Economy Guides / Item Prices
Item Prices Guide
How a single number gets computed from thousands of listings, what gets discarded on the way, and the reason that number tends to sit slightly above what things actually trade for.
Where the numbers come from
When a player marks a stash tab for sale, its contents and asking prices become public data. That feed is what every economy tracker reads — not sales, not trade logs, not anything a buyer confirmed. Just what people have currently written on the tin.
Which gives you the single most important fact on this page, and the one that reframes everything else: displayed values are asking prices, not sale prices. Nothing in the data records what anything actually sold for, because that information is never published.
That is why the filtering matters so much. An unfiltered average of asking prices would be useless, dragged upward by optimists and downward by people who quit the league in week two. The processing is what turns a pile of hopes into something usable.
How thousands of listings become one number
The exact implementation is not published anywhere, but the shape of it is visible in the output. Listings get collected, obviously bad data gets stripped, and what survives is reduced to a single representative value per item, per variant, per league.
Every copy in a public sale tab, at whatever price its owner set. Includes items listed months ago by players who have not logged in since.
Both directions. The absurdly high and the improbably low, which are usually mistakes, placeholders or someone testing the market.
Patterns consistent with someone listing many copies far below market to drag the displayed figure down, then buying up the real listings.
Nothing about that process is dishonest. It is the right thing to do with the data available. But it does mean the figure you read is a summary of intentions, and summaries inherit the biases of the pile they were computed from.
The listings you can see are the ones that did not sell
Here is the subtle problem, and it is not a flaw anyone introduced. It falls out of the fact that the data is a snapshot of what is currently listed.
Think about what happens to a well-priced item. Someone lists it slightly under market, a buyer whispers within minutes, and it leaves the pool. Now think about an overpriced one. Nobody whispers, so it sits there being counted, hour after hour.
The pool at 9amEvery copy currently listed
The pool at 5pmSame item, eight hours later
The pool constantly loses its cheapest members and keeps its dearest. Filtering removes the extremes, but it cannot recreate listings that no longer exist. So for anything in genuine demand, the displayed figure tends to sit a little above what the item actually changes hands for.
This is inference from how the data works rather than a documented behaviour, and it is worth holding loosely. But it explains something every trader notices eventually: the tracker price feels slightly optimistic when you are buying and slightly disappointing when you are selling.
The listing count is the confidence number
This is the most useful figure on any economy page and the most widely ignored. Two prices can look identical and rest on completely different amounts of evidence.
Read the count before you read the value. Most pricing mistakes disappear the moment that becomes reflex. The price confidence checker turns it into a one-field lookup if you want the habit reinforced.
Same name, different item
Seeing one item name listed several times at wildly different values is not a bug. Those are separate things that happen to share a label, and blending them would produce an average describing none of them.
Figures above are relative multipliers, not prices, and the spread varies enormously by item. The point is the shape: a factor of a hundred can sit between two rows sharing a name. Always read the small text identifying which version a row is pricing before you list or buy.
Why rare items are not on there at all
This confuses newer players constantly, and the reason is more interesting than it looks. Unique items have fixed modifiers, which makes two copies genuinely comparable. Rare items do not. A rare is a random combination drawn from a large pool of possible modifiers, each with its own value range.
The practical consequence is that no meaningful summary price exists for rares. There is no "the price of a rare ring" any more than there is a price of a random sentence. Two rare rings can differ in value by four orders of magnitude while looking superficially similar.
Which is why pricing a rare means going to the official trade site and searching for items with the specific modifiers yours has, then seeing what comparable ones are listed at. It is slower, and there is no shortcut, and any tool claiming to give you a rare's value from its name alone is not doing what it says.
What actually moves an item's price
When to stop reading and go to trade
A tracker is for orientation. The official trade site is for decisions. Four situations where the summary has done all it can and you should go look at real listings.
None of this makes the tracker less useful. It makes it useful for the thing it is good at — telling you roughly where a market sits in about two seconds — while keeping the final decision somewhere the actual listings live. If you are new to any of this, the beginner guide covers the workflow end to end, and common mistakes covers where it usually breaks down.
