What's the difference between a listing and a comp?

A listing is an asking price; a comp is a completed sale where money actually changed hands. Only sold data tells you what buyers really pay. Anchoring to asking prices, which often sit above the market, is the most common way dealers overprice and get stuck with cards that sit.

How do I handle one sale that's way above the rest?

Treat it as noise and ignore it. A single high sale is usually an outlier: a bidding war, a mislabel, or a nicer copy than yours. Read the middle of the cluster of recent comparable sales instead, which reflects what a typical buyer paid for a typical copy.

How recent do comps need to be?

As recent as you can get, because a card's market moves over weeks and months. Weight the freshest sales most and treat older ones as background rather than anchors. And always match the exact printing and roughly the condition of the copy you're pricing.

How to Read Comps Like a Pro

A comp is just a recent sale of the same thing. Simple in theory, and where most pricing mistakes are born in practice. The dealers who price with confidence aren't looking at different data; they're reading the same data more carefully.

Everyone pricing a card looks at recent sales. The difference between a dealer who prices well and one who guesses is what they do with them: which sales they trust, which they throw out, and how they adjust for the copy actually in hand. Read comps loosely and you either scare buyers off with a high number or hand margin away with a low one.

This is an evergreen method, not a price list. It's about how to interpret comparable sales for any card, so you arrive at a number you can defend in a few seconds and stand behind when a buyer pushes back. No specific figures here, because the moment they're printed they're stale; the skill is in the reading.

Only sold data is a comp

The first rule fixes the most common error: a listing is not a comp. Anyone can ask any price for a card, and unsold listings often sit far above what the market actually pays. If you anchor to asking prices, you'll price high, and your cards will sit.

A real comp is a completed sale: money changed hands. Those are the only data points that tell you what a buyer was actually willing to pay. Build the habit of ignoring the "for sale" column entirely and reading only the "sold" one. It's the single biggest upgrade to your pricing, and it costs nothing but attention.

Recent, and for the exact printing

Two filters do most of the work. First, recent. A card's market can move over weeks or months, so a sale from last year describes a different market than the one you're selling into today. Weight the freshest sales most and treat old ones as background, not anchors.

Second, the exact printing. The same art can exist across several sets, editions, and rarities that price very differently, so a comp for the wrong version is worse than no comp at all. Confirm you're comparing your precise printing before you trust a number. Getting the printing right is a pricing skill of its own; when in doubt, identify the version in hand first, then pull comps for that and only that.

Throw out the outliers, find the cluster

Line up recent sold comps for the right printing and you'll usually see a cluster with a few strays above and below it. The strays are noise: a bidding war between two collectors, a mislabeled listing, a fire-sale from someone who needed cash, a nicer or worse copy than average. One unusual sale doesn't set the market.

Ignore the extremes and read the middle of the cluster. That honest midpoint is your starting number, the price a typical buyer paid for a typical copy. Pricing to the middle, not the peak, is what keeps you fast and fair: you sell at a number you can defend, instead of chasing the one lucky sale you'll rarely repeat.

Adjust for the copy in your hand

The cluster gives you a price for an average copy. Your card is a specific copy, so adjust from there for condition. A clean near-mint example sits at the top of the range; a played one steps down, sometimes more than one tier. Read your own card honestly against the comps, because a buyer will. For how condition moves value, see how card condition affects value.

Make the same adjustment the comps themselves need: a sold price for a beat copy isn't a comp for your mint one, and vice versa. Match condition to condition as closely as you can, then land on your number. Done consistently, this is a few seconds of work, not a research project.

From comps to a number, fast

Put it together and reading comps becomes a quick, repeatable move: pull recent sold sales for the exact printing, ignore the outliers, read the middle of the cluster, adjust for your copy's condition, and price. A number reached that way is defensible, and defensible is what lets you hold it when a buyer tries to talk you down. For the at-the-table method around it, see how to price cards to market at a show.

The catch at a show is doing all this in seconds, in a hall with no signal, with a line waiting. That's why it helps to have market values already attached to your inventory by set and condition, prepared before the doors open rather than researched on the spot. CardOps carries a current value on each card and keeps it reachable in Show Mode even offline, so the reading you learned here is ready the instant a buyer asks.