Testing a deck before you sleeve it

The question worth asking first

Not "is this deck good" but "how often does it do the thing". Pick the card the deck is built around and look at the odds of having drawn it by turn three, turn five, turn seven.

• Choose the deck, then choose a card in it.
• The odds are shown by turn, on the draw and on the play, since being on the play costs you a card and the difference is not small.
• A single copy of a card in a hundred card deck is a much longer wait than people expect. Seeing the number is usually what prompts adding more ways to find it.

Cards that do the same job

Switch to group mode and tick several cards at once. This is the more honest question most of the time, because you rarely need one specific card: you need any of your four ways to remove a creature, or any of your six ramp spells.

• Tick everything that fills the role and the odds are for drawing at least one of them.
• Doing this per role is the fastest way to find the hole in a deck. Removal looks fine, ramp looks fine, and the thing that actually wins shows up on turn nine.

Lands

Group your lands and ask how often you have three by turn three. It is the single most useful number in deck building and almost nobody checks it before playing twenty games and forming an impression.

If the answer is uncomfortable, the fix is usually one or two more lands rather than a rebuild, and the tool will tell you exactly how much that changes things.

Sample hands

Numbers tell you the shape, and hands tell you what it feels like. Draw a few opening hands from the real deck list.

• Seven cards, as you would actually see them.
• Draw a dozen and count how many you would keep. That ratio is more informative than any single number, and a deck where you would mulligan half your hands has a problem no amount of card quality fixes.

Reading the odds properly

• These are exact probabilities for the deck as listed, not a simulation, so the numbers do not wobble between runs.
• They assume you draw normally. Cantrips, tutors and card selection all improve the real answer, so treat the figure as a floor.
• They say nothing about whether drawing the card wins the game. A deck that reliably finds a card that does not do enough is still a deck that loses.

When to use it

Before sleeving a new deck, and after any change to the mana base. It is also the quickest way to settle an argument about whether a deck should run three copies or four: check the turn you care about and see whether the fourth copy actually buys you anything.