I was inspired to make a cube after playing a pretty good conspiracy/chaos multiplayer cube at my friend's house.

I have tried to make a cube twice before. Once as a 180 card cube just using cards I had sitting around, which we played once or twice and it wasn't very fun, without high rarity cards the game usually comes down to a small amount of card advantage or board stalls. Another time I got about a third of the way through designing a nonblue cube with six archetypes, trying to have a mix of aggro, midrange and control. Eventually abandoned it. The hard part was not having a good idea of how to look for cards to add to it.

My new cube is complete and can be found here
viewtopic.php?f=33&t=20665
I decided to design it around 12 linear archetypes. This way it would be easy to choose cards since I just need to choose about 20 per archetype plus some all-around cards but drafts would still be interesting since there are only 8 players, not all decks will be made every time. Plus casual players enjoy playing proactive decks instead of a bunch of defensive cards with 2 win cons like you sometimes find in more generic drafts.

6 archetypes were 3 color and 6 were 2 color so each color would be represented 6 times. However, some 2 color pairs appear once more than the others.

The cube is 100% theorycrafted and took a few days to design. It's probably easier not to be too picky when first designing and then refine after testing, but it's hard for me to avoid overanalysing. Also I'm not sure how many people will want to give me feedback (irl players probably more helpful than online, but online easier to set up games), so I wanted it to be good on the first go.

Here are some things that were a struggle -

- Are there any OP cards? For a little while Blightcaster was in there, which would have been pretty unfair. But having a look on my last pass I noticed that Talrand, Sky Summoner will probably be almost as strong. I'm leaving it in for now.

- What if I gave strong cards to one color but not to others? Looking at the RG ramp archetype, red has some silly stuff like Scrambleverse and Dragon Roost while green gets Vorinclex, Voice of Hunger and Woodfall Primus. Does every color need a Vorinclex-level card?

- I'm aware of some of my biases, so what if I over-compensate for them? I know that I don't like counterspells and had to force myself to put some in, but the final cube ended up with 6.

- What if cards get used in other archetypes, like a non-mill deck taking Brain Freeze and building up a big enough storm count to win anyway?

- Some arbitrary restrictions meant I had fewer cards to choose from. I didn't want any transform cards or hybrid cards. I'm not sure if I had any cards with +1/+1 counters, which I generally tried to avoid because I had -1/-1 counters in there. Some cards had not been printed with updated text and might have been too hard for players to interpret (like Wicked Reward). Some cards would make players suspect others would be in the cube and they weren't, so they had to be left out (like Sphinx's Herald).

- I wanted to include some reactive cards especially since some archetypes were slower than others, but second guessed how much is too much. I found it hard to choose almost any red cards since at first I decided I only wanted a small number of burn spells, but this later got relaxed.

- I started to worry about the p/t of creatures; a lot of utility creatures were higher in toughness than power, so I worried things would stall too much and decided to put in some pump spells. This started to make the cube look less offbeat and more like a generic draft though.

- Related to these last two points is the order than I designed the cube in. If I had designed bant populate first I wouldn't have worried about putting in cards like Labyrinth Guardian and Aven Wind Guide (maybe still avoid the wind guide because I didn't want to confuse boros swarm players into thinking they needed blue), but since this was one of the last to be designed I was being more cautious about defensive creatures.

- Also not being sure what the reactive cards needed to counter. I wanted to make sure that some but not all removal could hit a big Thunder-Thrash Elder, but later I ended up taking Dragon Fodder down to a 1-of which meant the elder wasn't such a big deal anyway.

- Some archetypes had lots of cards on gatherer available to use; it was easy to liberally add red cards to the ramp and swarm archetypes, whereas there are very few cards with populate. So I'm pretty sure some archetypes got more useful cards than others.


Final thoughts

- Making stuff by yourself is hard. If I had someone else they might balance out my biases.

- Theorycrafting is hard. I'll have no way of knowing if this is cool, fun, or balanced till I actually play it.