WhitmanTrading

What Is Gap Trading?

A gap is a range of prices that never traded, created when a market opens away from its previous close. Gaps sometimes fill — price trades back through the range — and sometimes do not, and nothing visible at the moment one opens says which it will be.

What Is Gap Trading? — illustrated on a chart Watch me trade an open on a live chart (6:09)

A gap is the simplest thing on this site to define and carries the most confidently repeated claim in trading, which turns out not to be checkable.

How it works

A candlestick chart where one bar opens well above the previous close, leaving a visible hole.
A gap is a price range nothing traded in — 0.60 wide here. Illustrative chart - not real market data.

A market closes at one price and opens at another. The prices in between never traded, and the hole they leave is the gap. On this chart, 0.60 wide.

Gaps need a break in trading to exist. Stocks close overnight and at weekends, so daily stock charts are full of them. A continuously traded market has almost none, which is why this is largely a stocks topic.

Three gaps, three outcomes

A gap down that price trades back through within the same bar.
This one filled: price traded back through the whole gap.

Filled. Price traded all the way back through the range, in this case within the same bar.

A gap up followed by price continuing away from it.
This one did not — price carried on away from it.

Gap and go. Price never came back, and every bar after it moved further away.

A gap up that remains unfilled to the right edge of the chart.
And this one is still open at the right edge.

Still open. The third gap has not filled, and the chart ends.

Three gaps on one chart, three different outcomes, all of them 0.50 to 0.60 wide. Nothing about the gaps themselves separated them.

“Gaps always fill”

This is the most repeated claim in the subject and it cannot be tested as stated.

The problem is that no time limit is attached. A gap that has not filled in six years has not disproved the claim — it might fill tomorrow. A statement that no observation can contradict is not a finding, whatever its track record sounds like.

Make it testable and it becomes ordinary. “Most gaps under 1% fill within five sessions” is a claim you could check, and it would have a hit rate rather than a certainty.

The third gap on this chart is unfilled at the right edge. Under the unfalsifiable version, that is not a counterexample. Under any testable version, it is.

What is actually usable

The gap with both its edges drawn as horizontal levels.
The edges of a gap are ordinary levels, and that is the usable part.

Two prices, and both of them are real.

The previous close is a price the market agreed on, and a great many people are watching it — the crowding argument that makes any level work.

The opening price is where today’s trading started, which is where a lot of orders were filled.

Neither of those requires believing anything about fills. They are levels, they behave like levels, and the hole between them is where price will move quickly if it goes — the same thin-area reasoning as the volume profile page, arriving from the opposite direction.

Size is the one measurement

The chart with all three gaps annotated with their widths.
The three gaps measure 0.50, 0.60 and 0.50.

Measure the gap against normal daily movement, not in isolation. A 0.50 gap on an instrument whose ATR is 0.20 is an event; the same gap where average true range (ATR) is 2.00 is an ordinary open.

That ratio is the only number in the whole topic that carries information, and it is rarely the one quoted.

The named kinds, and what the naming does

There is a standard four-way classification, and it is worth knowing mainly for what it reveals about itself.

A common gap is a small one in a quiet market with no news behind it.

A breakaway gap starts a move out of a range.

A runaway gap happens partway through a trend that continues.

An exhaustion gap happens near the end of a trend that then reverses.

Now read those last three again. They are distinguished entirely by what price did afterwards — the same gap is a runaway gap if the trend continued and an exhaustion gap if it did not.

So the classification cannot be applied at the time. At the moment a gap opens, every one of them is simply a gap, and the label arrives with the outcome it is supposed to have predicted.

That is not a reason to discard the vocabulary, which is a useful way to describe a chart after the fact. It is a reason not to mistake it for a method — the same distinction the chart patterns page draws between a name and a level.

A worked example

Mark both edges before the open. Previous close, and — once it exists — the opening price.

Do nothing for the first bars. The open is the noisiest moment of the session and the gap is not going anywhere.

Then treat the edges as levels. Price rejecting the previous close from above is a level holding; price closing back through it is a level breaking, with the breakout page’s caveats.

And the invalidation is the far edge, which is a real price the gap handed you.

The original data

Across our study of 24,971 trading videos, 96 cover gap trading. The median one gets 8,535 views, 82% never pass 50,000, and the median length is 12.8 minutes.

The corpus carries description text for only 6 of those 96 — far too few to say anything about how the topic is written, so this page does not try.

96 videos is a modest field for a topic that applies to every stock, every day, and produces the single most repeated maxim in trading.

When it fails

The fill never comes

The unfilled gap with the distance from current price to the fill target marked.
Shorting this one to 'fill the gap' would still be waiting.

A position taken to capture a fill has no exit if the fill does not happen, because the thesis has no expiry date. That is the practical cost of an unfalsifiable rule: it cannot tell you when you were wrong.

It is the wrong instrument

Continuously traded markets barely gap. Running a gap strategy on 24-hour futures or crypto means waiting for events that structurally almost never occur.

The gap was the news

A gap is usually caused by something — an earnings release, a policy decision. The chart shows the hole and not the reason, and the reason is often exactly what decides whether price comes back.

You measured the fills you remembered

The chart cut off immediately after a gap opens.
A gap has just opened. Does it fill, or run?

Filled gaps are invisible afterwards — the chart just looks continuous. Unfilled ones stay visible forever. That asymmetry means memory over-counts one and under-counts the other, in the direction that makes the maxim feel true.

Support and resistance is what the two edges of a gap are, and why they work without any theory about fills.

Breakout is what a gap-and-go actually is.

And day trading is where gaps matter most, because the open is where a session inherits everything that happened while it was shut.

What I actually do

The thing that made this usable for me was giving up on predicting the fill and treating the two edges as levels like any other. The previous close is a price a lot of people are watching and the opening price is where today started - both of those are worth marking. Whether the hole between them closes is a question I stopped trying to answer, because I could never find a version of it that was falsifiable.

— Michael Whitman, from this video

This page is educational, not financial advice. Test every idea on your own charts before risking money.