Testing & evaluation
Why the same EA backtest differs between brokers
The same EA file and settings can produce different trades in different environments. Quotes, history coverage, sessions, contract rules and costs all influence what the program observes and what the tester accepts.
Freeze the common inputs
Use the same EA build, explicit dates, timeframe, deposit and sizing mode. Then record each broker’s symbol name, contract size, volume rules and timezone. A default value can depend on the environment.
Find the first divergence
Compare the first different signal or order, not only the ending return. Was an indicator value different? Was an entry rejected? Did one history set contain a tick that the other did not?
Isolate one cause at a time
Once you identify a likely difference, run a short interval around it. Keep costs and data distinctions visible. Adjusting many parameters until the profits match hides the explanation.
Use differences as information
A portable project makes assumptions easier to inspect. MQLume source access supports that investigation, while each published review remains tied to its stated broker and data conditions.
Start with a free editable EA project · Explore MQLume source code
Back to top