Why Are Cards Off-Center? Printing, Cutting and Sheet Layout Explained

Cards are off-center because printing and cutting are two separate operations, each with its own tolerance, and centering is the mismatch between them. The design is printed onto a large sheet, and a cutter then slices the sheet into 63.5 × 88.9 mm (2.5 × 3.5 in) rectangles. If the image lands 0.2 mm to the left of where the layout expects and the cut lands 0.2 mm to the right, you get a card whose left border is 0.4 mm wider than its right. On a 2 mm border that is 2.4 mm vs 1.6 mm, which is 60/40, the PSA 9 line.

That is the whole story in one paragraph. A few tenths of a millimeter is excellent precision for an industrial press running thousands of sheets per hour, and it is a large fraction of a border that is only 2 to 3 mm wide. Everything else in this article is detail: which step contributes what, why certain sheet positions and certain print runs are consistently worse, what a diamond cut is, why the back is usually worse than the front, and why none of it can be fixed after the fact.

If your immediate question is "how off is my card", measure it with the free card centering calculator and compare against the grading standards guide. If your question is why it came out of the pack that way, read on.

Step 1: the sheet is printed, and the image has to land in the right place

Trading cards are printed many-up on large sheets. Each card design sits in a fixed position in the layout, and each sheet passes through the press several times, or several units of a press, once for each ink color and again for foil, varnish or texture layers where used. Registration is the term for how precisely those layers line up with each other and with the sheet. CGC's Pristine 10 wording asks for "flawless color and registration" for the same reason: a slip between layers is visible as color fringing, and a slip of the whole image relative to the sheet edge becomes a centering error after cutting.

Two things can go wrong at this stage, and both are small:

• The image as a whole can be placed slightly off its nominal position on the sheet. This shifts every card on that sheet in the same direction, which is why a whole print run can lean the same way.
• The sheet can feed through slightly skewed, so the image is rotated by a fraction of a degree relative to the sheet edges. This is one of the two origins of the diamond cut.

The front and the back are printed in separate passes, often on separate sides of the sheet in different runs, so the back image has its own placement error relative to the front. Hold that thought for the back-centering section.

Step 2: the sheet is cut, and the cutter has its own tolerance

Modern production typically uses rotary die cutting: the printed sheet passes through a drum carrying a die in the exact size and shape of the cards. Older production, and some still today, uses guillotine-style straight cuts. Either way, the cutter has to be aligned to the printed image, and the alignment is never perfect.

A useful description of the classic process comes from a long-running collector discussion of how Topps cut cards: the sheet was gripped on the left side while horizontal slits were cut for the rows, and then, working from right to left, a vertical blade cut off a column of cards at a time. The same thread lists the failure modes that every press operator knows: slight misalignments as the ink goes onto the paper, sheets that were not squared when loaded, and blades dulling over a run. A sheet that is not flush when it is clamped produces diamond-cut cards, and cards at the edge of the sheet are described as the toughest to find centered, because cumulative error builds up across the sheet and the outer rows and columns absorb it.

Whatever the exact machine, the outcome is the same: the cut has its own error, independent of the print's error, and they add.

The arithmetic of tolerance stack-up

Manufacturers do not publish their placement tolerances, so the numbers below are illustrative, but the shape of the math is what matters. Suppose the image placement on the sheet is accurate to ±0.2 mm and the cut is accurate to ±0.3 mm. In the worst case they add to 0.5 mm of shift on one axis. Here is what that identical 0.5 mm shift does to cards with different border widths:

Same factory error, four different grades. This is the single most important thing to understand about off-center cards: the factory is not sloppier on thin-border sets, the same absolute error is simply a larger share of a thinner border. And a typical shift that is a mix of print and cut error, say 0.2 to 0.3 mm, is exactly the range that turns a 50/50 design into a 55/45 to 60/40 card on a standard border, which is why so many pack-fresh cards sit right on the PSA 10 line of approximately 55/45.

Sheet position: why some card numbers are always off

Because error accumulates across a sheet and the outer rows and columns absorb clamping and feeding error, cards from particular sheet positions are consistently worse than others. PSA's own article on the 1986-87 Fleer basketball set makes this concrete. It names centering and chipping as the set's "two main culprits", says "the serious 1986-87 Fleer set collector knows that the centering is the key", and notes that "poor centering, in some cases, will affect the common players even more so than the stars, depending on where the card was located on the sheet". It quotes set specialist Daniel Chang identifying cards 1, 11, 22, 33, 44, 55, 66, 77, 88, 99, 111, 121 and 132 as the most affected "because they were at the end of the sheet", adding that "they seem to always be off-center, have print problems and edge damage".

That is the pattern to look for in any set: a systematic lean tied to sheet position, with random sheet-to-sheet variation layered on top. It is why collectors report things like "every copy of this promo I have ever seen measures about 58/42": when the image placement for a run is off by a fixed amount, every copy inherits it, and only the cut's random component varies. It also explains why a population report can show one card number in a set with a dramatically lower gem rate than its neighbors without the card being any rarer or more handled.

Practical consequence: if the first three copies you check of a particular card all lean the same way, the fourth probably does too. Measure before you buy, rather than assuming the next one will be centered.

Diamond cuts and skew

A diamond cut is a card whose cut is rotated relative to the printed design. The borders are not just unequal, they taper: the left border might be 2.6 mm at the top and 1.9 mm at the bottom. The card looks subtly like a rhombus in the holder. It comes from a skewed sheet in the press, a skewed sheet in the cutter, or a cutting angle that drifted during a run.

Graders treat it as its own defect and tolerate it on a sliding scale. Beckett's published scale is the most explicit:

SGC's scale notes at 7 that a "card may exhibit a slightly skewed (diamond) cut", and CGC's 4.5 description allows "an extremely noticeable diamond cut". None of those companies publish a diamond-cut allowance for their 9s and 10s, which tells you what you need to know: at the top of the scale, a visible skew is a problem in itself.

For you as a measurer, the lesson is to take two readings per side, near each end of the edge. If they differ by more than about 0.2 mm, you are looking at skew, and a single centering ratio understates the problem.

Why the back is usually worse than the front

Collectors ask this constantly, and the answer is mechanical. The back is printed in a separate pass from the front, so there are two independent placement errors on the sheet, one per side. A single cut then has to serve both. Cutters are aligned to the front, so the front gets the cut error alone while the back gets the cut error plus the front-to-back registration error. More contributing errors, more spread.

Graders acknowledge this by publishing looser tolerances for the reverse:

Notice that Beckett's back tolerance at the very top is much tighter than PSA's or CGC's. If your front measures well and your back is 70/30, that is a PSA 10 candidate and not a BGS 10 or 9.5 candidate, which is the kind of decision the back number should drive. Whether it matters for your card is covered in does back centering matter.

Thin borders, eras and factories: what is reported vs what is known

A few patterns come up in every forum thread, and it is worth separating the known from the reported.

• Border width changes how the same error looks. This is arithmetic, not opinion: the stack-up table above shows a 0.5 mm shift landing anywhere from 58/42 to 75/25 depending on border width. A thin border makes centering both harder to hit on paper and, oddly, easier on the eye, because a 0.3 mm shift on a 1.5 mm border is still only 0.3 mm of visible asymmetry.
• Japanese vs English Pokémon. Collectors widely report that Japanese cards, with their thinner borders, gem at higher rates than English cards, and PokeInvest, which tracks PSA population data, states that recent Japanese cards routinely gem above 90 percent. The same source puts the overall PSA 10 rate for Pokémon at 43.9 percent, the modern (2023 and later) gem rate at 60.3 percent, and the vintage (pre-2003) gem rate at 9.9 percent as of August 2026. Those rates bundle centering with corners, edges and surface, so treat them as context rather than a centering statistic. Different factories, different eras and different border geometries all plausibly contribute, and none of the manufacturers publish process data that would let anyone say which dominates.
• Vintage and WOTC-era cutting. Older production is widely reported to be less consistent, and the sheet-cutting description above explains why blade wear and manual sheet loading would matter more on older equipment. PSA's published tolerances for mid grades, 70/30 at 7, 80/20 at 6 and 85/15 at 5, exist largely because a huge share of vintage cards need them. The vintage card centering guide goes through those grades.
• Modern chrome and borderless designs. The factory error is no different, but the visible reference is thinner, so the same shift reads as a worse ratio, and collectors report high centering-related 10 failure rates on some borderless sports products. Treat those reports as anecdotal; the mechanism, not the statistic, is the reliable part.

For where to measure on English and Japanese Pokémon frames specifically, see the Pokémon card centering guide.

Why you cannot fix it

Centering is a property of the cut, so the only physical way to change it would be to cut the card again, which is trimming. PSA's published standards list Evidence of Trimming as a no-grade reason (its N1 category), and every major grader treats a trimmed card as altered and ineligible for a numeric grade. Nothing else touches centering: pressing addresses surface waviness, cleaning addresses residue, and a sleeve or a slab changes nothing about where the borders are.

It also cannot be appealed away. An off-center card that is otherwise perfect may, per PSA's published policy, receive a numerical grade that reflects the minimum centering allowed for that grade, or in extreme cases an OC qualifier, and a card cut so far off that part of another card or the sheet alignment marks show is a miscut, which PSA designates MC or, when "the factory cut is abnormal for the issue", may return ungraded as N8 without charging the grading fee. The line between an off-center card and a true miscut, and what each does to value, is in off-center vs miscut: OC and MC qualifiers explained.

So the only fix is selection: pull or buy the best-centered copy and leave the rest raw. Measuring takes a minute, and because a single pack, box or print run tends to share a lean, checking the first few copies tells you quickly whether a well-centered example is realistic for that card.

Bottom line

• Cards are off-center because printing and cutting are separate operations with separate tolerances, and centering is the mismatch between where the image landed and where the cut landed.
• A shift of a few tenths of a millimeter is normal for a press and is a big fraction of a 2 to 3 mm border; the same 0.5 mm shift is 58/42 on a 3 mm border and 75/25 on a 1 mm frame line.
• Error accumulates across a sheet, so edge-of-sheet positions are consistently worse, and whole print runs can lean the same way; PSA's own 1986-87 Fleer write-up names the affected card numbers.
• The back is usually worse because it adds a front-to-back registration error to the cut error, which is why published back tolerances are looser.
• Diamond cuts come from skew in the press or cutter and are tolerated only in lower grades; measure two points per side to catch them.
• Centering cannot be fixed; trimming makes a card altered and ungradable, so measure and select instead.

Frequently asked questions

Why is my Pokémon card off-center?+

Because the card was printed on a large sheet and then cut out, and the cut did not land exactly where the design expected it. Printing places the image with one small tolerance and the cutter adds another; the sum shows up as one border wider than its opposite. A shift of a few tenths of a millimeter is normal in industrial printing but is a large fraction of a 2 to 3 mm border, which is why so many pack-fresh cards measure 55/45 or worse.

Can real cards be off-center, or is off-centering a sign of a fake?+

Real cards are off-center all the time, and an off-center border on its own says nothing about authenticity. Off-centering is a normal result of sheet cutting tolerances at every manufacturer and in every era. Fakes are identified by print quality, color, font, card stock, texture and light test, not by centering. If anything, a dramatic miscut with part of a neighboring card visible is usually genuine, since counterfeiters rarely reproduce factory errors.

Why is the back of a card often more off-center than the front?+

The front and back are printed in separate passes, so the back image has its own placement tolerance relative to the sheet, and then the single cut has to satisfy both sides at once. The cutter is typically aligned to the front, so the back carries the front's error plus the front-to-back registration error. Graders know this and publish looser back tolerances, for example 75/25 on the reverse for a PSA 10 against approximately 55/45 on the front.

Can off-center cards be fixed?+

No. Centering is set by the cut and the only way to change it is to trim the card, which makes it altered. PSA lists Evidence of Trimming as a no-grade reason, and every major grader treats trimmed cards as altered and ineligible for a numeric grade. Pressing, sleeving or cleaning does nothing to centering. The only fix is to buy or pull a better-centered copy, which is why measuring before you submit matters.

Are all cards from the same print run off-center the same way?+

Often partly, yes. Cards from the same sheet position tend to share a systematic offset, so collectors notice that nearly every copy of a particular promo or a particular card number leans the same direction. Cards at the edges of the sheet are widely reported to be the worst. On top of that systematic offset there is random variation from sheet to sheet, so two copies of the same card can still measure quite differently.

What is a diamond cut card?+

A diamond cut is a card whose edges are not square with the printed design, so a border is wider at one end than the other, giving the card a slightly skewed or rhombus look. It happens when the sheet is not flush in the cutter or the cutting angle drifts. Graders tolerate it on a sliding scale; Beckett, for instance, allows very slight diamond cutting at BGS 7 and moderate diamond cutting at BGS 4, and SGC notes a slightly skewed cut may exist at 7.

Sources & further reading

• PSA - Modern Masterpiece: A Look at the 1986-87 Fleer Basketball Card Set (official)
• PSA Grading Standards (official)
• Beckett Grading Scale (official)
• SGC Grading Scale (official)
• Collectors Universe forum - How did Topps cut cards?
• PokeInvest - PSA gem rates