Hidden Order: Size That Does Not Show
A hidden order is a resting order whose full size is not displayed in the book, most commonly an iceberg that shows a small slice and refills it as it trades. It is a standard exchange feature that exists because displaying a large order moves the price against the person placing it.
How it works
A hidden order rests at a price without displaying its full size. Some are entirely invisible; the common variety, an iceberg, shows a small visible slice.
When the visible slice trades, another appears. A 20,000-lot order displaying 200 at a time looks like 200 in the book, and keeps looking like 200 through 100 refills.
The reason is straightforward. A visible 20,000-lot bid tells everyone that a large buyer is present, and the price moves up before the order fills. Hiding the size is how a large participant transacts without paying for the information they would otherwise be broadcasting.
It is an order type the exchange offers, documented in the rulebook, usually with a fee difference. It is not manipulation and it is not hiding from regulators — the exchange knows the full size, and so does the clearing system.
What it means for reading a book
You find out about hidden size after the fact. The tell is traded volume exceeding what was displayed: a level showing 200 that absorbs 5,000 had something behind it.
Which makes traded volume the only evidence available, and it arrives one trade too late to help with the trade in front of you.
The practical consequence is simple and worth stating plainly: displayed size is a floor, not a count. Any method built on reading the depth display — Level 2 or a ladder — is reading a lower bound of unknown tightness.
Dark pools apply the same principle to an entire venue. Nothing is displayed at all; trades are reported after execution. A substantial share of equity volume trades that way, which is part of why an equity order book is a partial picture even before hidden orders are considered.
In practice
On any longer timeframe hidden size is invisible and constantly present. It is part of why price stalls at prices with no visible reason, and it is not something a daily-chart trader needs to model — it is already inside every bar.
Outside session hours there is nothing to hide, because there is almost nothing resting.
Your own stop order is hidden by construction, held at the broker and invisible to the market. The concern about stops being “seen” is misplaced; what is seen is the obvious price where a great many people would place one.
Hiding usually costs something. Many venues pay a smaller rebate, or charge more, for non-displayed liquidity — the exchange is compensating for the fact that the order is not contributing to the public quote. On top of that, the round trip is 2% of a median bar’s range on this history.
On a quiet bar a single hidden participant can be the entire session. With tenth-percentile bars at 0.17 against a median of 0.493 on this data, it does not take much resting size to hold a thin market still.
What a hidden order is not
It is not illegal. Spoofing — displaying size with no intention of trading — is illegal. Hiding size you do intend to trade is an exchange feature.
It is not undetectable. It shows up in traded volume against displayed size, after the fact.
It is not only used by institutions. Most retail platforms offer iceberg orders, though few retail orders are large enough to need one.
And it is not the same as a dark pool. One is an order type on a lit venue; the other is a venue that displays nothing.
When it fails
In a range you cannot distinguish a refilling iceberg from ordinary two-sided activity. Size appears at a price repeatedly because price keeps returning there, and that looks identical to a hidden order being worked.
The second failure is inferring intent. A large hidden buyer might be accumulating, or hedging, or unwinding a position from somewhere else entirely. The size is observable; the reason never is.
A third is trading against a wall you inferred. Assuming hidden supply at a level and positioning against it is acting on the least verifiable thing on the chart.
A fourth is expecting it to persist. An iceberg finishes when its total is filled, and there is no warning — price moves freely the moment it does.
And a fifth is over-explaining ordinary behaviour. Price stalling at a number is usually just a level a lot of people can see. Hidden size is one explanation among several, and it is the one you cannot check.
The original data
On this site’s shared 576-bar history, bar ranges span 0.17 at the tenth percentile to 1.10 at the
ninetieth against a median of 0.493, and the round-trip cost of 0.0098 price units is 2% of that median
bar and 45% of the smallest bar in the series. The figures are in research/series-measurements.json,
produced by site/measure_series.py.
The measurement that actually detects hidden size is one you can make and almost nobody does: for a given price level, compare displayed size against the volume that traded there. A level that showed 200 and absorbed 5,000 had something behind it, and that comparison is available in any platform that records both time and sales and depth. It converts a suspicion into a count — and once you have counted it a few times on your own instrument, you will have a much better sense of how much of the depth display is worth reading at all.
Related
Order book covers what a book shows and what it leaves out. Dark pools is the same principle at venue scale. And Level 2 is the display whose numbers this page says are a floor rather than a count.
Hidden size explained something that had annoyed me for years: price stalling at a number where the book showed almost nothing. Once I understood that the display is a floor and not a count, a whole category of confusing chart behaviour stopped being mysterious.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.