(Originally posted on my Substack on 12/18/25.)
As a techy nerd who got into cube design during its infancy, and whose writing helped to pave the ground to define modern cube design theory, I wanted to discuss this topic for my 60th article about cube design as I felt a strong connection when I saw the following classic User Interface (UX) meme. It captured what I’ve observed in cube design – the gap between a designer’s intentions and what actually happens in gameplay – from both the cube designer’s and the drafter’s perspectives:
You’ve probably seen them in parks, on college campuses, or around office complexes: user‑created routes that show how people and animals actually move from point A to point B. These unplanned tracks are called desire paths, and similar patterns often appear in software applications*.
One of the most famous examples of desire paths is at The Oval pathways at The Ohio State University’s Campus, pictured below, with text added by Zolthux.
The campus was originally laid out with paved paths planned by the designers, but over time students forged their own, more direct routes. These informal tracks, desire paths, are often cited in UX discussions under the theme of user experience vs designer intent: designers plan one way, and users will take whatever route works for them.
Like many memes, this one makes its point pithily but lacks nuance. Cube designers frequently imagine how cards will play or how a metagame will develop, yet actual gameplay and drafting often unfold differently.
As cube designers, we tend to have lofty aspirations for our cubes and how we think they’ll play out.
Most early cube content, when it wasn’t trying to convince readers that cube was worth exploring in a world without an MTGO cube available online, came from an aspirational angle: “look at the weird and awesome things you can do in cube,” paired with broad, designer‑focused views of what a cube should be.
Speaking for myself, my early writing took that same perspective out of necessity. In 2009 few people outside the hobby knew what cube was, I didn’t have a steady local group, and I drove to Grand Prix and SCG events just to get a draft going because local drafts rarely fired.
Thanks to the Magic Online cube, cube gradually moved out of its niche and began attracting broader attention. I appeared on several Limited Resources episodes when cube was still fairly underground, and I was proud to host Marshall Sutcliffe on a Third Power episode in 2015, which has been linked here. At the time, player‑focused discussion of cube was non-existent, and I wanted to show how to approach cube from that perspective.
One important podcast moment that stood out was Marshall describing Blue‑Black Tempo as an archetype. He explained his drafting approach: prioritize cheap, easy‑to‑cast blue countermagic because other drafters won’t value the black aggro creatures like Bloodsoaked Champion highly, since he could usually pick those up on the wheel.
I don’t recall an official document on the mothership that cataloged the cube’s archetypes, but if there had been a CubeCobra‑style overview, it likely would have labeled UB as a classic control color pair given how the colors have historically played.
Marshall’s take on UB was an eye‑opening example of a drafting desire path: players discovering and exploiting an archetype in practice even when the cube designer(s) hadn’t explicitly defined it that way:
Innistrad is a clear case of drafters carving out desire paths. Players built the beloved “Spider Spawning” deck, using cards such as Spider Spawning, Gnaw to the Bone, Runic Repetition, and Memory’s Journey to create an infinite loop that either won via decking an opponent or grinding them down with an army of spiders. That deck wasn’t part of the design team’s explicit plan, as it didn’t appear in the design handoff, nor were other emergent decks like the Clear the Mind Ravnica Allegiance deck (pioneered by Ryan Saxe and Deathsie) or the five‑color Khans of Tarkir morph deck pioneered by Ari Lax’s Pro Tour win.
None of these were listed as draft archetypes in the mothership archetype writeups; clever drafters simply found their own desire paths and made those archetypes real.
The same phenomenon shows up in Software QA: teams write test cases for expected behavior, only to have something unexpected throw a wrench in the works or produce an entirely different outcome.
This isn’t just about collecting feedback as a cube designer, but more about critically assessing your environment, which often is helped via outside perspective(s).
Conversations about “where the rubber meets the road” keep you grounded and force you to consider other viewpoints. I’ve heard designers say things like “in my cube almost any deck will happily play <insert card here>” but a drafter trying to maximize win rate often won’t agree.
Sometimes drafters discover the best play is to ignore designer assumptions and the intended rules of engagement. That’s why I treat CubeCobra overviews—especially long ones—with healthy skepticism: they usually reflect what the designer thinks the metagame is, not how it actually plays out. A somewhat recent episode of the Lucky Paper Radio podcast where Andy talks about how drafters discovered that a very good strategy is to simply play 0 artifacts and win via decking, and then making adjustments to the metagame from that discovery.
If you read my Micro and Macro POVs article, it’s the same idea: viewing the cube through a different lens. Think of “big picture” limited content like Limited Level‑Ups and Limited Resources state‑of‑the‑format pieces – they’re forged in the trenches and focuses on how the metagame actually behaves (which cards and archetypes work or don’t) and how that changes as the environment gets more reps.
Of course, anyone who’s run a cube for a while knows no one’s evaluations are perfect, including drafters’. Still, their perspectives matter and should be taken into account.
Something else that stuck out with me from Ari Lax’s article about the 5-color morph deck in Khans of Tarkir draft was another great example of how set designers can have one perspective, but drafters can find something else:
“The fixing in Khans of Tarkir is actually really poorly positioned for a three color deck. Rather than Shards of Alara, where the fixing was locked into your three colors, the fixing is very modular. It also is randomly distributed.
This means that you can’t rely on seeing your specific on color fixing. Need a green land in Temur but get passed two Swiftwater Falls? Guess what, you now have to play six Forests. You know what that means? You will draw three land hands that can’t cast your spells, be forced to keep them, and die. No one wants that.“
“(Creature sizing) is one of the most important principles of Five Color decks in Khans of Tarkir.
There is a lot of talk about Khans being defined by the three (cast) and five (flip) mana slots and the tempo around them.
That’s true, but that only holds for a small time frame each game. For the rest of the game it’s actually defined by sizing and blocking.
Just like Theros, X/5s rule the world. This is less because of removal lining up (see how Lash of the Whip and Rage of Purphoros worked against Nessian Asp) and more because there just aren’t a lot of 5 power things to break the various X/5 walls.
This X/5 sizing makes a lot of the four-drops really bad. ”
(emphasis mine)
You’ll see these insights from grinders across formats—how formats develop and how they adapt over time—and the same dynamics happen in cube as drafters shape the metagame. Yet cube design rarely emphasizes iterative adjustment. When adjustments do occur, they usually focus on individual card evaluations, which are often treated as a “set it and forget it” exercise. Those assumptions can be wrong. For example, before testing it out during Conflux previews, I assumed Noble Hierarch was primarily a Bant card, but the first person in testing played it well in a mono‑green shell. As silly as it may sound in retrospect, that experience changed my view on card evaluation, since Hierarch isn’t a “Bant card,” but a green card that becomes even better in decks that can also use white and/or blue mana.
One of the joys of designing a cube is seeing drafters play with cards in unexpected ways, which can change how you see things in your cube environment.
While the MTGO Vintage Cube isn’t my cube (I’m just one of the many people who help to maintain it) – it was nice hearing Joe Anderson say he found some unexpected lines and interactions like this:
“Also including enough fixing/fewer super narrow cards, so you get funky color combinations leading to cool emergent things – in the MTGO cube yesterday I used Arwen, Mortal Queen to give Broadside Bombardiers lifelink and then flip my Sorin, of House Markov to finish them.”
While generally, I’ve seen this happen more on a macro than a micro scale, there are times when it happens on an individual card level, like using Sorin of House Markov as lateral storm wincon and card in black midrange decks, using Fragmentize to blow up a Song of Creation in a 100 Ornithopters cube or using rituals like Seething Song to power out 5-mana dragons.
As a cube designer and drafter, you don’t necessarily have to act on those discoveries, but it’s useful to take note of them when they do to see if it’s something that can iterate on in your cube’s design or – as a player – winrate.
Going back to the original desire paths picture, the color picture is what the university’s campus layout looked like years later, as the campus designers created new paths based on the design paths created by the students. The designers used experience in their design taking into account how players play, resulting in iterative design.
“The story goes that, back in 1914, University Architect Joseph N. Bradford waited patiently for winter before using a hot air balloon to get a bird’s eye view of the intricate web of criss-crossing trails that students had carved in the snow. Their findings were then used in the configuration of the more formal network of paved walkways.” – Desire paths: How UI designers can learn from the ways we walk around | by Jeremy Brown | Medium
Don’t be afraid to forge new paths if you see draft experience not reflect how you’ve imagined things in your cube meta. As a cube designer, one of the most integral skills is using what you see in experience and applying it to your format.
Thanks for reading. You can find my socials and my cube lists on my Linktree, as well as other cube set review articles and design articles that I’ve written over the last 15 years.
*For example,X (Twitter) “paved” a number of desire paths by integrating them into the service, including @ replies, hashtags, and group discussions.
Leave a Reply