Trading the retest after the London open breakout fails

[FXS London Scalper]

A pattern I trade a few times a month: EURUSD breaks the Asian high at the open, fails, comes back inside the range, then retests the broken high from below. I short that retest.

From my log since April, 34 trades:

• Win rate 56% with a 1.5R target.
• The best trades came when the failed break was quick, under 15 minutes from the break to the return.
• When price spent more than 30 minutes above the high before failing, the retest short worked much worse.

The logic as I see it: a quick failure means the buyers who broke out are trapped, and the retest gives them an exit at breakeven. A slow failure is more like a range shift.

Stop goes above the failed-break high, not just above the Asian high. That's usually 6 to 9 pips.

Anyone else trading failed breakouts like this?

[Shadow Trader]

On a chart the first hour after the London open looks like one continuous move. In practice it is two different things that happen to be adjacent.

The first stretch, roughly the opening quarter hour, is about positioning and stop collection. Desks arrive with orders carried from overnight, spreads settle, and price frequently sweeps a nearby high or low before doing anything meaningful. Direction here is unreliable, and the candle that looks decisive at 8:05 gets reversed by 8:20 more often than people like to admit.

The remainder of the hour is about information. By then Europe has digested the overnight session, any early data has printed, and price has shown which side of the Asian range it wants to accept. A move that holds after the opening sweep carries more weight than one that started with it.

The two look alike because both involve fast candles and wide ranges. They differ in what the speed means. Early speed often comes from clearing resting orders, and later speed from actual commitment.

For a scalper this suggests different tools. The first window suits quick reactions to the sweep or no trading at all. The second suits trading the retest of a level that was just accepted.

Do you treat the first 15 minutes of London differently from the rest of the hour, and has that distinction ever survived a proper sample?