Card Centering Tool

PSA centering tolerances (front)

Centering sets the highest grade a card can reach. These are the widely-cited PSA front-centering thresholds, measured on the worse of the two axes:

• PSA 10 (Gem Mint) — about 55/45 or better
• PSA 9 (Mint) — about 60/40 or better
• PSA 8 (NM-MT) — about 65/35 or better
• PSA 7 (NM) — about 70/30 or better

Backs are judged more leniently — a Gem Mint generally allows around 75/25 on the reverse. Every figure here is approximate on purpose: a grader reads the card in hand, and these are the conservative end of published bands, not a contract.

The worse axis is the one that counts

This is where most people mis-read their own card. Centering is not one number — it is two independent measurements, left-to-right and top-to-bottom, and the tolerance has to be met on each of them. A card that is a flawless 50/50 horizontally and 62/38 vertically is a 62/38 card for grading purposes. It does not average out to 56/44, and there is no credit for the good axis.

That single misunderstanding is responsible for a lot of surprised 9s on modern Brunson chrome. The tool above reports both axes and then reports the worst one separately, because the worst one is the only one that sets your ceiling.

A reading on the line is not a pass

If your worst axis comes back at exactly 55/45, the honest reading is not "it is a 10" — it is "it is on the line with zero margin." Your measurement has error in it from several directions at once: a scan that sits a degree off square stretches one border relative to its opposite, soft or gradient border edges make the endpoint a judgement call, and rounding to the nearest tenth of a unit moves the ratio on its own. Any one of those can shift a reading by a point or two, and the grader is measuring their look at your card rather than your scan of it.

Treat anything within roughly two points of a threshold as genuinely uncertain, in either direction. A card reading 56/44 is not a comfortable 9 — it is a 10 missed by a point, which is well inside the error of your own measurement. That is why the tool prints the distance to the nearest threshold beside the verdict and marks it when it falls inside that band: on a borderline card, the gap is the more useful number of the two.

Do not skip the back

The reverse has its own tolerance, and while it is a wide one, it is not automatic. A card can clear the front threshold comfortably and still miss on a badly-shifted back, and that is a demoralising way to find out — after you have paid the fee. The optional back block in the tool checks your reverse against the ~75/25 figure commonly cited for a Gem Mint and flags it inside or outside.

It is reported as its own line rather than folded into the headline verdict deliberately. The Gem Mint back figure is widely published; per-grade back bands below it are not something we can source, so we do not print numbers for them.

How to measure accurately

• Scan straight-on at high resolution, or use calipers on the physical card. A tilted scan is the single largest source of false readings, because it lengthens one border and shortens the one opposite it.
• Measure to the same reference on each side — typically the edge of the printed image to the edge of the card. Switching reference points between the left and right border invents a miss that is not there.
• Do both axes. Horizontal alone is half an answer.
• Be consistent with units. Ratios are what matter, not the absolute numbers — millimeters and pixels give the identical result as long as all four borders use the same one.
• Measure before you buy, not just before you grade. On a raw Brunson rookie listed with a straight-on photo, the same four measurements tell you whether the copy you are bidding on has a ceiling worth paying for.

What centering cannot tell you

Centering is a ceiling and nothing more. It rules cards out reliably — a 68/32 front is not going to be a 10 however clean the rest of it is — but it never rules a card in, because corners, edges and surface are graded alongside it and any one of them can pull the number down on its own. On refractive stock in particular, surface is usually what does the damage: chromium shows print lines and scratches that plain card stock hides, which is part of why Brunson's Prizm and Optic rookies miss a 10 more often than the base Donruss paper version.

For the other three factors and the submission process itself, see should you grade your Jalen Brunson cards. And once you know the grade is achievable, run the card grade estimator for a weakest-link read across all four factors, then the grading ROI calculator to see whether it pays.