WhitmanTrading

Gap and Go: A Screening Problem First

Gap and go is a strategy of buying a stock that has gapped up at the open and continues in that direction. Finding candidates is a screening problem rather than a charting one, and the names that gap most are the thinly traded ones where execution costs are highest.

How it works

A candlestick series containing several opening gaps, with the largest opening gap marked. The headline on the chart reads: Buying a gap that keeps going.
Buying a gap that keeps going. Illustrative chart - not real market data.

A stock opens well above its previous close and keeps rising. The trade is to buy into that continuation rather than fade it.

A candlestick chart of the site's shared price history. The headline on the chart reads: It starts as a screening problem, not a chart one.
It starts as a screening problem, not a chart one. Illustrative chart - not real market data.

The first task is finding candidates, and it is not a charting task. Out of thousands of listed names, a handful gap meaningfully on any given morning. A scanner filtering on premarket percentage change, premarket volume and float produces the list, and the chart work starts afterwards.

That ordering matters because most material on this strategy skips it. The pattern on the chart is the easy part; assembling the list of candidates before the open, every day, is the actual work.

What separates the gaps that continue

A strongly rising stretch of the long price series. The headline on the chart reads: The gaps that continue usually have a reason behind them.
The gaps that continue usually have a reason behind them. Illustrative chart - not real market data.

A gap on real news is a repricing. Earnings, a contract, a trial result — the market’s estimate of the company changed, and there is no reason for price to return to a level that reflected the old estimate.

A flat, quiet stretch of the long price series. The headline on the chart reads: A gap on nothing is the one that comes straight back.
A gap on nothing is the one that comes straight back. Illustrative chart - not real market data.

A gap on nothing is thin overnight trading. A handful of participants moved price in a session with almost no volume, and the wider market disagrees when it arrives.

A candlestick chart with a volume histogram beneath it, with the volume histogram emphasised. The headline on the chart reads: Participation at the open is the one real filter.
Participation at the open is the one real filter. Illustrative chart - not real market data.

Volume at the open is the best available separator, and premarket volume relative to the name’s average is the version you can check before the bell. It is the one filter with a mechanism behind it rather than a shape.

The problem the charts do not show

A flat but volatile stretch of the long price series. The headline on the chart reads: It lives on small, thinly traded names, which is the risk.
It lives on small, thinly traded names, which is the risk. Illustrative chart - not real market data.

The names that gap most are the smallest and thinnest. A large, liquid company rarely gaps 20%; a small-cap with a low float does it regularly, because it takes very little buying to move it.

A declining stretch of the long price series. The headline on the chart reads: And the spread on those names is the whole problem.
And the spread on those names is the whole problem. Illustrative chart - not real market data.

And on those names the spread can be a substantial fraction of the price. The round trip on this site’s shared history is 2% of a median bar’s range; on a thin small-cap at the open, the spread alone can exceed the move being targeted.

That is the honest reason this strategy is harder than it looks. The selection criteria that make a name a good candidate — small, thin, moving fast — are exactly the criteria that make it expensive to trade.

A 72-bar candlestick section of the shared price history. The headline on the chart reads: There is no book to speak of in the first minute.
There is no book to speak of in the first minute. Illustrative chart - not real market data.

The order book in the first minute is nearly empty at any individual price. Heavy volume and thin depth is the combination that produces large slippage, and a market order into it fills across several levels.

In practice

A gently rising stretch of the long price series. The headline on the chart reads: Seventy percent of ten-bar highs came back within ten bars here.
Seventy percent of ten-bar highs came back within ten bars here. Illustrative chart - not real market data.

On this site’s shared 576-bar history, of 53 closes above a 10-bar high, 70% closed back below the level within ten bars, and 43% had a higher close ten bars later against a 54% base rate for any bar. That is the null model for buying a recent extreme — what it looks like with no mechanism at all — and a continuation method has to beat it.

A calmly advancing stretch of the long price series. The headline on the chart reads: And a volatility halt can stop you trading entirely.
And a volatility halt can stop you trading entirely. Illustrative chart - not real market data.

Circuit breakers matter here more than anywhere else. A single-stock volatility halt suspends trading for several minutes, mid-position, in either direction. You cannot exit during a halt, and price on the reopen can be a long way from where it stopped.

A declining stretch of the long price series, with the entry price and the level at which a stop would trigger drawn as horizontal lines. The headline on the chart reads: The stop is under a bar that is already enormous.
The stop is under a bar that is already enormous. Illustrative chart - not real market data.

The stop sits under a bar that is already very large. A stock up 15% premarket has a first bar that may be several percent tall, so the invalidation level is far away and the position has to be small — which is the correct response and the one that makes the strategy feel not worth the effort.

A long-horizon candlestick view of the same price series. The headline on the chart reads: It does not exist on any timeframe longer than minutes.
It does not exist on any timeframe longer than minutes. Illustrative chart - not real market data.
A candlestick chart of the site's shared price history, annotated with the round-trip cost. The headline on the chart reads: And the round trip at the open is worse than a share of a bar.
And the round trip at the open is worse than a share of a bar. Illustrative chart - not real market data.

What gap and go is not

It is not a chart pattern. It is a screen plus an entry.

It is not the same trade as fading a gap. They are opposite views of the same event, and both are described on the gap trading page.

It is not available on liquid large-caps. Those rarely gap far enough for the trade to exist.

And it is not risk-controllable in the usual way. Halts, slippage and gap continuation mean the loss on a failed trade is not reliably bounded by the stop distance.

When it fails

A sideways, range-bound candlestick series. The headline on the chart reads: In a range-bound name the gap fills and proves nothing.
In a range-bound name the gap fills and proves nothing. Illustrative chart - not real market data.

The most common outcome is the gap that fills. Price opens up, sellers arrive, and the whole move unwinds inside the first half hour — which is the ordinary behaviour of a gap with nothing behind it.

The second failure is the fill quality. Being right and paying several percent in spread and slippage produces the same result as being wrong, and it does not look like a mistake in a trading journal.

A third is the halt. A position frozen mid-trade with no way to exit is a risk that has no equivalent in most other strategies.

A fourth is chasing. The move is already large by definition, and entering late means the stop is enormous or arbitrary.

And a fifth is the survivorship in the teaching material. Every example shows a gap that continued. The far more numerous gaps that faded are not memorable and never get written up.

The original data

On this site’s shared 576-bar history: 53 closes above a 10-bar high, of which 43% were higher ten bars later against a 54% base rate, and 70% closed back below the broken level within ten bars. At a 20-bar lookback, 39 events, 38% higher and 85% back below. The round-trip cost of 0.0098 price units is 2% of the median bar range and 45% of the smallest bar. All in research/series-measurements.json, produced by site/measure_series.py.

A 72-bar window of the shared price history, cut short at the decision bar. The headline on the chart reads: Up twelve percent premarket on a press release. Buy?
Up twelve percent premarket on a press release. Buy? Illustrative chart - not real market data.

The measurement this strategy needs, and which no article supplies, is your own slippage. For every gap trade, record the price you clicked and the price you got, in both directions. The average of that difference is the real cost of the style, and on thin gapping names it is frequently larger than the planned risk on the trade — which means the strategy can be losing money for execution reasons while the analysis is working perfectly. Nobody else’s backtest measures your fills.

Opening gap explains what a gap is and why the two kinds behave differently. Gap trading covers the wider set of approaches. And penny stocks is the corner of the market where most of these candidates live.

What I actually do

The thing that made this style unworkable for me was not the losses on the trades - it was the fills. On the names that gap hardest, the difference between the price I clicked and the price I got was routinely larger than my entire planned risk.

— Michael Whitman

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