I've written about many aspects of Cube design, such as looking at cards and your Cube from a holistic point of view to how Signets and bounce lands aren't unhealthy for Cube. However, I've never really talked about how to get a Cube started. It's a big, imposing process that can be overwhelming, and it's daunting enough to make many just give up. In my next few articles, I'll talk about how to break down that barrier to entry and get a Cube started. This article will focus on some of the first steps, talking about constructing Cube sections and some ratios for sections.
But first, a history lesson.
When Cubes were first created in the early 2000s, their construction was ruled by rigid constraints. Because there weren't many Cubes in the beginning, people held steadfast to the rules created by the originators of the format. The format was called "Cube" because it was a format designed to have the game's best cards with each section having the same number of cards. The rule that was held to was that a Cube was supposed to be 410 cards, each section containing 50 cards (WUBRG, artifacts, lands, multicolor) with multicolor having ten additional cards, one for each tricolor pairing. As each section was of equal size, it emulated a "Cube" (this may not be technically correct, but it's much catchier than asking people if they want to Octahedron Draft).
While this may have been for strict symmetry, it may have also been because many tricolor pairings had their first really playable cards in Apocalypse. The "Ceta" pairing of U/R/G had Guided Passage and, well, Guided Passage (as Intet, Riku, and Maelstrom Wanderer wouldn't be printed for another six years) since Cube designers back then didn't consider putting cards like the volver cycle into their respective triads or cards like Kird Ape into Gruul. However, the inclusion of cards like Guided Passage and Overgrown Estate were likely because of "new card optimism" or a general false level of optimism to convince themselves that those cards were good enough to complete that cycle of triads.
Sections were further defined by rigidity as white, black, and red were supposed to have creatures and noncreatures at the same size (25/25 split), whereas blue was allowed to have more spells and green was allowed to have more creatures. However, this eventually went by the wayside years before creature power creep as white ended up being more creature-centric along with red and black, although not as much as white. Using my Cube as an example, I'm running 37 creatures and 24 spells in white.
Why the history lesson? It's to illustrate how thinking in terms of rigid construction used to be the norm and how that's no longer the case. Cubes come in varying sizes nowadays, and the sections aren't rigidly defined either. As I discussed in my Pauper Cube format overview article, restrictions like that can be useful as a self-disciplining measure to make sure that you aren't overloading on noncreatures, but such measures are no longer necessary. Although creating a Cube from scratch is imposing, we can start Cube construction by looking at how we define it. In other words, from the ground up.
While I've talked about Pauper Cubes and how they differ from regular Cubes (and in a future article, I may talk about Peasant Cubes), in this article I'll assume that you're using rares since most Cubes use rares.
One of the biggest things to consider is the size of your Cube. If you have a consistent playgroup that you're going to Cube with, it helps to define the minimum number of people that will draft it. Since a three-pack Cube set is 45 cards, we can use that as the basis for how big a Cube can be. Many people use 360 as a minimum to allow for eight drafters, and because of this it's the generally accepted minimum for a Cube size, although there are Cubes at sizes like 180 and 270.
When first building a Cube, I generally advise starting on the medium side, between about 500-650 cards. Don't feel that you have to make your Cube size in terms of multiples of 45 cards (360, 405, 450, etc.). For example, I've personally been happy with having 460 cards because it allows me to tinker with multicolor and mana-fixing options in the format.
Although smaller Cubes make it so that changes are easier to see in terms of requiring fewer cards to create a tangible change, which can help in terms of learning how changes effect the format, it gives you more wiggle room for customization and can help to make initial cuts easier when learning the format since making choices of what to cut can be excruciatingly hard at the smaller levels. There's the factor of seeing many of the same cards repeatedly, which can be a plus or minus depending on how you and your drafters feel about seeing the same cards frequently (it can get old for some, but others may enjoy the consistency of the format).
I'm a fan of a medium list as opposed to a large (720+) list for people starting with Cube since those sizes can have their own problems, such as with mana-fixing lands that enter the battlefield untapped, which I will discuss later. That said, there really isn't a "wrong" size for a Cube, either when starting out or once you have learned the format more. Smaller Cubes tend to have more tight interactions and make it easier to support cards that require other cards (big dumb robots for Tinker, equipment for Stoneforge), but finding room for those is more difficult, which is why I shy away from recommending smaller lists for those who are new to the format.
For the individual W/U/B/R/G sections, we are assuming that the color sections themselves are equal.
As I discussed earlier, a lot of the rigid measures helped people with questions like how big each section should be. Like the limited finisher options in Pauper, it was self-disciplining; you can't overload your Cube with multicolor cards if you can only include 60 out of 410 cards, after all. However, in the past few years, we've gotten much better options. Advent of the Wurm has been virtually ignored for Cube because while a 5/5 flash token for 1GGW is quite the deal in terms of raw mana efficiency, there just isn't the room for it.
How much room, therefore, should we devote for each section? When discussing them, although I use a type of guild system in my Cube, I'll do it in terms of traditional Cube construction: W/U/B/R/G, artifacts, multicolor, and lands.
Courtesy of the "Card Comparison" link on MTGS, there's an approximation of types used per section thanks to user eidolon232:
One thing that you'll note about these numbers is that even though the sums don't exactly equal the size of a Cube (there's some wiggle room, likely due to rounding), the monocolored sections are if not the largest sections close to it.
Looking at another site, Cube Tutor, there's an "average" listing for various sizes.
Although the lists show different averages for 360 and 450 cards, they show that most Cubes put an emphasis on monocolor cards and lands. I'll discuss why this is correct.
Many Cube decks are multicolor, and it may seem intuitive to tailor your Cube with this in mind by shifting emphasis towards multicolor cards. The problem with this is that the cards themselves are naturally restrictive in terms of what decks can use them. Since tracking data, I've found that multicolor cards get maindecked about 56% of the time, as opposed to colorless cards being maindecked about 76.8% and the W/U/B/R/G sections having a maindeck percentage of about 65%.
Due to the increased quality of multicolor cards, there's the temptation to on overload multicolor in Cubes—even Pauper Cubes got a lot of their weaker guilds fleshed out in Return to Ravnica block. However, due to the cards being naturally restrictive, having a lot of these cards can result in a lot of late picks and cards that don't get drafted. I've found that in Cubes with a heavy multicolor emphasis, the late multicolor cards end up mainly as freebies for Five-Color Control since no other deck really wants cards like Intet, the Dreamer because it isn't really worth it for decks like Izzet to splash green. Because of this, I recommend aiming for about 10% multicolor cards in your Cube (about 1.5 per pack). This helps reduce the number of cards that can lend themselves to being either unplayed or in sideboards.
On the topic of multicolor cards, there are a few options to consider when thinking of multicolor cards.
One promotes using uneven multicolor cards because some color combinations like Selesnya and Golgari have a very high concentration of solid cards while other color combinations like Dimir don't quite have that same density (though don't get me wrong, its good cards are very good). This usually happens more with larger Cubes as the options get thinner for color combinations like Dimir whereas Selesnya, Golgari, and Rakdos have a lot of powerful options. The argument is that the differences are small enough that it won't be felt on an overall Cube power level, much like how the inclusion of Loam Lion in Worldwake didn't push Selesnya in that format (as red and black were the powerhouses of it).
Personally, I don't ascribe to this model since I want to make sure that each color section is even to give each color combination an equal chance, but those who have used it (such as the aforementioned Mayor of Avabruck) note that there aren't discernible differences between the power of, say, Rakdos over Dimir due to the multicolor cards. Using this method can affect the multicolor section by making it uneven with the other sections, but I find that isn't something to worry about.
Another promotes putting hybrid cards into the color combination where they would most likely fit, such as including Boggart Ram-Gang in red (due to the aggressive nature of the color) and Murderous Redcap in black (due to the sacrifice-heavy nature of the color). The argument is that because these cards have hybrid as a benefit rather than a drawback, they shouldn't be competing against the naturally restrictive cards—a card like Murderous Redcap can be played in many more decks than even something easy to splash like Olivia Voldaren or Falkenrath Aristocrat. This can also affect the size of a multicolor section by allowing for more multicolor cards to be used (Kitchen Finks going into green allows for another Selesnya card to be included), but as long as you remain disciplined by keeping the overall section small, that isn't something to worry about either.
Lands are the glue that hold a Cube together because, as mentioned before, most Cube decks are multicolor and the primary mana fixing for most Cube decks is found through mana-fixing lands. There's a fear that having too many of them makes multicolor-based control decks "too good," but I haven't find this to be the case at all—it's more about the lands themselves than the number of lands.
Justin Parnell discussed the role of the importance of lands that enter the battlefield untapped for aggro decks; since while most decks naturally want to curve out anyway, it's imperative for aggressive decks to do so since their game plan is to get as much use out of their mana during each step of the game. Think of the original Sligh decks and their structured mana curve so that they could make the best out of their mana every turn.
In Cube, the XX and 1YY mana cost of aggressive creatures historically was a problem for aggressive decks, although in recent years the printing of cards like Thalia, Guardian of Thraben; Young Pyromancer; and others with easy-to-cast costs replacing the mana-intensive costs of cards like White Knight / Silver Knight help take some of that burden off. But that being said, being able to get the right mana on curve is crucial for aggressive decks.
Therefore, it's much more about managing the lands themselves and making sure that there are sufficient mana-fixing lands that enter the battlefield untapped for aggressive decks than blaming the number of lands that are there. Mana-fixing lands like pain lands tend to help aggressive decks more, whereas bounce lands and vivid lands tend to help control decks more. As mentioned before, one of the drawbacks of a larger Cube is that the options for lands that enter the battlefield untapped tend to run out quickly; the Magic Online Cube suffered when the Ice Age pain lands and the Shadowmoor filter lands weren't included, as I noted in my initial impression article, because there weren't enough lands that entered the battlefield untapped. In larger Cubes, the options for mana-fixing lands that enter the battlefield untapped can run thin, which is a drawback of larger Cubes.
This can also be seen in non-rare Cubes, as the options for mana fixing that doesn't enter the battlefield tapped are extremely rare; because of this, the context may change. I'd use a bounce land or a Borderpost in a Pauper Cube aggro deck but wouldn't want to be anywhere near one in a regular Cube aggro deck.
Other lands that don't fix for mana like Mutavault / Mishra's Factory, Ancient Tomb, and Maze of Ith generally play similarly to artifacts in that they're not multicolor-tied but can support different archetypes (I wouldn't play Maze of Ith in an aggro deck, nor would I play Thran Dynamo); because of that, it's much harder to overload on those kinds of cards since they don't frequently make dead picks and they're maindecked often. I've found that having at least 10% for lands is a good starting point and a level between 10-13% is a good number to end up with to aptly provide mana fixing, but in formats like Pauper where options are limited, you have to work with what you have.
Lastly, I've found that artifacts can be tricky due to their non-binding nature to colors and color pairs. In formats like Pauper Cube, there aren't many good options available aside from staples like Bonesplitter, Signets, and Borderposts, but like multicolor, Scars of Mirrodin made artifacts in regular Cubes much better. As mentioned before about how non-color aligned lands play very easily across archetypes and because of it tend to see high maindeck play (I tend to put color-aligned lands in their color section, like Shelldock Isle in blue).
This means there isn't a need to try to minimize artifact slots in a Cube because the cards are incredibly flexible, especially the non-archetype specific ones like the Mirrodin/Scars Swords. Even artifacts that aren't universally playable are very useful in design because they can go a long way—a Grafted Wargear can supplement an aggro strategy in any color, whereas Wurmcoil Engine and Myr Battlesphere work in just about any non-aggro archetype.
I've found that an artifact section works well when it is about the same size as the W/U/B/R/G section since it supplements all of the colors and archetypes, but it's hard to include too many artifacts unless you go way too far and typically the dropping in power level can help to prevent that.
The main lesson to learn here is that even though the sections of your Cube may provide the building blocks of the sections, they aren't rigid; if you feel like a section size needs to be changed to meet the needs of your Cube, you can do so. If you feel that you need more lands than the 10-13% figure states after you've become more accustomed to the format, change it and see how you like the change. For example, I recently cut ten multicolor cards to see how their loss impacts the format.
Building a Cube is a wonderful journey of environment crafting. I hope that this article has given you some helpful tools for breaking down the imposing wall that may have stopped you from creating a Cube.
May your opening packs contain Sol Rings!