Buy order or sell order

Two numbers, not two guesses

Every market in this game keeps two prices for every item. The cheapest sell order standing, which is what you pay if you buy it right now. And the highest buy order standing, which is what you receive if you sell it right now. They are the two edges of the order book, and the space between them is a gap nobody is trading in.

Read them the wrong way round and you have not made a small error. You have priced your exit at the entry, and the difference does not show up until the bag is already in Caerleon. This guide is short and it is arithmetic, so you can check every line of it yourself.

The Black Market has no sell price

It only ever generates buy orders. Nothing there is for sale. That is the mechanic, and it is the entire reason the place exists: the Black Market buys gear from players so that the game can drop it back to somebody else.

I pulled Black Market cells straight from the public API, Normal quality, spread across the whole T4 to T8 gear range. Two runs, 1,435 cells and 600, because the API rate limits and stops you where it wants. Both say the same thing. Essentially every cell carried a buy order. Fewer than one in five also showed a sell price: 14% on the large run, 18% on the small one.

That number, where it exists, is not the Black Market offering to sell you something. It is a player who posted a sell order there, into a market that never buys, and it has been sitting untouched ever since. So on more than four cells out of five there is no sell price at all, and on the rest it is a price nobody has ever paid. Where both exist, the phantom sell price sits a median 1.4x above the buy order on both runs, above 5x at the ninth decile, and the worst single cell I found was 54x.

Cities have both sides, so you can measure the gap

On the current snapshot, across the 7,637 gear cells that carry both:

By city the median gap runs from 1.47x in Thetford to 1.78x in Bridgewatch, so the royals are all much of a muchness. Then there is Caerleon, at 2.74x, with a ninth decile close to 16x. The worst place in the game to confuse those two fields is the one city where every flip ends.

What the mix-up invents

Take those same cells and price your exit at the sell order instead of the buy order, net of the 4% tax. The gap alone, with no other mistake anywhere, invents a median 74% of margin. At the third quartile it invents 147%. At the ninth decile, 307%.

I want to be careful here, because it would be easy to be unfair. I am not saying any particular tool is wrong. I am saying that a flipping tool quoting margins that sit around 79% on high value gear is not showing you a lucky market: that is roughly the number this specific mistake produces. And it takes thirty seconds to find out which one you are looking at.

A test that takes half a minute

It takes half a minute. Take any line a tool is showing you, open that item in game at the Black Market, and read the buy order it is offering you. Compare it to the exit price the tool used. If the tool's exit sits above that buy order, it is reading the wrong field, and the difference is not going to be paid to you.

Two things make this test better than it looks. It needs no data of your own and it does not care how the tool is built, so it survives every "but mine uses a live feed" answer. And it works in reverse: a tool quoting an exit below the visible buy order is being conservative, which is a much better failure to have.

The fees are a separate mistake

While we are on things that get miscounted, here is the fee table. It is the same at every marketplace, including the Black Market.

A flip fills an existing buy order, so it pays the tax and nothing else. If a tool charges you a setup fee on that exit, it is modelling a sell order you never placed. That is a small error and it is in your favour, so it is not the one to worry about.

It does mean something worth keeping in mind, though: a tool can have its fee arithmetic exactly right and still be out by a factor of five, because these two mistakes have nothing to do with each other. Correct fees on the wrong price is still the wrong price.

The half that costs you a plan

I said cities have both sides. Most of the time they do not. Across the whole snapshot, only 17% of city cells carry a buy order at all.

So for more than four cells out of five there is nothing to instant-sell into. If your plan for a trade that goes wrong is "worst case I dump it back on the city market", that plan does not exist for most items. You either post a sell order and wait up to 30 days, or you carry it. Worth knowing before you need it rather than after.

Where I could be wrong

• The Black Market samples are 1,435 and 600 cells rather than the full universe, because the public API rate limits hard and cuts the run where it likes. That is also why the two runs disagree by four points on the headline share: a sample stopped by a rate limit is not a random sample. Both are spread across the gear range rather than cherry-picked.
• A cell with no buy order may just mean nobody has browsed that market recently. The public database only sees what players look at.
• Public data carries no order depth, so a buy order can be one single unit.
• Europe server.

How this was measured

The city figures come from the public snapshot published at 22:08 UTC on 30 July 2026, which lists both the cheapest sell order and the highest buy order per item, quality and city. I kept the cells that carry both, restricted to T4 to T8 gear for the headline numbers, and compared the two fields directly. The invented margin is the sell order taxed at 4%, measured against the buy order, which is what you would actually have received.

The Black Market figures come from a separate call to the public API, because the snapshot deliberately does not store a Black Market sell price: it is a phantom, and storing it invites exactly the mistake this guide is about.

Back to all guides