tests/reference.py vs the Prior Day Breakout (Free) EA in the MetaTrader 5 Strategy Tester
Symbol XAUEUR, model "1 minute OHLC", out of sample 2023-07-01 to 2026-10-09, broker M1 history (server time).
Executed 2026-10-21. The broker price history and the EA's trade lists are not published.

Preset H1, target 2 R:   reference 612 trades, EA 612 trades; same entry time and side: 612 (0 only in one of them)
                                  same exit minute: 521 of 612
Preset M15, target 1 R:  reference 690 trades, EA 690 trades; same entry time and side: 690 (0 only in one of them)
                                  same exit minute: 502 of 690

(tests/compare_trades.py on the EA's own TRD lines gives the same 612 / 612 entries for the H1 run.)

The same reference was also run against the research simulation the preset was selected with (a separate Python engine,
not published): 612 / 612 and 690 / 690 trades, identical entry time, side, exit time and exit price (largest difference 0.0).
Exit minutes differ from the tester because the tester builds its own tick path from M1 bars and fills at tick prices.