PSA, BGS, CGC and SGC centering: standards compared

How do the Gem Mint front and back limits compare?

The table separates the worse and better front axes. For houses without a two-axis combination rule, both axes must remain inside the same limit. “Back” is a separate check; passing the front does not make an unmeasured reverse acceptable. SGC's blank reverse entry means this model lacks a defensible separate numerical rule, not that any reverse is acceptable to SGC.

Limits apply to both axes separately. For BGS front, the worse axis must meet the loose limit and the other axis the tight limit. — means no separate numerical back rule in this model. A ceiling is not an overall grade.

These ceilings cover centering alone. Read the front-versus-back examples before combining measurements, and retain the evidence status when quoting a limit.

Which PSA bands does the calculator actually use?

The PSA table follows the numerical reference recorded in the model. It intentionally does not manufacture a separate threshold for every possible grade. If a ratio lands in a coarse band, the listed ceiling is the output of this reference; it is not a claim that intermediate final grades never occur. The PSA grading standards also describe condition requirements beyond centering.

Bands are intentionally incomplete where numerical evidence is missing. Do not interpolate unlisted grades. Status applies to each row, not just to the house.

Why can the same worst ratio produce different BGS ceilings?

For BGS front centering, sort the axes by imbalance, then check the worse axis against the loose limit and the other against the tight limit. The axes can swap: horizontal is not inherently the loose axis. A single “worst ratio” discards information needed by this rule.

Example one: 55/45 horizontally and 50/50 vertically gives 9.5 Gem Mint. Example two: leave the horizontal ratio at 55/45 but change the vertical to 55/45; the ceiling becomes 9 Mint. The physical difference is on the second axis, even though both cards have the same worst ratio.

Bands are intentionally incomplete where numerical evidence is missing. Do not interpolate unlisted grades. Status applies to each row, not just to the house.

The upper BGS bands are interpreted; lower ones are approximate. The source endpoint was unavailable at the recorded verification, so these are explicit modelling choices rather than a freshly confirmed official numerical scale. Neither this calculation nor a centering subgrade alone establishes the overall BGS label. The ratio guide includes a compact two-axis example table.

Is Pristine 10 the same as Gem Mint 10?

A numerical 10 does not uniquely identify a designation. CGC and SGC distinguish Pristine from Gem Mint, while BGS Gem Mint is a different numerical band from Pristine. Keep the label alongside the number in any comparison.

CGC's published Pristine description asks for perfect centering. Centering Lab uses 52/48 as a practical measurement tolerance for its perfect-centering model; that is an interpretation, not a published allowance that guarantees Pristine. The CGC reverse bands do not establish a separate numerical Pristine reverse rule. A model result retaining the Pristine front designation is therefore conditional, not proof of full Pristine eligibility.

Bands are intentionally incomplete where numerical evidence is missing. Do not interpolate unlisted grades. Status applies to each row, not just to the house.

CGC explicitly describes several lower numerical limits for sports and non-sports cards. Applying those bands to TCGs remains interpreted here. The model also omits some intermediate grades, so do not interpolate them or compare an older label without checking the service's current scale and legacy key.

What can and cannot be concluded from the SGC table?

All numerical SGC bands in this reference are approximate. Its official scale link and the separately identified secondary reference allow readers to inspect the evidence instead of mistaking the table for a promise from the house. Pristine uses the same model convention discussed above. The absence of a reverse band prevents a full numerical reverse screening.

Bands are intentionally incomplete where numerical evidence is missing. Do not interpolate unlisted grades. Status applies to each row, not just to the house.

A tool can display a front ceiling while lacking a back rule, but the scope must remain visible. Do not translate “not modelled” into “passed”. If the reverse is visibly displaced, consult SGC directly and retain that uncertainty in your submission decision.

Why is TAG not included, and how should these sources be quoted?

TAG publishes a grading rubric, including numerical centering criteria and distinctions between sports cards and TCGs. TAG is absent because its scale and precision behaviour have not been implemented and verified in Centering Lab's four-house model. Its omission does not mean TAG lacks public standards, and borrowing another house's row would give an unsupported answer.

Use official for a recorded published numerical rule, interpreted for a measurable translation or an extension of incomplete evidence, and approximate for a working estimate. Verification status is a separate field: retrieving a source does not turn every model row into an official rule. The standards page gives sources and the recorded verification date; the method page explains rounding and measurement uncertainty. Check the live house policy before a submission if a particular boundary matters.

Frequently asked questions