Backtest vs live results: why we publish Myfxbook instead of screenshots
Every EA seller shows you a beautiful equity curve. The uncomfortable truth of this industry is that a beautiful backtest proves almost nothing: with enough parameters, you can fit a strategy to any historical data and produce a curve that goes up and to the right — right up until it meets live markets.
The three gaps between backtest and live
Execution: backtests fill at the price you ask. Live trading has spread variation, slippage, and requotes. On XAUUSD during volatile sessions, the difference is measurable on every single trade.
Data quality: most backtests run on broker M1 data with interpolated ticks. Strategies that rely on precise intrabar behaviour — tight trailing stops, quick partial closes — behave differently on real ticks.
Overfitting: the silent killer. If a strategy needed 40 parameter tweaks to look good on 2019–2024 data, it has memorised the past, not learned the market.
How we validate before release
Our pipeline is: multi-year backtest on quality tick data first, then a forward test on demo, then live capital — our own — before any commercial release. Golden Tiger EA traded live for months before the first license was ever sold.
The live account is published on Myfxbook with broker verification enabled. Myfxbook reads results directly from the trading account — we can't edit a losing week out of the record, and that's exactly the point.
How to evaluate any EA (including ours)
Ask for live verified results, not screenshots. Check the track length — three good weeks means little, six-plus months means something. Look at drawdown, not just gain. And be suspicious of anything promising fixed monthly returns: markets don't pay salaries.