How to Shuffle and Cut a Tarot Deck
The mechanics, the common conventions, and why there's no single 'correct' method.
New tarot decks come with noticeably stiffer cardstock than a well-worn deck of playing cards, and that single fact explains most of the disagreement about how to shuffle one. A tight riffle shuffle — the kind that bends and interleaves two halves of the deck — works fine on a broken-in deck but can crease or wear a brand-new deck's edges faster than gentler methods, which is why plenty of experienced readers avoid it entirely for tarot even though it's their default for playing cards.
The overhand shuffle is the most common alternative: hold the deck in one hand and repeatedly pull small packets off the top with the other, dropping them onto a growing pile. It's gentler on the cards and mixes the deck thoroughly with enough repetitions, though it has one quirk worth knowing — an overhand shuffle that never rotates a packet end over end preserves each card's original upright/reversed orientation through the whole process, which means if you care about reading reversed cards, pure overhand shuffling alone can under-produce them.
For decks with delicate or aging cardstock, or simply for a gentler physical ritual, many readers use a 'spread and mix': fan the entire deck face down across a table and swish the cards around by hand for ten to twenty seconds before gathering them back into a stack. This method has the practical side benefit of naturally flipping a genuine, roughly even mix of cards end over end, which is the most reliable way to get authentic reversed-card odds if pure overhand shuffling has been your habit.
Once the deck feels sufficiently mixed — there's no fixed number of shuffles required, just genuine mixing rather than a token gesture — the cut is the next step, and conventions vary here too. The simplest is a single cut with the non-dominant hand, splitting the deck into two piles and restacking them in either order. A more ceremonial variant, borrowed from the Golden Dawn tradition, splits the deck into three piles left to right and restacks them in reverse order (right pile on the bottom, then the middle, then the original left pile on top); it's a nice ritual flourish but functionally no more 'correct' than a single cut.
If you're reading for someone else, it's common practice to let that person shuffle and cut the deck themselves while holding their own question in mind, on the reasoning that a genuinely personal question benefits from personal contact with the deck before the cards are drawn — though plenty of readers shuffle for their querent instead and report no meaningful difference in outcome. Neither approach has any documented advantage over the other; it comes down to what feels right for the reading, not a rule either method is breaking.
What actually matters across every method described here is genuine mixing, not the specific technique used to achieve it. A hurried three-second riffle followed by an unthinking cut mixes a deck far less than a slow, twenty-second overhand shuffle done with real attention — and since the deck's shuffled order is the entire mechanism producing a meaningful draw, rushing this step undercuts everything that follows it more than any 'wrong' shuffling style ever could.
A related but distinct process worth knowing is 'washing' a brand-new deck — spreading every card face down across a large table surface and mixing them thoroughly by hand, sometimes for a full minute or more, before the deck ever gets its first proper shuffle. Many readers do this once, specifically with a new deck, partly to physically soften stiff new cardstock faster than ordinary handling would, and partly as a one-time ritual marking the start of their relationship with that particular deck. It's a one-off practice rather than something repeated before every reading, distinct from the regular spread-and-mix shuffling method described above, which is meant for ongoing use.
In a group setting, where several people are taking turns reading with the same deck, it's worth reshuffling fully between each person's reading rather than simply cutting the deck once and moving on — a full reshuffle resets the deck's order genuinely, where a quick cut alone mostly preserves whatever order the previous shuffle already produced. If time is short, even a brief twenty-second overhand shuffle between readers is enough to meaningfully randomize the deck again, more so than skipping straight to a cut.
The three-pile cut, if you choose to use it, works concretely like this: split the full deck into three roughly even piles moving left to right across the table, then restack them in reverse order — the rightmost pile goes on the bottom, the middle pile stays in the middle, and the original leftmost pile goes on top. It's a small physical ritual borrowed from ceremonial magic traditions rather than a mechanically necessary step, and skipping it in favor of a single simple cut changes nothing about how randomized the deck actually is.
Frequently Asked Questions
Do I need to shuffle a specific number of times?
No — there's no fixed count. Shuffle until the deck genuinely feels mixed rather than counting to an arbitrary number; a slow, attentive shuffle for fifteen to twenty seconds mixes a deck more thoroughly than a fast one repeated many times.
Does it mean something if I drop a card while shuffling?
Some readers treat a dropped card as significant and read it as an extra message; others treat it as a purely physical accident with no bearing on the reading. Both are common, honestly held positions — there's no tradition-wide rule settling which is correct, so pick whichever convention feels right to you and stay consistent with it.
Should the person I'm reading for shuffle the deck themselves?
It's common practice, especially for a personal question, but far from mandatory — many readers shuffle for their querent instead. Neither approach has a documented advantage; it's a matter of preference rather than a rule.