Premarket gainers and losers: how the lists work

How a movers list is built

A premarket gainers or losers list is a ranking, not an analysis. Almost every version follows the same procedure:

1. Take a universe of symbols: every listed stock the provider covers, or whatever its scanner includes.
2. For each one, take the latest premarket trade price.
3. Compute the change from the previous regular-session close: (latest price − previous close) ÷ previous close × 100.
4. Sort by the largest gain for gainers and the largest loss for losers, apply any filters, and show the top of each list.

The ranking uses percentage change and nothing else, which is why it favours small, thinly traded stocks. For example:

Stock A tops the list, although Stock B's move involved roughly 200 times as much money. Neither number is wrong; they answer different questions. If what you care about is whether you can get in and out at a sensible price, the dollar volume row matters more than the change row.

Why two providers show different lists

Open three premarket movers pages at 07:30 and you will often see three different lists. The usual reasons:

• Venue coverage. A consolidated feed combines trades from all US exchanges; a single-exchange feed sees only that venue's share. For example, Alpaca's free IEX feed only shows premarket activity from 08:00 ET, so a stock that moved at 06:00 doesn't appear there until IEX itself trades it.
• Start time. Some feeds begin at 04:00, others later. Before a feed's start time, a list built on it reflects old data.
• Stale endpoints. Some movers endpoints keep returning the previous day's movers until the regular open. A "top gainers" list at 08:15 can therefore be yesterday's list, unless the tool computes premarket moves itself.
• Reference price. Most lists use the previous close, but a corporate action that isn't reflected in the reference creates a fake move. For example, after a 1-for-10 reverse split, a stock that closed at $0.50 and now trades at $5.00 shows +900% against an unadjusted close, although a holder's position is worth the same.
• Filters. One provider excludes stocks under $1, OTC securities, warrants or ETFs; another includes all of them.
• Refresh rate. One list updates every few seconds, another every minute, and in a fast premarket the order changes between refreshes.

Caveats before you trust a mover

Low float. A company with few freely tradable shares can move a long way on a small amount of buying, and fall just as fast. A big percentage on a low-float stock tells you more about supply than about demand.

Thin volume. A move made on a few thousand shares hasn't been tested. The next sizeable order, in either direction, may reprice it.

Stale prints. A list shows the last trade, not the current market. If that trade happened at 05:12 and it is now 08:45, the percentage is a historical fact. Check the time of the last trade and whether there is a live bid and ask.

Halts. A halted stock may sit on a list at its last price. When it reopens, often after news, the first trade can be far from that price.

Bad prints. Occasionally a trade is reported far outside the prevailing bid and ask, by error or as a late report. One such print can put a stock at the top of a list until the next real trade.

Reading a list critically

For example, suppose a list at 07:45 shows a stock up 38%. Before treating that as momentum, you check and find that:

• the last trade was 1,200 shares at 06:58, 47 minutes earlier;
• the quote is $3.10 bid and $3.45 ask, a spread of about 10.7% of the midpoint;
• premarket volume is 9,000 shares, under $30,000 in total.

The 38% rests on a handful of small trades and a quote you couldn't trade through without giving up around a tenth of the price in spread. The stock may still be interesting, but the list hasn't shown that.

A practical checklist:

1. Check the time of the last trade, not only the price.
2. Check premarket volume and dollar volume against the stock's normal activity; see relative volume in the premarket.
3. Check that there is a two-sided quote and how wide the spread is.
4. Check company size: market cap and float.
5. Find the catalyst, or confirm there isn't one.
6. Check for a halt.
7. If something looks odd, compare a second source.

A scanner can do much of this filtering for you; the premarket stock scanner guide covers which filters to use. For why the percentage itself can mislead, see what premarket momentum is.

How Momentum Lab discovers movers

PatternX's Momentum Lab starts from movers lists too, so it is worth being precise about how:

• For each data source, it requests the 50 top gainers and 50 top losers and refreshes them every 15 seconds.
• From each list, only the top 15 on each side are quoted, meaning fetched with a full bid and ask.
• At most 25 new candidates per strategy are added in any one cycle.
• With Interactive Brokers data, the list comes from IBKR's own scanner: top percentage gainers and losers, price above $1, with a market-cap floor applied to the scan.
• With Alpaca data, movers are computed from snapshot data rather than read from a movers endpoint. On Alpaca's SIP feed, which needs the Algo Trader Plus subscription, premarket data starts at 04:00; on the free IEX feed it starts at 08:00.

The list only nominates stocks. A nominee then has to pass a data-quality gate, the mandatory market-cap filter and the liquidity filters before a strategy can act, and a stock that fails is skipped with a labelled reason such as "Skipped — Market cap unavailable". The Momentum Lab guide lists every skip reason, and its market data section covers feed requirements.

What this approach does not do

Momentum Lab does not scan all US-listed stocks continuously. It works from the top of each movers list, so on a busy morning a stock that ranks outside the top 50 on its side isn't discovered until it climbs the ranking. Its view also depends on the data source: on the free IEX feed, the first four hours of the premarket simply aren't visible. The same is true, in different ways, of any list you read. Knowing what your source can't see is part of reading it, and the market data methodology explains how PatternX treats data sources.