an equally important piece could be written about the exact same elephant from the opposite side. I just don't think I'm the person to write it.

When I think I could have most used this lesson is when I was first thrust into the crucible of chronic pain. My friends and family will be familiar with the health struggles I've had over the past few years, but the short version is that back in September 2020, I started feeling sharp pain in my hands and wrists, which has since spread to my arms, neck, jaw, shoulders and back. When the pain first started, I remember clinging onto the idea that it would, had to, go away at some point. I would eventually figure out What was Wrong and How to Fix It, either through the help of a doctor or by my own experimentation. For years, I remember being frustrated whenever I had a bad day of pain (which were more common than the good ones). I was constantly analyzing my behavior and hopelessly grasping for something I could do or figure out that would make it better. Unsurprisingly, this cycle of picking apart my own behavior led to a lot of self-blame and stress, which I'm sure didn't help me feel any better (not that it was really possible to avoid or that not doing it would have fundamentally made a difference -- See, there I am doing it again).

I expect that a lot of folks with chronic illnesses and disabilities that happen to us later in life go through the same type of arc. It takes us a while to realize the lesson that those who have been disabled their whole life know as a basic fact: We are in many ways powerless to control our circumstances through reason, observation and effort alone. You can't think, analyze or even act your way into being able-bodied, but overthinking your disability will definitely make it shittier to experience. Moreover, we are (all) fundamentally dependent on others. Yes, even those of us who have not yet had to learn this fact through brutal, tangible, and loving experience.

Before wrapping up, I do want to take a moment and try to better outline what I'm trying to say and what I'm not trying to say. I don't think this lesson, this elephant, can or should be boiled down to "trust your intuition" and "allow yourself to be human and make mistakes". Both are certainly good advice in many contexts, but there is a reason why I started with the usefulness ****of 17Lands data in helping players make better decisions, counter to the intuitions of many top players. If we revisit the strategy employed by the most successful players, the true lesson here is something more along the lines of "trust, act, but then verify". "Trust, act, and then try to falsify" might be more accurate, but it doesn't quite roll off the tongue.

Analytical reasoning is crucially important because it is a really powerful tool for surfacing when and how our fuzzy maps of reality are incomplete. If you find someone claiming to have a perfectly logical map free from guesswork and hypothesis that works for any journey, you should be worried - But if you find someone whose map hasn't at least been tested and adjusted based on its clear internal conflicts or conflicts with the real world, you should be worried too. The search for truth (and, less importantly to me, the living of a human life) is like any good mutualistic relationship. All partners (in this case, conscious logical processing and subconscious intuition) are full individuals with different capabilities. The goal isn't to divvy up the work 50/50 or even to find the right "division" -- It's to figure out how to turn the whole thing into more than the sum of its parts, with each party bringing what they have, getting what they need, and the output being far greater than what any of them could have accomplished on their own.

I really have come to believe this applies across longer timescales and in contexts where we have far more time to think and process and encounter different parts of the elephant. Magic, as 17lands demonstrates, is one of them. So is political discussion, so is computer science, so is math, and yes, so is love. Before wrapping this up, I want to visit one more area where this lesson has come to deepen my understanding of the world, but more importantly, has deepened my experience of it.

When I was younger, I hated visiting art museums with my parents. Part of it was certainly that my smaller legs had a hard time standing for hours on end, but part of it was also just "not getting it". Everyone has this experience with some art, I had it with most of it. My parents told me that I would enjoy art more when I was older, and I thought this meant that I would become informed enough about the reasons behind its value that I would be able to understand and reason why it was "good".

Of course, now that I am older, I realize this is not actually why people enjoy art (at least, when we are being honest and unpretentious). We enjoy art because we see ourselves in it - a messy child who is ignorant of the world but nonetheless exists in it and is trying to figure it out. We invent new languages to try and share our view of the elephant in ways that words cannot. Some are popular, some are niche. Some of us develop into highly skilled inventors of these languages, and it's no wonder we admire those people. But in many ways, the fundamental process is the same no matter if you're a total amateur trying clumsily to paint a landscape by watching a Bob Ross video or an established professional artist scamming a museum out of $84,000. We're all just trying to say something to each other. We know that the elephant is vexing, indiscernible, sometimes cruel and often maddening, but at the end of the day it is beautiful. We want to share it.

I'd like to share with you some art that I enjoy. The first is a video game called Outer Wilds. It's exceptionally hard to say anything about this game without diminishing the experience for someone who is playing it for the first time, because the challenge of the game is entirely knowledge and exploration-based. In fact, a new player who simply happened to go to the right places and do the right sequence of things on their first try would be able to beat the game in about 20 minutes. Here's what I can tell you:

Outer Wilds is a game about a small group of aliens who live in a tiny solar system near the edge of the universe

The game controls are simple but difficult to master - This is because the mechanics of the game are in many ways an honest and realistic portrayal of the physics of the real universe.

You will die a lot. This is an important part of the game's experiential design, which is extremely good at giving you a tangible feeling of the universe's coldness and impartiality. Sometimes things happen that are completely beyond our control or understanding, things that affect us in powerful ways, and there is nothing we can do but experience them.

Still, the game is incredibly hopeful and heartwarming. It makes the case that trying to understand and experience the world is in and of itself a noble goal -- Even if the process is doomed to be short and incomplete.

If you're interested in playing, here is a video I would recommend watching first that might deepen your experience of the game.

Another thing I want to share that my enjoyment of has only deepened over time is what might lovingly be referred to as "sad teenage girl music". One of my favorite dopamine-regulation techniques (ways to procrastinate) is scrolling through the Chinese Communist Party spying app commonly known as Tik Tok. Quite a lot of my feed has become artists sharing clips of songs they are working on. Here are some of my favorite recent examples. Plus a few more. (note: Tiktok seems to have changed their linking system and these may no longer work). Yeah, the algorithm has learned to give me a lot of these. Yeah, a lot of them are probably just well-designed marketing. Yeah, I will totally admit that a lot of them basically sound the same. Music theory is a complicated subject I know next to nothing about, but I do know many of the most popular songs follow a well-trodden formula. So why aren't we bored of them? It's like I said before - Every new song is trying to say something that's been said a million times before, using techniques that have been repeated since the beginning of human civilization. But every single one is written and performed by a person who has never existed until now, a person whose interactions with The Elephant are so similar to and yet so different from our own. It's something each of us enjoys accessing in different ways, but the experience of receiving someone's attempt to convey what logic and language cannot is a deeply meaningful one, and it always will be, even if we keep using the same 4 chords to do it.

I'm going to end this with one last song. Hopefully this overlong dissection of why 17lands.com data has revolutionized the Magic the Gathering limited community will help you understand why I like it so much.

https://www.youtube.com/watch?v=gn-wGaM7x5c