Pull up a release page on Discogs to price something you're about to list, and the Statistics box hands you three numbers before you've thought about it at all: Low, Median, High. It reads like the site already did your pricing for you. It didn't — it did a narrower job than that, and selling at "the median" without knowing what went into it is how a clean VG+ copy gets priced off a run of beat-up G+ sales, or a genuine NM pressing gets left on the table twenty dollars under what it was actually worth.
What the Statistics box is actually built from
The Low/Median/High figures on a release page are calculated from that release's 30 most recent Marketplace transactions — not an all-time average, not a snapshot of current asking prices, but the last 30 sales that actually went through. Discogs tightened this calculation in an October 2023 update specifically to weight toward more recent sales, so what you're looking at is meant to track current market price, not whatever a copy sold for five years ago. You'll need to be logged into a free Discogs account to see the live numbers; logged out, the box doesn't fully populate.
That's a real, useful number. It's also doing less than it looks like it's doing, because of what it doesn't separate out.
The gap: the median doesn't know what condition sold
Here's the part that catches sellers who treat "Median" as a single price to list at: those 30 transactions cover whatever conditions actually sold, lumped into one figure. A release where the last 30 sales were a mix of NM, VG+, and VG copies produces a median that isn't the price of any one of those conditions — it's a blend across all of them, weighted by however the market happened to sell that particular month.
That matters most for anything you'd already grade carefully, because condition changes value on roughly the fixed percentages the Goldmine Standard uses — a VG+ copy is worth roughly half of Near Mint, VG roughly a quarter. If the median you're looking at happens to be built mostly from VG+ sales and you're holding a genuine Near Mint sealed copy, pricing at "the median" leaves real money on the table. Price a VG copy at that same median because it's "what records like this go for" and you've priced above what your actual condition supports, which is the direction that gets a listing sit unsold for months.
The fix isn't complicated, it's just a step most sellers skip: click through to the sales history behind the number instead of taking the headline figure at face value.
The number worth clicking: "Last Sold"
Under the Statistics box, the most recent sale date is a link, not just a label. Clicking it opens a detailed breakdown of those recent transactions — the price each one actually sold for, the condition it was graded at, and a chart plotting price over time. That's the number that actually answers "what does a copy in my condition sell for," instead of "what did records at this release sell for regardless of condition." A release with a $40 blended median might show $60 NM sales and $22 VG sales sitting inside that same number — two very different list prices for two copies of the same record, both technically "at the median" if you never open the breakdown.
It's an extra click every reseller pricing a record should be making before they list, and most aren't, because the headline Low/Median/High box is right there and looks complete on its own.
Want/have: a demand signal, not a sales signal
The other number on the same page — how many Discogs users have this release in their collection versus how many have it on a want list — gets read as a straightforward scarcity score, and it's a real signal, just a noisier one than sellers treat it as. A release with 4 copies logged as "have" and 400 "want" is telling you something genuinely different from one with 400 "have" and 4,000 "want," even though both work out to a 10:1 ratio — the absolute numbers matter as much as the ratio itself, because a handful of collectors owning a genuinely scarce pressing is a different market than a widely-owned release with a large, loose fanbase.
The bigger caveat: wanting a record and buying one aren't the same action. A want list is free to add to and nobody prunes it, so a high want count tells you a lot of people would like to own a copy someday and very little about how many of them are actively shopping this week. Treat want/have as a rough scarcity indicator worth factoring into a price, not as a demand forecast you can price against with any precision — the Last Sold history is still the harder evidence of what buyers actually paid.
A worked example
Say you're pricing a genuine Near Mint copy you're confident in, and the release page shows: Low $18, Median $32, High $58, want/have of 210:9.
- Listing flat at the $32 median would underprice a real NM copy if that
- median is pulled down by VG and VG+ sales mixed into the same 30-sale
- window — which is the default assumption until you check.
- Click "Last Sold." Say the breakdown shows the $58 high was an NM sale
- two weeks ago, the $32s cluster around VG+, and the $18s are VG or
- worse. Your NM copy now has a real comp — something close to $55–60,
- not $32.
- The 210:9 want/have ratio doesn't set the price, but it does tell you
- demand is real enough that pricing toward the top of that range,
- rather than undercutting to move it fast, is a reasonable call for a
- genuinely well-graded copy.
What to actually do
- Don't list at "the median" without opening Last Sold. The headline
- number blends every condition that sold — it isn't the price for your
- specific grade.
- Grade honestly first, then find the comp that matches that grade,
- not the other way around. Pricing off a number before you've confirmed
- what condition it represents is backwards.
- Read want/have as a rough scarcity signal, not a demand forecast.
- A high want count with almost nobody actually owning a copy is a
- stronger signal than the same ratio at higher absolute numbers, and
- neither one tells you how many of those people are buying this month.
- Recheck before relisting a stale price. The 30-sale window Discogs
- calculates from moves as new sales happen — a comp you pulled six months
- ago isn't necessarily the comp today.
PalmFlow doesn't set your Discogs price for you — it can't, since Discogs is import-only on PalmFlow's side, listings and orders come in but nothing pushes back out to a Discogs listing. What it does do is import that order once it sells, so the actual sale price against what you paid for the record shows up as real margin instead of a guess sitting in a notebook. Free plan, 50 items, no card required.
Disclaimer: this is general information about how Discogs' own pricing tools work, not a substitute for checking a release page directly. Discogs' calculation methodology and the data it's built from are set by Discogs and can change — confirm what a release's Statistics box currently reflects on discogs.com before pricing a specific item.