ITL-001 Turnaround Tuesday \ EURUSD, XAUUSD, SP500 — the results are different, there is no universal effect
Testing the Turnaround Tuesday effect: if Monday's daily candle closes bullish, a short position is opened on Tuesday. If Monday closes bearish, a long position is opened. The baseline version enters at the start of Tuesday and exits at the end of the trading day without Stop Loss or Take Profit.
This hypothesis is quite popular among equity index traders, so I wanted to see whether it also works on Forex and Gold.
Test Conditions
- Instruments: EURUSD, XAUUSD, SP500
- Test period: 2016–2026
- Entry: Tuesday open, against Monday's direction
- Baseline exit: Tuesday close
- Additional variations tested:
- ATR Filter
- Stop Loss as a fraction of ATR
- Take Profit based on Risk/Reward
The experiment code is open.
If you'd like to reproduce the tests, modify the parameters, or try the idea on other instruments, the EA is available in CodeBase: 001 - Turnaround Tuesday
Or download the compiled version directly: 001_Turnaround_Tuesday.ex5
Results
| Instrument | Version | Trades | Return | PF | Max DD | Recovery |
|---|---|---|---|---|---|---|
| EURUSD | Baseline | 518 | -1.34% | 0.87 | 1.55% | -0.84 |
| With Filters | 470 | -8.23% | 0.92 | 12.00% | -0.66 | |
| XAUUSD | Baseline | 496 | -2.31% | 0.93 | 7.32% | -0.29 |
| With Filters | 470 | +38.03% | 1.08 | 26.44% | 1.11 | |
| SP500 | Baseline | 413 | +0.03% | 1.06 | 0.05% | 0.55 |
| With Filters | 413 | +40.39% | 1.38 | 11.02% | 2.11 |
Observations
Without additional filters, the effect is practically nonexistent across all tested instruments.
EURUSD remains negative both in the baseline version and after parameter optimization. Adding the ATR filter, Stop Loss, and Take Profit did not produce a positive expectancy.
XAUUSD is also unprofitable in its baseline form. However, after adding volatility filtering and position management, the results become positive. That said, drawdown exceeds 26%, so the robustness of the edge still needs to be verified.
SP500 delivered the strongest result. The baseline version is essentially flat, while the filtered version produced a return of +40.39% with an 11.02% maximum drawdown. However, the result cannot be considered stable, as most of the profit was generated during the first half of the test period, while the second half remained largely flat.
An interesting observation is that the baseline version shows no meaningful edge on any of the three instruments. Positive results only appear after introducing additional trade selection and risk management rules.
Conclusion
The original Turnaround Tuesday hypothesis does not hold up as a universal market effect.
EURUSD remains unprofitable both in the baseline version and after optimization.
XAUUSD and SP500 only become profitable after adding volatility filters and position management rules.
At this stage, the key question is not profitability, but whether the optimized parameters are robust.
The positive results were obtained after parameter selection on historical data. The next mandatory step is out-of-sample testing on an independent period. Without it, there is no reliable way to separate a genuine statistical edge from curve fitting.
Follow the project so you don't miss the next research report
Telegram channel: @it_trader_ru
Telegram chat: @it_trader_chat_ru
