In short: Put a card you know is real from the same era beside it and look at both under a 10x loupe. Magic: The Gathering cards are offset printed, so mid-tone color breaks into small rosettes of dots and the black text is flat black with hard edges. A counterfeit builds that black out of colored dots and the letters go soft at the edge. The light test is faster and much weaker. Failing it proves a fake. Passing it proves almost nothing.
Find a real card from the same era before you test anything
The card on the desk is a Revised dual land bought online. Before any test, dig out a second card: anything from Revised or the sets around it, a bulk Swamp is fine, as long as it is white bordered and you know where it came from. Every check on this page is a comparison. Without the second card you are testing a suspect against your memory of something you last held in 2019, and memory is where counterfeits win.
Match the era, not the value. Magic has been printed at several plants across three decades, and gloss, stock color and ink density all drifted over that time. Hold a 2024 mythic against a 1994 dual land and the two disagree on half a dozen points, none of which mean anything. With the eras matched, the differences that remain are the ones worth reading.
The light test can only rule a card out
Hold the card flat against a phone flashlight and look at the front. A Magic card is two printed paper faces glued either side of an opaque core, and that core is what stops the back being read through the front. A counterfeit run off on plain card stock has no core at all, so the art turns luminous and the printing on the back starts to show through it. The check takes two seconds and risks nothing.
It only works in one direction. A card that fails is fake. A card that passes has proved that its printer bought the right paper and nothing beyond that, because core stock is cheap and the people counterfeiting Revised duals stopped failing this years ago. Treat the lamp as a filter, then get the loupe out.
A 10x loupe settles what the lamp cannot
Genuine Magic cards come off an offset press. Cyan, magenta, yellow and black go down from separate plates, each screened at its own angle, and where the screens overlap the dots gather into small flower-shaped clusters. Those clusters are rosettes. At 10x they show up in any mid-tone area of the artwork, a shaded rock or a patch of sky, repeating evenly across the whole card.
Black text is where a copy gives itself away. On a real card the name, the type line and the rules text come off the black plate alone, so the letters have hard edges and the black is flat and solid. Under the same loupe a fake often builds its black from cyan, magenta and yellow dots, or carries a faint colored halo along each letter edge, or breaks into a square grid instead of rosettes. Start at the edge of a capital letter in the type line.
One row of the table below is dead on this card. Revised is white bordered, so there is no black edge to read, and the art and the type line carry the whole answer. Same light, same magnification, both cards. You need to see it once.
- A 10x jeweler's loupe, or a phone camera in macro mode at full zoom.
- Daylight or a white desk lamp. A warm bulb buries the cyan.
- Both cards out of their sleeves and flat on a pale surface.
- Look in this order: a mid-tone in the art, the type line, then the border on a black-bordered card.
Sources: TCGplayer: how to spot counterfeit Magic: The Gathering cards
| Where to look | Genuine card | Counterfeit usually |
|---|---|---|
| Mid-tone area of the art | Rosettes of cyan, magenta and yellow dots | A square dot grid, or a smooth inkjet spray |
| Edge of a capital in the type line | Hard edge, flat solid black | Colored dots or a cyan halo along the edge |
| A black border, where the set has one | Even black with a fine dot structure | Gray or mottled, visibly made of colored inks |
| Mana symbols in the cost | Clean circle, sharp symbol inside | Soft circle, symbol filled with stray dots |
| The cut edge of the card | Printed face, dark core, printed face | White all the way through, or two layers |
Read the cut edge, then measure the card
Turn the card on its side and put the loupe on the cut edge. A genuine card shows three layers: printed face, core, printed face. The core reads blue-black against the white paper either side of it and runs the full length of the edge without breaking. Single-ply counterfeit stock shows white all the way through. A fake glued from two faces with no core in between shows two layers where there should be three.
Then the measurements. Square the suspect against the known-real card, corner to corner, and sight down the two outer edges: a standard card runs 63 by 88 mm, about 2.5 by 3.5 inches, and anything trimmed or oversized leaves a line of overhang. Counterfeits come out slightly small when they have been trimmed to hide a bad cut, or fractionally thick when the stock is wrong. Thickness is easier to feel in a stack of five than in one card.
Two marks on modern cards that a counterfeiter has to reproduce
Type weight needs no equipment at all. Body text is set at one weight across a set, and type that has been scanned and printed again drifts off it: the letters fatten until the counters inside them start to close up, or they thin until a stroke drops out mid-word. Hold both cards at arm's length. The fake usually looks darker in the text box before you can say why.
Since the Magic 2015 core set, every rare and mythic rare carries a small oval holofoil stamp centered in the bottom border. On a genuine card the stamp shifts color as you tilt it and sits under the card's gloss. Fakes print it in flat silver ink that stays one color from every angle, stretch the oval a little too wide, or set it off center. Compare it against any real rare from a recent set.
Cards printed from 2018 onward also carry a line in the bottom-left corner of the front: collector number over set total, the three-letter set code, a rarity letter and the language. That line has to agree with the card above it. A counterfeiter working from one template and swapping the art produces a number that belongs to a different card, or a rarity letter that disagrees with the stamp.
Neither mark helps here. Revised predates both, which is why the dual land rests on the loupe.
A foil blocks the light test, so read the finish instead
A foil layer sits above the printed image, so it softens the rosettes and it kills the light test outright. Read the finish. A card sold as a traditional foil throws a rainbow across the whole face as you tilt it, where a flat silver sheen that holds one color is wrong for what it claims to be. The holofoil stamp and the type weight still work through foil, and on a modern rare those are the checks left.
The curve proves nothing in either direction. Foil and card stock do not settle at the same rate, so a genuine foil often bows along its length, and just as often it comes out flat after a few months pressed in a binder. Newer foils bow less than the old ones did. A flat foil is a flat foil.
The bend test goes last because it can cost you the card
Nothing so far has touched the card. This does. Bow the card gently along the short edge and genuine stock springs back flat with no memory of it, while brittle counterfeit stock creases along the bend or lifts at a corner. The tell is real. So is the crease.
Which is why it sits at the bottom of the page and stays off the dual land. A crease on a Revised dual land is money gone, and the loupe has already answered the question the bend would answer. Keep it for a card you have written off, or for one card out of a cheap bulk lot that tells you about the rest of the lot.
The cheap tests are the ones counterfeiters fixed first
Counterfeits are built against the published checklists. Core stock, a crisper back, a holofoil stamp that tilts: each was a reliable tell until enough people wrote it down, and each is now cheap to add. What survives is comparison, because a printer can copy a description of a real card and cannot copy the real card sitting on your desk. That is the whole argument for the known-real card and the loupe.
Which makes the purchase, not the inspection, the real defense on an expensive card. A graded copy has been through a company that authenticates before it grades and comes sealed with a certification number you can look up. A seller who takes returns is carrying the risk instead of you. A store with a name to lose is worth the few percent it charges over the cheapest listing. And a price far under the recent sold range for that printing is the oldest warning in the hobby.
Plenty of fake-looking Magic cards were never meant to fool anyone. A proxy is a stand-in so a player can test a deck without buying a dual land, usually marked with a note on the face, a different back or an obvious watermark. Wizards of the Coast has printed its own playtest cards as well. The difference is intent. A proxy announces itself, a counterfeit is priced and sold as the real card, and a seller who calls a proxy a proxy is not the problem.
A genuine card still needs its printing named before it has a price
The dual land passed and the return window can close. What is left is the price, and in Magic the price belongs to a printing rather than to a card name. One name can carry half a dozen set codes, and the regular frame, the borderless and the foil of that same card trade as three separate markets.
The set code and collector number in the corner, or the border and copyright line on a card old enough to lack them, decide which copy is in your hand. CardLens matches the exact printing from a photo and shows a value range built on recent eBay sales of that printing.
Those sales sit one tap below the number, so the range can be judged rather than taken on faith. Save the card and the vault total updates. Slab labels from PSA, BGS and CGC are read as well, so a graded copy is filed with its grade. The app is free on the App Store, with an optional premium subscription.
Questions collectors ask
How do you tell if a Magic card is fake or not?
Compare it with a genuine card from the same era under a 10x loupe. Real cards show rosettes of colored dots in the art and flat solid black in the text. Check the cut edge for the dark core, and the stamp and collector number on modern cards.
How can you tell if a foil Magic card is fake?
The foil layer blocks the light test and softens the print pattern, so work on the finish, the holofoil stamp and the type weight instead. A traditional foil throws a rainbow across the whole face as you tilt it. Ignore how the card curves: genuine foils bow, and plenty of them flatten out in a binder.
What is the rosette pattern on a Magic card?
It is the flower-shaped cluster of cyan, magenta, yellow and black dots left by an offset press, where each color is screened at a different angle. At 10x it repeats evenly across any mid-tone in the art. Counterfeits from digital printers show a square grid or a smooth spray instead.
Does the light test work on Magic cards?
Partly. A card that lets the artwork glow through has no core layer and is fake, so the test is worth the two seconds. A card that blocks the light has only passed the cheapest check, because counterfeit printers have used core stock for years. It cannot confirm a card.
How do I check a Magic card without damaging it?
Every check on this page except the bend test leaves the card as you found it. The lamp, the loupe, the cut edge, the stamp, the collector number and a size comparison against a known-real card are all non-destructive. Keep the bend test for a card you accept losing.
Free on the App Store
Match the printing and see what copies of it have sold for.
CardLens reads the set, the collector number and the finish from a photo, then shows a value range with the recent eBay sales behind it.