Stop hunts or just spread widening?
Posted: Thu Oct 29, 2026 3:54 am
Almost every scalper has a story about a stop getting hit by a spike that reversed immediately. The default explanation is a stop hunt. The less dramatic one is often a spread widening for a few seconds.
Here is the mechanical detail. A sell stop on a long position triggers on the bid. A buy stop on a short position triggers on the ask. When spreads blow out during thin moments, such as the rollover window or just before a data release, the ask jumps while the bid barely moves, or the reverse. A chart showing only bid prices makes the spike on the other side invisible.
The fix for diagnosis is simple. Turn on the ask line on the chart, and check whether the spike appears on the bid, the ask, or both. A spike visible only on the ask was a spread event, and the stop on a short would have been triggered by it.
That does not make it fair, but it does make it explainable, and explainable problems can be traded around.
The tick history gives a further check. If the platform allows exporting bid and ask ticks, the spike can be replayed at full resolution. A true price move shows both sides shifting together. A spread event shows the gap between bid and ask opening up, often by several pips, with the mid price barely changing. Looking at ten or twenty stop-outs this way separates the genuine sweeps from the spread artefacts, and the proportion in each group says a lot about where the risk really comes from.
It also helps to note the time of each incident. Spread events cluster around rollover, news, and the very start of sessions, which means a simple schedule filter removes most of them. Genuine sweeps are spread more evenly and call for a different response, such as placing stops beyond obvious clusters or reducing size near them.
Which pairs and times of day have caught your stops most often through spread spikes rather than genuine price moves?
Here is the mechanical detail. A sell stop on a long position triggers on the bid. A buy stop on a short position triggers on the ask. When spreads blow out during thin moments, such as the rollover window or just before a data release, the ask jumps while the bid barely moves, or the reverse. A chart showing only bid prices makes the spike on the other side invisible.
The fix for diagnosis is simple. Turn on the ask line on the chart, and check whether the spike appears on the bid, the ask, or both. A spike visible only on the ask was a spread event, and the stop on a short would have been triggered by it.
That does not make it fair, but it does make it explainable, and explainable problems can be traded around.
The tick history gives a further check. If the platform allows exporting bid and ask ticks, the spike can be replayed at full resolution. A true price move shows both sides shifting together. A spread event shows the gap between bid and ask opening up, often by several pips, with the mid price barely changing. Looking at ten or twenty stop-outs this way separates the genuine sweeps from the spread artefacts, and the proportion in each group says a lot about where the risk really comes from.
It also helps to note the time of each incident. Spread events cluster around rollover, news, and the very start of sessions, which means a simple schedule filter removes most of them. Genuine sweeps are spread more evenly and call for a different response, such as placing stops beyond obvious clusters or reducing size near them.
Which pairs and times of day have caught your stops most often through spread spikes rather than genuine price moves?