forex-scalping.com

My entry rules for the first pullback after a London impulse, as they are written on my card:

1. The level must exist before the open. Asia high or low, previous day's high or low, or a round number that mattered overnight. Not something I draw after the move starts.
2. The impulse must be clear: at least 1.5 times the average five-minute range, in one direction, through the level, with normal spread.
3. The pullback must come back toward the level without breaking it, and it must not coincide with a data release.
4. Entry on the first M1 candle that closes in the impulse direction. Stop beyond the pullback low or high. Time stop of 20 minutes.

If any of the four is missing, no trade. The most common failure is point 1: I see a strong move and then go looking for a reason. That is not a first pullback setup, that is chasing.

After a messy open, with two or three fake breaks before 08:30, I skip the whole setup for the day. Pullbacks in chop are just more chop.

Win rate on this is around 50 percent for me, but the winners run further than the losers, which is the entire point.

LondonScalper wrote: at least 1.5 times the average five-minute range

The 1.5 times rule needs its averaging window pinned down, or it will drift. If the average five-minute range is taken over the last 20 or 50 bars at 08:00, it's mostly Asia bars, which are small, so almost any London bar clears 1.5 times and the filter does nothing. A more stable version is the average range of the same five-minute slot over the previous 20 days, so a bar at 08:05 is compared with other 08:05 bars. Then 1.5 times means genuinely unusual for that time of day, and it also copes better with Mondays and news days. A second detail worth writing down is whether the impulse is one bar or a sequence, since three medium bars in one direction can cover more ground than one big bar and still fail the rule as written.

LondonScalper, the rule I would add to the card is a depth limit on the pullback. On my EURUSD notes the good ones retraced between about 30 and 60 percent of the impulse, while anything that gave back more than 70 percent usually broke the level within the next few bars.

Timing helps too. If the pullback takes more than 45 minutes to arrive, I no longer count it as the first one, since the London flow that drove the impulse has often faded by then.
The illustration shows how I read it on M5, with the Asia high marked before 07:00, the impulse through it, and the entry zone on the first proper pullback. I ignore the small dip straight after the break, because that is spread and stops being cleared rather than a real retest.

On your 20 minute time stop, do you measure it from entry or from the M1 close that triggered it? With a stop beyond the pullback low, a slow first leg can eat most of the window before anything happens.

Your point about skipping after two or three fake breaks matches my log. Those days produced most of my losers on this setup, even when all four conditions looked met.