That produces a bad combination: poor fill rate on the trades that would have worked, and fills on the ones that don't.
How to test it on your own orders: tag every round-number limit order, filled or not, and record:
- Whether it filled.
- If not, the closest price reached.
- If filled, the result in R.
The common adjustment is placing limits 1 to 2 pips in front of the round number. Run it side by side, or on paper, and compare fill rate and average outcome. The offset should raise fills on reversals while reducing fills on breaks.
The same logic applies to stops: placing them just beyond a round number, not on it, avoids being taken out by the obvious sweep.
Do others offset entries from obvious levels? How far?