I spent all that time sleeving up the Commander Cube and then I gave up.
Of course, I can still use the Commander Cube to easily build myself new Commander decks. Nothing lost there. But when we actually tried to draft the thing, it was disheartening. The reasons it didn't work are obvious in retrospect.
Oh hindsight, thou art twenty twenty.
Here's what I learned:
Up first, unless your friends and drafting partners are very familiar with Magic, passing foreign and textless cards around the table will ruin your experience.
I'm that guy that can recognize thousands of cards by sight. I like this game. I play it a lot. Not everyone is like this. In fact, most people are not like this. So, while a textless Disenchant looks cool and is an iconic card that "everyone knows," remembering if it is an Instant or Sorcery off the top of your head when it gets passed to you in a draft is not a reasonable expectation for all players.
It may be tough to accept, but some of the people you play with might not have ever cast Disenchant. It last appeared in cardboard in Time Spiral, circa 2006. That's almost 10 years ago.
Oh God, I'm old.
Up next, there are lots of ways to tweak the drafting rules to make them quicker and more friendly. Not everyone wants to dedicate 3+ hours to drafting and playing. We have fun using a "pack war" or mini-master format, where cards are dealt out from the cube and the game starts. More cards are added between rounds.
This makes it easier for newer players to get started since they don't have to make deckbuilding decisions until they get a chance to see at least some of the cards in action in a real game first. It also means that the time you would have spent drafting, you spend playing.
For a more Commander experience, deal out 30 face-down cards to each player. Then, give out 4 of each basic land for a total of 20 lands. That makes a 50 card deck with at least 20 lands. Next, deal out a couple of legendary creatures to each player and let them pick one to act as their Commander. It's a half-size deck, so cut the life total in half, too.
After the first game, deal out another 30 cards to each player and rebuild decks. This time, enforce the color-identity rule. Now, players will know at least some of the cards to expect, but with some surprises from the newly dealt out cards.
Surprise!
Another thing I learned is that is it much easier to evaluate cards for the cube if you pick a strict number of total cards and stick to it. After I chopped out all of the foreign and textless cards, I was able to stick to a 360 card total. The most people I would play this format with is 4, so 360 cards is enough for everyone to get 6 "packs" of 15 cards if we wanted to play a traditional sealed deck format.
If you have a well-tuned Commander deck, don't make cards from that deck part of your cube. There's a good chance that you are running cards that work really well in the context of that particular deck, but that don't play nice with the cube's card pool. Again, this seems obvious in retrospect, but there are cards that we consider "good" in Commander that are only good in combination with one particular card or strategy.
For example, compare Necropotence to Read the Bones. Necropotence is a very powerful Magic card. It ruined an entire tournament cycle for a lot of players because it was so good that you either played it or played against it. But, it requires a heavy commitment to black to even cast, which is not always easy to achieve in a limited environment. Read the Bones doesn't have the raw power level of Necropotence, but it is much easier to cast in combination with colors besides black, draws cards, and even gives you some library manipulation in the process.
Speaking of drawing cards and library manipulation, you want to run a lot of these effects. Drawing cards covers a multitude of sins, and library manipulation has a similar effect of fixing terrible draw sequences that knock players out of a game. Because of the singleton nature of the format Commander is random. That's cool. Not drawing lands is not cool, though. If you can give players more ways to draw cards, loot cards, or rearrange cards, it lessens the randomness.
With that said, take the tutors out of your cube. Tutors require a shuffle, which is time-consuming. They also severely lessen the randomness, to the point that the game becomes about getting to and casting the tutor. Look at it this way, the guy who gets his tutor first gets to find the best card in a limited card pool. Because the card pool is limited, and there is a high degree of randomness to the decks already, one tutored card can entirely dominate the rest of the game.
Finally, find editions of cards where the wording has been corrected. Newer players may not understand that the printing of the card is different from the official Oracle text. This can be confusing and may take time away from actually playing the game while you have to look stuff up. Take a look at the original printing for the Impulse vs. the corrected version of Impulse.
See that last line, "Shuffle your library afterwards."? That's not supposed to be there. This is clear in the reprinted version on the right, but the original printing contains the error.
Another example of this is when the older wording on the card didn't make much sense. They are much better at designing cards now, but we still have access to old cards that do cool things. Try to find a reprint. Take a look at Animate Dead old vs. new.
Oh geez. That card still doesn't make any sense!
Maybe just use Reanimate instead. :)
