Premarket trading measured: 0.65% of the day's volume, and 90% of its minutes never trade

You can watch a stock rise 20% before the open. Whether you could have traded it is a different question, and it is the one nobody measures. So we measured it. Not whether premarket moves predict anything — that is a strategy question. Just whether there is a market there at all. The short answer: for the kind of stock a day trader actually watches, the median session did 0.65% of its dollar volume before the open, spread across 33 of the 330 premarket minutes. Nine minutes out of ten contained no trade at all.

1. What we measured, and on what

Picking "stocks that went on to move" would be reading the future, so the sample is defined the way a trader would have had it: symbols that appeared in a day trading scan on that same day, measured on that day only. From those we took every eighth symbol alphabetically, which is a sampling rule that cannot be nudged by hand, giving 84 symbols and 115 symbol-days across ten sessions from 2026-08-07 to 2026-08-20. For each symbol-day we pulled minute bars including extended hours, converted every timestamp to Eastern, and split them at the bell. Premarket is 04:00 to 09:30, the regular session is 09:30 to 16:00. Dollar volume per bar is volume times the bar's volume-weighted price. One check worth stating, because everything here depends on it: the regular session came back as exactly 390 minutes per symbol-day, 09:30 through 15:59. If the timezone handling were wrong, that number would not land on the nose.

2. Premarket is a rounding error in volume terms

56.5% of symbol-days did less than 1% of their dollar volume before the open. The spread between the quartiles is the interesting part. The 75th percentile is forty times the 25th. Premarket activity is not a property of the session, it is a property of the individual stock-day: either something happened overnight to that specific name, or essentially nothing traded. There is no typical premarket.

3. Nine minutes out of ten have no trade at all

The premarket session is 330 minutes long. On the median symbol-day, 33 of them contained a trade. That is 10%. For the other 90% of the premarket session there was no transaction in that stock at all — not a wide spread, not a thin book, simply nothing. On 3.5% of symbol-days there was no premarket trade at any point. This is the number that explains the experience people describe as "the price jumped and I could not get filled." There was no continuous market to get filled in. The quote you were looking at was the residue of the last trade, which may have been several minutes earlier, and the last price is not a price you can trade at.

4. What little there is arrives at the end

Premarket volume is not spread evenly across those 330 minutes. On the median symbol-day, 20.9% of all premarket dollar volume happened in the final 30 minutes before the open. Thirty minutes out of 330 is 9% of the session carrying 21% of the activity. The earlier you go, the thinner it gets, and the 04:00 to 07:00 stretch that looks like a full trading window on a chart is close to empty in transaction terms.

5. Per traded minute, regular hours are 10x denser

Comparing whole sessions is unfair to premarket, since it has fewer minutes. So compare only the minutes that actually traded: Even discounting all the dead time, a premarket minute carries about a tenth of the money a regular session minute does. The gap is not only that premarket is shorter or patchier. The minutes that do trade are themselves an order of magnitude lighter. Put that next to what the order book holds during regular hours — the quoted price is typically good for about $1,155 to $5,150 depending on price band — and a normal position size is a much larger share of premarket activity than of daytime activity.

6. What this means if you trade premarket

We are not saying do not do it. We are saying the constraint is different from the one people plan around. Size is the binding constraint, not direction. In a session where the median stock trades $31,030 per active minute, an ordinary retail position is a meaningful fraction of the flow. You are not a price taker in that market in the way you are at 11:00. Market orders are the wrong default here. Everything above says wide, discontinuous, and thin. The market order cost we measured during regular hours is already the highest at the open, and premarket is worse on every dimension we can see. The chart is misleading in a specific way. A minute with no trades still draws something on most platforms. Ten percent of the premarket session generated the entire price path you are looking at, and gaps between trades can be many minutes wide. Waiting 30 minutes changes the market you are in. A fifth of premarket volume arrives in the last half hour before the bell, and the regular session is another order of magnitude beyond that.

7. What we did not measure

Direction. We did not test whether premarket moves predict regular-session returns. That is a strategy question and this is a market-structure measurement. Spreads and depth. This measurement uses trade data, so it says how much traded and when, not what the quote looked like in between. Our depth work covers regular hours only. Mega caps. The sample is stocks a day trading scan surfaced, which skews to smaller and more volatile names. Large caps have materially more premarket activity, and these numbers should not be read as applying to them. Sample size. 115 symbol-days across ten sessions is enough to establish the shape and not enough to pin down tails. The quartile spread in section 2 says the distribution is wide, so treat the median as a centre of gravity rather than a forecast for any particular name.