Japanese vs English Pokémon Cards: Centering, Borders and Gem Rates

Japanese and English Pokémon cards are graded against exactly the same centering standards. PSA's Gem Mint 10 line is approximately 55/45 on the front and 75/25 on the back for both, BGS wants 55/45 both ways for a 9.5 for both, and TAG's Gem Mint TCG band is approximately 55/45 front and 65/35 back for both. No grading company publishes a Japanese tolerance and an English tolerance.

What differs is the card. The English yellow border is roughly twice as wide as the Japanese border, and because centering is a ratio, that single geometric fact produces two opposite effects at once: the same ratio looks better on a Japanese card, and the same ratio is harder to measure on a Japanese card. Layer on top of that the reported gap in gem rates, and you get one of the most frequently asked questions in Pokémon collecting, usually phrased as "should I just buy Japanese?"

This article separates what is geometry, what is reported data, what is collector observation, and what any of it means for what you buy and what you send in.

The rules are identical; only the design changes

Start here, because a surprising number of guides get it wrong. Every published centering tolerance is written for cards, not for regions.

Figures from the companies' official standards pages, verified August 2026. The worse of the left/right and top/bottom axes governs; PSA states that centering is designated as the percent of difference at the most off-center part of the card. The full ladder is in the centering tolerance by grade table.

So when a Japanese card gems and an English card with the same measured ratio does not, the standard did not change. Either the rest of the card differed, or a grader exercised the discretion PSA publishes: its standards page allows a small variance at the grader's sole discretion based on the card's overall eye appeal, and states a 5% leeway on front centering minimums for cards grading PSA 7 or better.

The border-width difference, in millimeters

This is the whole mechanical story, and it is worth doing the arithmetic rather than repeating the slogan.

A standard English Pokémon card carries a yellow border of roughly 2.5 to 3 mm per side, so the combined left-plus-right border across an axis is roughly 5 to 6 mm. A Japanese card's border is much thinner, on the order of 1.5 mm per side, so roughly 3 mm combined. Both cards are the same physical size, 63.5 x 88.9 mm; the difference is entirely in how much of that size is border.

Two consequences follow, and they pull in opposite directions.

Cosmetically, the Japanese card wins. A 0.6 mm imbalance on a thin border is a subtle thing; a 1.2 mm imbalance on a fat high-contrast yellow band is the first thing your eye lands on. PSA's own published examples of eye-appeal discretion turn on exactly this: it describes a low-contrast 1955 design where off-centering is hard to see receiving an unqualified grade, and a high-contrast 1957 design where a card that technically measures for a grade may be held back because the market would view it as off-center. A thick yellow border against white card stock and a dark holo interior is about as high-contrast as trading card design gets.

Mechanically, the Japanese card is harder to measure. The same physics that hides the shift from a grader's eye hides it from your ruler. On a roughly 3 mm English border, a 0.25 mm reading error takes you from 50/50 to about 54/46, a four-point swing. On a roughly 1.5 mm Japanese border, the same 0.25 mm error takes you to about 58/42, an eight-point swing that crosses PSA's published 10 line. A handheld phone photo with two or three degrees of tilt can generate that much error on its own.

The practical rule that falls out: scan Japanese cards. A flatbed at 600 dpi has zero perspective error and a known scale of roughly 24 pixels per millimeter, and on a border this thin that precision is the difference between a decision and a guess. More conversions in what centering ratios mean in millimeters.

What the reported gem rates actually say

The numbers people quote in this debate come from community trackers, not from the grading companies, and they deserve to be handled carefully.

Start with what is not published. Neither PSA nor any of the community trackers breaks its gem rates out by language, so there is no Japanese-versus-English figure to quote. Collectors widely report that recent Japanese cards gem more often than their English counterparts, and that report is consistent enough to take seriously, but it is an observation rather than a measured statistic, and anyone who hands you a precise Japanese gem percentage has made it up.

What is published is the overall picture. PokeInvest, which derives its figures from PSA's population report, as of August 2026 and across more than 15 million graded Pokémon cards, puts the overall Pokémon PSA 10 rate at 43.9%, the modern (2023 and later) rate at 60.3%, and the vintage (pre-2003) rate at 9.9%. For industry-wide context, GemRate data reported by cllct put PSA's overall gem rate at 43% across about 8.9 million cards in the first half of 2025, with TCG cards at 50% and 2020s cards at 59%, against 34% for sports.

Three caveats before you use any of that.

These are submission statistics, not production statistics. A gem rate is computed on cards people chose to send in. Submitters pre-screen, and they pre-screen hardest on centering because it is the one criterion you can measure at home. So every gem rate you will ever see is inflated relative to what a random pack would produce, and it is inflated by different amounts for different products depending on how carefully that product's collectors screen.

They bundle all four criteria. Centering, corners, edges and surface all sit inside that one percentage. Card stock, gloss, and how a product is packaged and shipped all feed into corners and surface. It is entirely plausible that centering is the dominant term in the Japanese-vs-English gap, and it is not established by these numbers.

They are community-compiled. PSA does not publish gem rates by region. PokeInvest is deriving figures from the public pop report, and cllct is reporting GemRate's compilation. That is good, useful work, and it is not the same as an official statistic. Cite it as what it is.

What the numbers do support is a direction, and the direction matches what collectors report consistently: modern Japanese product is the easiest Pokémon to gem, modern English is next, and vintage English is by a wide margin the hardest. That vintage figure of 9.9% is the one worth staring at, and centering is usually the first thing that fails on a WOTC-era holo. The reasons live in why trading cards are off-center.

Why the two might genuinely differ, and what we cannot claim

Collectors offer several explanations for the gap. Here is what can be said honestly about each.

• Border geometry. Established. Thinner borders make the same ratio less visible, and PSA publishes a discretion clause tied to eye appeal. This one is not speculation.
• Separate print runs and facilities. Japanese and English Pokémon cards are produced for different markets on different schedules, and collectors widely report differences in card stock, gloss and cutting consistency between them. What nobody outside the manufacturers has is process data, so the sensible framing is: the products are made separately and behave differently, and no one can say from published information which step in the process dominates.
• Packaging and handling. Collectors report differences in how sealed product is packed and how much edge and corner wear cards arrive with. This affects the other three grading criteria, not centering, but it does affect gem rates, which is a reason not to read a gem-rate gap as a pure centering statistic.
• Grader behavior on Japanese backs. Some collectors report that PSA is harsher on Japanese backs. That is anecdote with no published support, and it should be treated as such.

What we will not do is attach set-level gem rates or centering statistics to specific Japanese sets. Those numbers get repeated confidently in collector spaces and there is no published source behind most of them.

Measuring each one correctly

The method is the same; the tolerances for error are not.

English cards. Measure the yellow border: from the physical cut edge to the line where yellow stops and the interior frame, art box or colored background begins, on all four sides. On ex, V and VMAX cards it is the silver band; on gold secret rares the gold band. On GX-era double-border designs, use the outermost border only; the inner frame is printed in a fixed relationship to the outer one and adds nothing except a way to fool your eye. Lay the card on a dark matte surface so the yellow edge is unmistakable.

Japanese cards. Same reference lines, higher standard of technique:

1. Card out of the sleeve. A penny sleeve seam is a meaningful fraction of a 1.5 mm border.
2. Scan at 600 dpi, or use a fixed straight-down mount. Do not freehand it.
3. Measure each border at two points along its length. On a thin border, a slightly skewed cut shows up as two readings that differ by more ratio points than you would expect.
4. Compute larger border divided by both combined, larger first. Take the worse axis.
5. Treat anything within three points of a threshold as unresolved and re-measure.

Both, on the back. English and Japanese backs both carry the blue frame around the central design. Measure cut edge to the inner edge of that frame on all four sides. Remember that PSA's back line for a 10 is 75/25, not 90/10; 90/10 is the PSA 9 back line, and a lot of submitters have it backwards. Full detail in does back centering matter?

A worked pair, to make the sensitivity concrete. An English card measures left 3.2 mm, right 2.8 mm: 3.2 / 6.0 = 53.3%, so 53/47, inside the PSA 10 line. A Japanese card measures left 1.7 mm, right 1.3 mm: 1.7 / 3.0 = 56.7%, so 57/43, outside it. The Japanese card's borders differ by 0.4 mm and the English card's by 0.4 mm as well. Identical physical shift, four points apart in ratio, on opposite sides of a published line. That is what "centering is a ratio" means in practice.

What this means for buying and submitting

• Do not buy a category, buy a copy. A gem rate of any kind is a statement about a population, and you are buying one card. Measure the copy in front of you; a thin border does not make a badly cut card good.
• Use the gap to set expectations, not strategy. If you are deciding whether a modern English card is worth a fee, a 60.3% modern gem rate is a useful prior. If you are deciding between two cards, the market for each matters far more than the odds of gemming.
• An easy 10 is a common 10. The flip side of a very high gem rate is a large population of 10s competing with yours. Scarcity is what makes a 10 worth a multiple of a 9; a card that gems nine times out of ten does not generate much scarcity. That mechanism deserves its own treatment, and it is the reason a high gem rate is not automatically good news for a seller.
• Screen from listing photos before you bid. For a Japanese card, a straight-on scan in the listing is worth far more than extra photos from angles. If the seller only has handheld shots, assume the worse axis is worse than it looks.
• Check the back on both. Fronts and backs are printed in separate passes and registered independently, so a tight front guarantees nothing about the back. A card with a 52/48 front and an 80/20 back fails PSA's 75/25 back line outright, in either language.

For where to put the guide lines on every Pokémon card style, English and Japanese, the full walkthrough is in the Pokémon card centering guide. The free card centering calculator measures both axes from a photo in the browser and maps the result to the PSA, BGS, CGC, SGC and TAG lines, with guide lines you drag onto the border yourself, which is what you want on a thin Japanese border where automatic edge detection has the least room to work. The full published thresholds are collected on the grading standards guide.

Bottom line

• The standards are identical. PSA 10 is approximately 55/45 front and 75/25 back for Japanese and English alike; no grader publishes a region-specific tolerance.
• The English yellow border is roughly twice as wide, so the same ratio produces about twice the visible millimeter gap. That is why an off-center English card looks worse than a Japanese card at the same number, and PSA publishes an eye-appeal discretion clause that can act on exactly that.
• The same geometry makes Japanese cards harder to measure: a 0.25 mm error is about four ratio points on an English border and about eight on a Japanese one. Scan them.
• Collectors widely report that Japanese cards gem more often, but nobody publishes a gem rate split by language, so there is no figure to cite. The published Pokémon numbers are overall: PokeInvest, deriving from PSA's pop report, gives 43.9% overall, 60.3% modern and 9.9% vintage as of August 2026. Those are community, population-derived figures that bundle all four criteria and are computed on cards people chose to submit.
• Separate print runs, card stock and packaging plausibly contribute, but no one outside the manufacturers has process data, so treat single-cause explanations as speculation.
• Measure the copy you are actually buying or submitting, both axes, front and back. The category tells you the odds; the card tells you the grade.

Frequently asked questions

Do Japanese Pokémon cards really gem more often than English ones?+

Collectors widely report that they do, but no grader or tracker publishes a Japanese-versus-English gem rate, so treat the gap as community observation rather than a measured figure. What is published is the overall picture: PokeInvest, deriving its figures from PSA's population report, puts the overall Pokémon PSA 10 rate at 43.9 percent, the modern 2023-and-later rate at 60.3 percent and the vintage pre-2003 rate at 9.9 percent as of August 2026. Those are population-derived community figures, not official PSA statistics, and they bundle centering with corners, edges and surface.

Is the centering standard different for Japanese Pokémon cards?+

No. PSA's Gem Mint 10 standard is approximately 55/45 on the front and 75/25 on the reverse for every card it grades, Japanese or English, and the same is true at BGS, CGC, SGC and TAG. There is no language-specific or region-specific tolerance published anywhere. What differs is the card design, not the rule applied to it.

Why does the same ratio look better on a Japanese card?+

Because centering is a ratio and the Japanese border is physically thinner. On an English card with roughly 6 mm of combined border across an axis, a 60/40 cut means 3.6 mm against 2.4 mm, a visible 1.2 mm difference. On a Japanese card with roughly 3 mm combined, the same 60/40 means 1.8 mm against 1.2 mm, a 0.6 mm difference that is easy to miss at arm's length. PSA's standards explicitly allow graders a small variance based on eye appeal, so how obvious a shift looks can matter.

Are Japanese Pokémon cards harder to measure?+

Yes, and this is the flip side of the thin border. On a roughly 3 mm English border, a 0.25 mm measurement error moves you from 50/50 to about 54/46. On a roughly 1.5 mm Japanese border, the same 0.25 mm error moves you to about 58/42. A handheld phone photo with a small tilt can introduce that much error, so scan Japanese cards or shoot from a fixed straight-down position.

Should I buy Japanese cards instead of English if I want PSA 10s?+

For modern product, the reported gem rates favor Japanese cards, but that is only half the decision. Population, market demand and resale differ by card and by set, and a PSA 10 that is easy to obtain is competing with many other PSA 10s. Use the gem-rate gap to set expectations about your odds, not as a buying rule on its own, and measure the specific copy in front of you rather than trusting the category.

Do Japanese and English Pokémon cards have the same back?+

Both carry the familiar blue back with a frame around the central design, and you measure it the same way on either: cut edge to the inner edge of the blue frame, all four sides. Back tolerances are the same for both because they are the same published standards: 75/25 for a PSA 10, 90/10 for a PSA 9, 60/40 for a BGS 9.5, approximately 65/35 for a TAG Gem Mint TCG card.

Sources & further reading

• PSA Grading Standards (official)
• Beckett Grading Scale (official)
• TAG Grading Rubric (official)
• PokeInvest Gem Rates (PSA population-derived, community data)
• cllct: How many cards grade as 10s? Gem rate at 43% so far in 2025 (GemRate data)