Is yes or no tarot accurate?

As a forecast, no: one card read against a yes or no chart lands close to a coin flip. It can still show you something true, which is your own reaction to the answer in the second you see it, before you start arguing with it.

A yes or no reading is a single card drawn against a chart that sorts the deck into yes, no and sometimes maybe. The most common chart is also the simplest: upright means yes, reversed means no. Run the math on that one. A deck holds 78 cards, every card can land either way up, and so 78 of the 156 possible faces say yes. That is 50 percent before you have asked anything. Charts that sort by card meaning do not rescue the method, because they disagree with each other. Compare any two published charts and you find cards filed as yes on one and maybe on the other.

We know of no controlled test that shows a deck forecasting events better than chance. We make tarot readings, so saying this costs us something, and it is still the accurate thing to say.

Three things make a yes or no draw feel accurate even at coin-flip odds. One, selective recall: the draw that came true stays with you for years, and the 6 that missed are gone by Friday. Two, the Barnum effect: a statement loose enough to fit nearly anyone feels personal, and an answer like "no, not yet, not this way" fits nearly anything. Three, flexible scoring: when the answer was yes and the thing fell through, people decide it meant yes, but later. A forecast you can rescue after the fact cannot be wrong, and a forecast that cannot be wrong tells you nothing.

Re-drawing flatters the score further. If the first card says no and you keep drawing until one says yes, you have flipped a coin until it agreed with you. Asking the same thing a second time is a different act when the second question changes something, and a re-draw with identical words changes nothing.

You can test all of this at home without taking our word for it. Write 20 yes or no questions with answers you can check inside 7 days: whether the parcel arrives by Thursday, whether a colleague replies before noon. Draw one card for each, write the answer beside it, and score the sheet on day 7 with no reinterpretation. Chance gives you about 10 out of 20. A score of 12 or 13 sits well inside ordinary luck. Keep the sheet, since paper is harder to argue with than memory.

So what is a single card good for? Your reaction. Before the card turns you do not always know which answer you want, and the instant you see a no, something moves: relief, or a drop in the stomach. That response is information about you, and it exists whatever the card happens to be. A coin produces the same flinch, and a coin costs nothing, which is worth remembering.

Picture Marta, a nurse with a new contract on her kitchen table and a Friday deadline. Her first question is "Should I sign this contract by Friday?" and the chart says yes. Her stomach drops, which tells her more than the card did. She rewrites it as "What am I protecting by signing?" and draws three cards, two of which carry the read. The Four of Pentacles lands in the first seat, the one that holds what is moving now, and she sees her grip on a safe income. The Eight of Swords lands in the second seat, the one that holds what she is not looking at: a limit she has never tested. Friday stays her call.

That rewrite is the whole move, and it explains why open questions hold up better than closed ones. A closed question has 2 possible answers, and a card can only land on one of them. An open question starts with what or how, so there is no score to keep, only something to notice and check against your week. The questions that go wrong before any card is turned tend to share this fault. If a yes or no question is on your mind right now, turn it into one sentence about what is moving inside it, then pull three cards against that sentence and treat them as prompts, with the decision staying yours.

Some questions need a different tool altogether. The cards do not know whether a lump is serious, whether a clause in a lease is enforceable, or whether a pregnancy test would read positive. For those, the right tool is a doctor, a lawyer or the test itself. Use that tool first, and read afterwards if you still want to.

A single card does have one fair use, and that is the small question with nothing riding on it, like which of 2 errands to run first. It makes a decent nudge and a good party game. One rule covers the rest: if you would not let a coin settle this, do not let a single card settle it either.

Take the yes or no question on your mind and write it in one sentence. Imagine the card came up no and write the first word that arrives, then do the same for yes. Now rewrite the question so it begins with what or how, and stop once the sentence names no outcome.