Lim Wes Chun / 个人资料

"Every tick" vs "Every tick based on real ticks": why your backtest might be lying to you

Many traders run an EA in the MT5 Strategy Tester, see a beautiful equity curve, and buy or go live. Then the live results look nothing like the backtest. Very often the reason is the modelling mode.

Every tick
The tester does not use real market data at tick level. It takes M1 bars and generates artificial ticks inside each bar using an interpolation algorithm. Price moves inside the candle are estimated, not recorded.

Fast, and fine for slower strategies on higher timeframes
Price paths inside the bar are synthetic
Spread behaviour is simplified, so news spikes and spread widening are not realistic
Scalpers and tight-SL/TP systems can look much better than they really are

Every tick based on real ticks
The tester replays the broker's actual recorded tick history: every bid/ask change, with the real variable spread at that moment.

Closest thing to live trading the tester can offer
Real spread widening during news, rollover and low liquidity
Shows whether stops and targets would really have been hit, and in what order
Slower, and needs a lot of downloaded tick data
Quality depends on the broker's tick history, so check that the report shows high modelling quality

Why it matters on XAUUSD
Gold moves fast and its spread can jump sharply around news and the daily rollover. A strategy that survives on generated ticks can fall apart once real spread and real price sequence are applied. For gold, especially for scalping and short-term systems, real ticks are the test that counts.

My rule:
Every tick → quick idea checking only.
Every tick based on real ticks → the only result I trust before going live.

When you look at any EA's backtest, mine included, always ask: "Was this tested on real ticks?"

GEA-X — a note on where my head is at 🟡

GEA-X has been live on the MQL5 Market for a few days now. It's my first product here, and I want to share honestly how I'm thinking about it.

I didn't build it to look good in a backtest. Anyone can tune an EA until the curve is a perfect straight line. I built GEA-X for the real gold market, with all its volatility and surprises. Risk control comes first in every decision I make, and I won't compromise on that.

Every update comes from watching it trade, not from the optimiser. I follow how GEA-X behaves on real charts, and when I see something that can be done better, I fix it. Version 1.3 is the result: cleaner trade management, lower drawdown and a smoother equity curve.

Don't take my word for it — test it yourself. Download the free demo and run it in the Strategy Tester on your own broker's data, or put it on a demo account for a few weeks. That's exactly what I'd do before buying any EA.

About the price. I've set GEA-X at what I believe is a fair price for what it delivers, with ongoing updates included. This is the launch price, offered only for this early stage. As GEA-X builds its track record, the price will go up. Early users get it at its lowest, and that's my way of thanking the people who trust it first. If you'd rather start small, the rental options let you run it live at low cost before committing.

Early users matter most to me. If you're testing GEA-X on demo or live, message me with anything you see: a strange trade, a broker issue, a question about settings. Your feedback shapes the next version.

Slow and steady. Build it right, show it honestly, and let the results speak.

— Wes

🔗 https://www.mql5.com/en/market/product/198012

已在MQL5.community注册