So is waiting two bars even better? No, it is worse

Adding extra waiting time to an algorithmic strategy actually degrades performance, confirming that the current “one-bar confirmation” method is the optimal sweet spot for my systems. I recently tested two variations against my standard setup: requiring two consecutive candle closes to confirm a breakout (confirm2) and delaying execution by one bar to isolate the cost of waiting (delay1). By keeping the core engine identical and comparing these against my previous research, I could see exactly what that second bar of information was worth.

Why waiting hurts

In the Trend FX strategy, forcing a second confirmation bar didn’t just filter out false signals. Instead, it aggressively trimmed the profitable trades, leading to a drop in both return and profit factor (PF: the ratio of gross profit to gross loss, where a value over 1 indicates profitability). Interestingly, the “delay1” test showed that the pure cost of waiting one bar is negligible at H1, H4, and D1 timeframes. This proves that the information gained from that second candle is effectively negative. For the TJL-Gold system, the results were even clearer. The performance dropped significantly when I added a second confirmation bar. This suggests that the core edge of this strategy relies on the drift occurring immediately after the breakout. Waiting even a single day causes the potential profit to evaporate.

The trade-off in indices

I did see a slight improvement in the profit factor for indices when delaying the exit (confirm2-out); it moved from 5.52 to 6.36. However, this came at a cost: the monthly return dropped from 0.169% to 0.150%, and the drawdown worsened from -4.5% to -6.5%. In other words, I was sacrificing both volume and safety for a marginal increase in quality, which is not a trade-off I want to make.

Verdict

The performance curve for waiting time is a single peak, and that peak sits squarely at “one candle confirmation.” If I reduce the wait time, I pick up too much noise. If I increase it, I erode the initial momentum that drives the profit. Both the FX and gold engines are currently operating at their optimal point, so I will be keeping the logic exactly as it is.

How this connects

This verification builds on earlier ones (what failed before and what I tried this time, comparisons between approaches).

• Restricted to gold, the viral method became a real edge
• Waiting for the closed bar beats jumping in mid-bar, everywhere