How to price a trading card: the five-source comp method

Most dealers price cards using one source: the "Sold" filter on eBay. That single check is fine for $3 commons. It's wrong for everything above $20, and it's catastrophic for graded cards or anything regional, language-variant, or short-printed.

I check five sources on every card priced above $20. They take about 90 seconds combined once you have the workflow down. They've saved me four-figure mistakes more than once.

The five sources, in order

1. PriceCharting median — anchor.
2. eBay sold comps, last 30 days, condition-matched — the floor of "what will actually clear today."
3. eBay active asks, condition-matched — the ceiling.
4. TCGPlayer Market Price — a sanity check on raw cards.
5. 130point or live sold-comp tools for graded cards — the truth for slabs.

Order matters. If I check eBay first I anchor to whatever the most recent buyer paid — which might be a panic sale, a private negotiation, or a $0.99 auction that closed at $40 because two specific buyers showed up. PriceCharting median gives me the frame first; eBay comps confirm or break it.

Source 1 — PriceCharting median

PriceCharting is the closest thing the hobby has to a Blue Book. They aggregate sold comps across multiple marketplaces and surface a median by condition tier (Ungraded, Grade 7, Grade 8, Grade 9, Grade 9.5, Grade 10).

What I use it for:

• The median is my anchor price. If everything else clusters around it, that's my listing price.
• Their "loose" / "complete" / "new" distinction is a sanity check on condition. A raw NM card should price between "Loose" and the next tier up.
• They cover Japanese sets. Most US competitors don't.

What I don't trust it for:

• Cards under $5 — too noisy.
• Anything sold in the last 48 hours — their crawl has a lag.
• Cards that just spiked from a Whatnot show or a YouTube break — the median lags reality by days.

Source 2 — eBay sold comps, last 30 days, condition-matched

Open eBay. Search the exact card name + set + number. Filter to Sold, last 30 days. Then — and this is the part most dealers skip — open the first five sold listings and verify each one is condition-matched to your card.

A PSA 9 sold at $180 is not the same data point as a raw NM card. A holo with a back scratch sold at $40 is not the same data point as a near-mint holo. You're looking at five-to-ten matched comps, not the top of the unfiltered list.

What this tells me:

• The floor. If five matched NM comps cleared at $40-$48 in the last 30 days, $42 is a price that will sell this week. $55 is a price that might sit for a month.

What to watch out for:

• BIN-with-Best-Offer sales display the accepted offer, not the asking price. Click into them.
• International buyers paying in non-USD currencies sometimes show inflated USD totals. Verify with a few comps.
• Promoted-listing fees aren't in the displayed price but they're real money out of your pocket. Adjust your floor by the promo percent you run.

Source 3 — eBay active asks, condition-matched

Same search, but uncheck Sold and look at Buy It Now actives sorted by price + newly listed. The lowest 5-10 condition-matched actives are the ceiling. If the lowest comparable BIN is $52, you can't list at $80 and expect a sale — buyers see the cheaper one first.

This is also where you find the over-priced cards that have been sitting for 90+ days. If three sellers are at $80 and nothing has sold above $50 in the last month, the $80 listings are wishful thinking and you should ignore them when setting your price.

What active asks tell me:

• The competitive ceiling.
• Which sellers are dumping (lowest 1-2 actives, often relisted weekly).
• Whether the card is in undersupply (only 1-2 actives = pricing power) or oversupply (15+ actives = race to the bottom).

Source 4 — TCGPlayer Market Price

For raw Pokemon and Magic cards, TCGPlayer's Market Price is the second-most-important data point after eBay. TCGPlayer's audience is different from eBay's — more LGS-flavored, more set-builder, slightly lower average price tolerance on commons but slightly higher willingness on chase cards.

When I check it:

• Raw Pokemon / Magic cards above $5.
• Anything I plan to cross-list to TCGPlayer (~25% of my volume above $10).

What it tells me:

• If TCGPlayer Market and eBay 30-day median diverge by more than 20%, something is happening. Usually it's a TCGPlayer set-builder spike that hasn't hit eBay yet, or vice versa. Investigate before pricing.

Skip for: sports, sealed product, slabs (TCGPlayer doesn't index graded), Japanese (limited coverage), Pokemon promos with low sample size.

Source 5 — 130point and graded-comp tools

For graded slabs only. eBay's sold filter is useful but limited — 130point shows you the cert-resolved comp history when available, including private auctions and PWCC weekly sales.

Other graded-comp tools worth the bookmark:

• PWCC marketplace — for high-end graded cards. Their weekly auction prices are the strongest signal for $500+ slabs.
• Goldin Auctions — same tier, different audience.
• MyCardsell — pop-report-adjusted comps.

The reason this matters: a PSA 10 from 2019 with population 4,200 is a very different card economically than a PSA 10 from 2024 with population 380. The eBay sold filter doesn't show you pop. Graded-comp tools do.

Putting it together — a worked example

Say I'm pricing a 2023 Crown Zenith Galarian Gallery Pikachu V-Union GG70/GG70 (raw, NM).

Decision: list at $58 with Best Offer enabled, accept down to $52. That gives the buyer a small win on the negotiation, prices above the cheapest active (so I'm not first-to-cut), but stays within the cleared range from sold comps.

If I'd just looked at eBay sold and seen the $48 outlier, I might have priced at $50 — and left $6-$8 on the table per copy.

Where this approach breaks

Cards that don't have five sources to check:

• Brand-new sets. Days 1-30 after release, the sold data is too thin. Use TCGPlayer Market + my own gut.
• One-of-one parallels. No comps exist. Find the closest analog (same set, same parallel tier, different card) and adjust for player/character demand.
• Highly regional or language-variant cards. Japanese promos, Korean Lottery prints, foreign-language Black Star Promos. PriceCharting + eBay JP + my own dealer network is what I use.
• Severely off-condition cards. A 1999 Charizard with surface damage doesn't comp against NM Charizards. Find your own comp set within "damaged" sold listings.

The thirty-second version

When you can:

1. Check PriceCharting median. Anchor.
2. Verify with five condition-matched eBay sold comps in the last 30 days. Confirm the cluster.
3. Glance at active asks. Set your ceiling.
4. Cross-check TCGPlayer Market for raw Pokemon/Magic. Sanity.
5. For slabs, hit 130point. Truth.

Then price at the anchor unless something specific tells you to deviate.

This is the entire pricing methodology behind the vault's auto-suggested prices. The software does it in 1.7 seconds per card. If you're doing it by hand for a $200+ card, it's still 90 seconds. Either way, the framework is the same.

— Jamie

The vault's Market Price feature runs this exact comp logic on every import, weighted toward the most recent matched sold listings. If you're already a member, you've seen the suggested prices on your imports. If you're not — the trial is real and the first batch is free.