forex-scalping.com

Sometimes gold moves significantly in the late US session or early Asia, creating a gap from the London close. By the time London opens, price is far from where it was the previous afternoon.

Options:

Trade it in Asia. If you're awake and the move has a clear driver, Asia can offer good entries. But liquidity is thinner and reversals sharper.

Wait for London. Let Europe show whether the overnight move is accepted or rejected. More information, better liquidity, but you might miss part of the move.

Fade the gap at London. If the overnight move had no clear driver, London sometimes pushes price back toward the previous close.

Questions that help decide:
- Was there a news driver? Fed comments, geopolitical events, US data revisions?
- Did the move happen on high activity, or did price drift there in thin trade?
- Is the new price near a significant level from the higher timeframes?
- Did yields and the dollar move in the same way?

A driverless gap in thin trade is a candidate for a fade. A news-driven gap with confirmation from yields is more likely to hold.

What's your default when gold has moved overnight?

A common belief says weekend gaps almost always fill, that price returns to Friday's close within a day or two. It's repeated confidently and rarely checked.

On major pairs, gaps are usually small, a few pips, and many do close within the first hours simply because the opening spread and thin liquidity create a distorted first print. That is the part of the story with some truth.

Larger gaps, those caused by genuine weekend news, tell a different story. They can stay open for weeks or never close. A trader who fades a large gap expecting a fill can sit on a losing position as the market continues away from it.

The statistics depend heavily on how a gap is defined. Using bid prices at different brokers, with different weekend closes, produces different results. Measuring from the first tradable ask to Friday's last bid inflates the gap.

A proper test needs consistent data and a clear definition: gap size in pips, filled within an hour, a day, a week, never. Then group by size.

Most people who run it find that the answer depends on size, and that the profitable zone is narrow and eaten by the spread.

How have gap-fill results looked on real data, split by gap size?