PSA centering tolerances by grade, with worked examples

Centering is the only part of PSA's standards with a number attached, which makes it the one thing you can check before you submit. It is also the attribute most people get wrong, because a card that looks centered to the eye can measure 58/42 and miss a 10. This article lays out the tolerances PSA publishes for every grade, shows how to measure, and walks through examples.

The tolerance table

Two things stand out. The back tolerance is loose for a 10 (75/25) and then constant at 90/10 for every grade from 9 down to 1.5, so the back almost never decides the grade below a 10. And the front tolerance widens by five points per grade at the top of the scale, which means a card's centering grade is decided within a few tenths of a millimetre.

What a ratio means

The ratio compares the two opposite borders. Measure the left border and the right border, add them, and express the larger one as a percentage of the total. Do the same for top and bottom. The worse of the two axes is the card's centering. A card that is 52/48 left to right and 61/39 top to bottom is a 61/39 card.

1. Measure the left and right borders at the same height, from the card edge to the edge of the printed frame or artwork. Use the same reference line on both sides.
2. Add them. Divide the larger border by the sum and multiply by 100.
3. Repeat for top and bottom.
4. Take the worse axis and compare it with the table.

Worked example 1: the near miss

Left border 2.0 mm, right border 1.6 mm. Sum 3.6 mm. The larger border is 2.0, and 2.0 divided by 3.6 is 0.556, so the card is 56/44 left to right. Top 1.8 mm, bottom 1.8 mm is 50/50. The card's centering is 56/44, which sits just outside PSA's 55/45 for a 10. PSA writes approximately and allows a small variance for eye appeal, so a card like this can go either way. If everything else is perfect, it is a coin flip on centering alone.

Worked example 2: the clean 9

Left 2.4 mm, right 1.6 mm. Sum 4.0 mm, larger border 2.4, ratio 60/40. This card meets the 9 tolerance exactly and is well outside the 10. It does not matter how sharp the corners are: the centering caps it at a 9, and a grader could reasonably put it at 8 if they measure the borders slightly differently than you did. On the plus side, a 60/40 card that is otherwise flawless is a strong candidate for the half-point bump PSA describes for high-end cards within a grade.

Worked example 3: the back that matters

Front 53/47 both axes, a textbook 10 front. Back: top border 3.1 mm, bottom border 0.9 mm. Sum 4.0, ratio 77.5/22.5, which rounds to 78/22. That is outside the 75/25 back tolerance for a 10, so the ceiling is a 9. Vintage sets and some modern Japanese prints are often cut with a strong back offset, and this is the single most common way a card that photographs perfectly from the front fails to gem.

Measuring from a photo instead of a ruler

A ruler works, but a 0.2 mm read error swings the ratio by several points on a card with 2 mm borders. A photo taken straight down, with the whole card in frame, lets you measure in pixels instead, which is more repeatable. The method is the same: find the card edge and the frame edge on each side, count pixels, compute the ratio. The two things that ruin a photo measurement are perspective (the camera not square to the card) and a frame edge that is hard to find, such as full-art Pokemon cards and borderless parallels where the printed edge is not a straight line.

Three cards that trip people up

• Rotated cuts: the left border is wider at the top than at the bottom. Measure at the middle of the card and note the tilt; a visibly rotated cut hurts eye appeal even when the midpoint ratio passes.
• Yellow-bordered Pokemon: the yellow border is the frame, not the card edge. Measure the yellow, not the artwork box inside it.
• Chrome and Prizm parallels: the reflective surface makes the frame edge jump in photos. Shoot under diffuse light, never flash.

What this means for a submission decision

Because centering is decided by numbers you can measure at home, it should be the first check on any card you are thinking of submitting. A card outside 60/40 on the front cannot be a 10 and cannot be a 9. If the price gap you are chasing lives between the 9 and the 10, and the centering says 9 at best, the grading fee is money you do not need to spend. See the companion piece on grading ROI for the arithmetic.

GradePeek measures both axes on the front and, when you upload it, the back, and reports the ratio alongside its corner, edge and surface scores. Try it on a card at /scan. It is an estimate, not a grade, and we are not affiliated with PSA.

Why the back decides so many 10s

Collectors measure the front and stop. That habit quietly rejects good cards. The back tolerance for a 10 is 75/25, which sounds generous until you remember how backs are cut: the printing on the reverse is often placed with less care than the art on the front, and vintage prints in particular were cut in stacks where the back drift moved with the sheet. A front that measures 53/47 and a back that measures 76/24 is a 9, and there is no way to see the back problem without flipping the card over and measuring it.

The reverse asymmetry also compounds. Because the 9 tolerance for the back is 90/10, a back that misses the 10 tolerance almost never costs you the 9. The realistic damage of a bad back is exactly one grade: it converts a 10 into a 9. That is still enough to matter on any modern card, where the 10 is the only grade worth chasing, but it should also stop you from discarding a card entirely. A card with a soft back is not a bad card; it is a card whose ceiling you now know.

Practising without spending a fee

• Measure ten cards from a recent bulk lot against each other. The spread of ratios you record tells you how repeatable your own measuring is before you trust it on a submission candidate.
• Check your measurement against a slab you own. PSA prints the grade on the label, not the ratio, but a PSA 9 slab you measure at 63/37 front is a useful calibration point.
• Write the four numbers down (left, right, top, bottom, both sides) instead of the ratio. When a card comes back a grade different from your estimate, the raw numbers are the only way to learn what your error was.
• Measure the worst axis twice. If the two readings differ by more than a point, your reference edge is not clean; find the actual printed frame before you trust either reading.

One last caution on the numbers themselves. PSA writes approximately on most grades and reserves the right, at the grader's sole discretion, to permit a small variance based on the card's overall eye appeal. A visibly rotated or skewed cut can be judged harsher than its midpoint ratio suggests, and a beautiful card that measures a hair outside can be judged kinder. The ratios are the best public guide to the decision, not the decision itself.