Is this card worth grading? The ROI math with current PSA fees

Grading a card is a bet. You pay a fee, wait, and receive a number that sets the card's price. Whether the bet is worth placing depends on three things you can look up and one you have to estimate: the fee, the price of the card raw, the price in each grade, and the probability of each grade. This article does the arithmetic on two real cards with prices read on 7 October 2026.

The fee, as of October 2026

PSA's pricing page (psacard.com/services), re-checked on 8 October 2026, lists the card tiers as follows. The Value service levels (Value Bulk, Value and Value Plus, each insured to $500) are temporarily paused. Standard is $59.99 per card with a maximum insured value of $1,000 and an estimated turnaround of 90 to 100 business days. Priority is $79.99 ($1,500, 70 to 80 business days). Express is $199 ($2,500, 20 to 30 business days), Super Express is $349 ($5,000, 10 to 15 business days) and Premier is $599 ($10,000, 7 to 10 business days). Above that sit the Premium tiers, from $999 for cards insured to $25,000 up to $9,999 and more. Every card tier includes imaging of the front and back of the finished slab, and the page states plainly that the turnaround times are estimates, not guarantees. $59.99 is the cheapest fee listed today.

Card one: 2018 Panini Prizm Luka Doncic #280 base

Sports Card Pro's price page for this card, read on 7 October 2026, listed: ungraded $55.00, grade 7 $44.86, grade 8 $46.49, grade 9 $79.03, grade 9.5 $110.50, PSA 10 $201.00. Its recent raw eBay sales for September and October 2026 ran from $41 to $65.

Only one outcome makes money, and it makes $86. Every other grade loses money on the fee alone, and a graded 8 sells for less than the raw card does. To break even on expected value, assuming every non-10 comes back a 9 (optimistic), the card needs a probability of a 10 above about 30 percent: 0.30 times $201 plus 0.70 times $79.03 is $115.62, minus $59.99 is $55.63. If half the non-10s are 8s, the required probability climbs to about 38 percent. A base Prizm with any corner softness or centering outside 55/45 does not clear that bar. This card is a keep-raw card unless it is a near-certain 10.

Card two: 1999 Pokemon Base Set Charizard #4

PriceCharting's page for the Base Set Charizard, read on 7 October 2026, listed: ungraded $446.29, grade 7 $787.50, grade 8 $1,350.00, grade 9 $3,309.38, grade 9.5 $7,888.50, PSA 10 $24,077.50. Better Pop's August 2026 analysis of the same card used a raw price of $1,300, a PSA 9 price of $3,071 and a PSA 10 price of $28,145. The two raw figures differ a lot because PriceCharting's ungraded average includes played copies; a raw copy that is a plausible 9 candidate sells closer to Better Pop's figure.

Here the fee is noise. A 9 pays for itself several times over, an 8 sells for a little less than a strong raw copy, and a 10 is a different asset class. The question is not whether to grade, it is which grade the card can actually get. Note the service level: PSA's insurance ceilings mean you cannot declare this card at $3,309 on Standard ($1,000), Priority ($1,500) or even Express ($2,500). Declaring it at the 9 price requires Super Express at $349. Declaring it at the 10 price of $24,077 fits inside Premium 1's $25,000 ceiling, so the cheapest tier that covers it costs $999. The fee still does not change the answer.

The gem-rate problem

A PSA 10 Base Set Charizard is priced at over seven times a 9 because almost none exist. Better Pop's pop-report pull on 1 October 2026 counted 106,701 unlimited Base Set Charizards graded by PSA, of which 491 were PSA 10, 8,547 PSA 9 and 16,446 PSA 8, a gem rate of 0.46 percent. Pikawiz's Base Set pop page, read the same week, showed 489 PSA 10, 8,417 PSA 9, 15,339 PSA 8, 15,750 PSA 7 and 16,306 PSA 6 out of 92,856; the totals differ because the two sites count variants differently, but the shape is the same. The most common grade for this card is a 6, 7 or 8, and a 10 is a rounding error.

That pop is cumulative and includes every beat-up copy ever submitted, so it understates the odds for a carefully selected raw card. It still tells you the honest prior: for a 27-year-old card, assume 8 unless the card gives you a specific reason to assume 9, and never price your decision on a 10.

A simple rule

1. Look up the raw price and the price at each grade from 7 to 10 on a site that separates them.
2. Find the lowest grade at which the price, minus the fee and shipping, beats the raw price. Call it the floor grade.
3. Inspect the card against PSA's standards: centering ratio on both sides, corners under magnification, edges, surface under raking light.
4. Submit only if you are confident the card clears the floor grade. If the floor grade is 10, you need a 55/45 front, a 75/25 back, four perfect corners and a clean surface, and even then PSA's standards allow grader discretion.

What PSA says about typical outcomes

PSA's own get-started page gives a useful calibration: it says 90 percent of ultra-modern sports cards (2017 to present) and modern TCG cards (2003 to present) have graded at PSA 8 or higher, 80 percent of modern sports cards (1986 to 2016) at PSA 8 or higher, and 80 percent of vintage TCG cards (pre-2003) at PSA 7 or higher. Those are floors, not ceilings. An 8 or higher includes a great many 8s, and for a modern base card an 8 is usually worth less than raw.

Where an estimate fits

GradePeek is built for step three. It measures centering on both sides, scores corners, edges and surface from your photos, and tells you which attribute is the limiting one. If the limiting attribute caps the card below your floor grade, you have saved a fee and three months. It is an estimate, not a grade, and we are not affiliated with PSA.

Run the card through /scan before you fill out a submission form. Three estimates a month are free.