Dark pool

A dark pool is a private trading venue, formally an Alternative Trading System (ATS), where buy and sell orders are matched away from the public, displayed order books of exchanges like the NYSE or Nasdaq. The defining feature is pre-trade opacity: unlike a lit exchange, a dark pool doesn't publish the resting orders sitting in it before they trade, so other market participants can't see the size or price of interest waiting to transact. Large traders, most often institutional managers, use dark pools mainly to work a big order, buying or selling a large block of a stock, without signaling that size to the wider market ahead of time, which can move the price against them before the order is even filled.

Because dark pools don't publish pre-trade quotes, there's no live order book for a dark pool the way there is for a lit exchange. What becomes visible is after the fact: an individual dark pool trade still gets reported to a FINRA Trade Reporting Facility and shows up on the consolidated tape shortly after it executes, so the price, size, and time become public within seconds, same as any off-exchange trade. Separately, FINRA publishes weekly ATS transparency data: aggregated share volume for each individual dark pool, broken out by stock, published roughly two weeks after the week it covers. That delayed, venue-level report is the closest thing to a real 'dark pool volume' figure, and it's what most estimates of a stock's dark pool activity are built from.

A dark pool print, whether seen on the tape or in FINRA's delayed ATS report, doesn't tell you which side of the trade was the aggressor: whether it was driven by an urgent buyer or an urgent seller. That's a real gap, since a lit-exchange trade can usually be classified as buyer- or seller-initiated by comparing its execution price against the prevailing quote at the time, a reference point a dark trade never had to begin with. A dark print also doesn't identify either counterparty, doesn't say whether the trade opened or closed a position, and the venue-level ATS report pools every stock and every participant in that venue together, so it can't attribute a specific print to a specific fund. That combination, no direction and no identity, is exactly why dark pool prints are read as a weaker signal than a block trade of comparable size on a lit exchange, even though both can represent large, genuinely informed positioning.

Suppose a mid-cap stock trades 40 million shares of total consolidated volume over a given week. FINRA's ATS transparency report for that week, published about two weeks later, shows the stock's two largest dark pools together crossed 3.2 million shares, roughly 8% of the week's total volume. That figure is genuinely informative: it says a meaningful chunk of the week's trading happened away from the lit exchanges' visible books. It says nothing about whether that 3.2 million shares was mostly buying or mostly selling, whether it came from a handful of large institutional orders or many smaller ones, or which participants were on either side. A trader working from this data alone knows that off-exchange interest existed and roughly how large it was, and stops there.