How to find a One Piece card's real price — and why the price apps are often wrong

The short answer: Most price apps — Collectr, Shiny, and the rest — build their number from asking prices and thin, easily-distorted data, so it's often well above what a card actually sells for. TCGIntel prices to real sold comps (where many independent buyers actually transact), backs grading calls with EV math you can check, and discloses every position. Here's how we do it, the three traps that inflate the apps' numbers, and how to find a card's real price yourself.

This is independent analysis of physical collectibles — not financial advice.

Why the price apps are often wrong

A price app's job is to show a number, fast. To do that, they lean on data that's easy to scrape but easy to distort:

• Asking prices, not sold prices. A card listed at $1,200 isn't worth $1,200 until someone pays it. Blend live listings into the "value" and you inflate it above the real clearing price.
• Thin markets. Most cards trade rarely. When only a handful sell, one or two high sales drag the "market price" up without broad demand behind it.
• Comp-fed loops. Some retailer buyback / trade-in programs price off public comps — so a few inflated sales can lift a "value" that then feeds back into the apps. We documented exactly this on OP13-108 Bonney, and turned it into a reusable checklist: how to spot a manufactured card pump.
• Stale data. A pop count or price from months ago, presented as current.

None of this is malicious — it's what happens when you optimize for a number instead of the right number.

How TCGIntel prices a card

1. Anchor to real sold comps. Where copies consistently sell to many different buyers — not the highest listing, not a single outlier.
2. Discount thin or inflated comps. A four-figure print built on two sales is a headline, not a value. The thinner the market, the harder we discount.
3. Run the grading-EV math. Every grading call is backed by an expected-value calculation you can run yourself — raw price, PSA-10 rate, grading cost, graded price.
4. Disclose our position. If the founder holds the card, it's labeled on the call with the entry price and date — so you can weigh our incentive before our logic.

The three traps to check before you pay

• Ask vs. sold. Look at the last few sold prices, not the active listings. A wall of high asks with few real sales is a red flag.
• The outlier. Don't let one inflated sale become "the price." Anchor to the cluster, not the peak.
• The thin-market spike. A big move on low volume can reverse the moment real liquidity returns. Wait for broad, repeated sales before you trust a new level.

Find a card's real price yourself

You don't need us for this — and we'd never paywall the method:

1. Pull the last 5–10 sold prices (eBay sold, Cardmarket sold), not the listings.
2. Throw out the highest and lowest; anchor to where most copies actually cleared.
3. For a graded play, run the grading-EV calculator on conservative graded prices.
4. If the market is thin, discount hard and wait for confirmation.

That's the whole method. The apps give you a number; this gives you the real one.

See it in practice

• Every call we publish uses this method, dated and public — the track record.
• A worked example of an inflated "value": why PSA 10 Bonney's price is a manufactured pump.
• Run the grading-EV calculator on any card.

Independent analysis of physical collectibles. Not financial advice. Card values are approximate secondary-market estimates, not live quotes.