Gap fill and gap-and-go
BreakoutIntermediate
A gap is an open that jumps past the prior close, usually on overnight news. Two opposite rules are both widely used. Gap-and-go trades in the direction of the jump. Gap fill bets that price will drift back toward the prior close. They are not the same strategy, and the open has to be classified before either rule is applied.
Not investment advice. Nothing here is a recommendation to buy or sell anything. Backtests and historical studies are not guarantees, and a rule that worked in one sample can fail in the next. Trading can lose money. Links are for education, not endorsements. Do your own research.
- Rules
- Measure the gap from the prior close to the open, and ignore small gaps inside the prior day's range if you only want full gaps. Gap-and-go buys a gap up that holds above the opening bar, with a stop under that bar, and is more plausible when volume is heavy and the gap clears a known level. Gap fill fades the open back toward the prior close, with a stop beyond the opening extreme. News gaps and exhaustion gaps are handled differently from ordinary opening gaps.
- Where AI fits
- Classifying the gap is the AI job: news versus no news, gap size versus recent volatility, and whether the open is a breakaway or a fade candidate. An LLM can read the overnight headline and tag the event, while a simpler model decides if that tag has historically been followed by continuation or fill. The order logic itself should stay deterministic.
- Markets
- Stocks and index futures with a daily halt or a clear session open. Continuous FX gaps mainly over the weekend.
- Timeframes
- The decision is at the open, managed on 1- to 15-minute bars, usually flat the same day.
- Risks
- Fading a real news gap can be a loss that never fills. Chasing a gap that immediately reverses is the other loss. Stops do not protect you from a further jump at the next halt. Historical fill rates are not a promise about the next open.