If you think you’ll profit by moving TP far away, the PF kept falling—stopping the fixed pips and continuing until reaching ATR

“If profits extend, you’ll earn more.” When I extended the take-profit, performance kept dropping.

Up to the previous part, the entry conditions (First Touch of M15, BB2.9, ADX22.5) were solidified. Next, I tackled the exit: where to take profits and where to cut losses. At first, I also thought, “the more it goes, the more it grows.”

When TP was extended, PF kept falling

Keeping the entry the same, I extended only the TP (take-profit width). Here are the PF (Profit Factor) results.

※ The bar length indicates the portion exceeding PF1.0 (breakeven) from the baseline.

From TP100 to TP300, it shows a clear downward slope. At TP300, PF is almost 1.00. Whether winning or losing, it nearly breaks even.

“Stretching further yields bigger profits” did not hold.I interpret that with far TP targets, cases where it returns before reaching them increased.

Averaging (grid) add-ons improved numbers, but were not adopted

Next, I tried adding 0.01 lot once when the unrealized loss reached -20 pips. I compared five currencies across 643 cases.

PF and average P/L were better with additions. Looking at the numbers alone, additions win.

Still, I didn’t adopt it. The loss-burden when losing would grow from about -40p to about -60p, which is 1.5 times larger.Backtest numbers being good and a product that can be safely released are separate matters.Boundary does not include this addition.

Fixed-width trailing was not a decisive factor

“Because it fixes the TP, you miss opportunities in trending markets. If it trends, you should chase it.” With that thought, I compared fixed TP100 with trailing after reaching +100 pips at EURUSD in 2025.

※ Amounts are based on the validation settings.

Both the number of trades and win rate and PF were identical. Net profit was almost the same, slightly lower by a few hundred yen. The maximum drawdown improved slightly from 12,232 yen to 11,629 yen.

The slight improvement in DD is not bad, but profits did not increase. Fixed-width trailing did not become a decisive factor in this configuration.

Adjusting fixed pips hit a dead end here

Putting the results side by side up to this point yields:

• Extend TP → PF kept decreasing
• Add once via grid → numbers improved, but loss burden increases 1.5x
• Fixed-width trailing → profits barely changed

Adjusting fixed pips hit a dead end here. So the next change was not to adjust numbers, but to change the yardstick of the exit itself.

Using the H4 ATR (14 bars) as the yardstick, the exit distance now varies with market size

This is the current exit for Boundary.

Summary

• Extending TP caused PF to fall from 1.221 → 1.106 → 1.055 → ≒1.00
• One extra grid addition improved numbers, but the loss burden increased from about -40p to about -60p, so it was not adopted
• Fixed-width trailing did not become decisive as net profit remained almost unchanged
• Adjusting fixed pips reached a dead end; changed the exit yardstick to H4 ATR

Next:It isn’t just about choosing currencies with high win rates. It took until it was EURUSD-only, compared against USDJPY, GBPUSD, and GBPJPY.

※Past validation results are not a guarantee of future performance.