Advertisement IC Markets

Demo to live: a checklist before attaching an EA to real money

Share, develop, and backtest custom MQL4/MQL5 Expert Advisors, Python data-scraping scripts, trading bots, and automated market alert systems.
Post Reply
FXS London Scalper
Posts: 976
Joined: Sun Oct 18, 2026 12:25 pm

Demo to live: a checklist before attaching an EA to real money

Post by FXS London Scalper »

EAs moved from demo to live very often do worse at first. A checklist catches the usual reasons before money is at stake.
  • Same broker and the same account type on demo and live? Demo servers can have different spreads and execution.
  • Same symbol names, digits and contract sizes? Suffixes like .r or .pro break symbol checks.
  • Commission included in the backtest and demo? Some demo accounts charge none.
  • Stop level and freeze level identical? Read them from the live specification.
  • Lot size calculated from live balance and the live account's minimum step?
  • Magic number unique, so the EA doesn't touch manual trades or other EAs?
  • Behaviour after a restart: does it recognise open positions?
Then run live at minimum size for a few weeks and compare each live trade with the demo trade at the same moment. Differences in fills and timing will show up quickly.

A worked example: a strategy making 0.3 pips per trade on demo with no commission becomes negative with $7 round-turn commission (0.7 pips). The checklist catches that before the first live trade.

What's on your checklist?
FXS Official Team | London Session Desk
Recommended broker for automated trading & scalping IC Markets
Amanhasnoname
Posts: 2060
Joined: Fri Oct 30, 2026 9:45 am

Re: Demo to live: a checklist before attaching an EA to real money

Post by Amanhasnoname »

I'd add a point about time. Check that the server time offset on the live account matches the demo, because I once moved an EA between two brokers where one used GMT+2 and the other GMT+3 in the summer. My session filter, which only allowed entries between 08:00 and 11:00, was shifted by an hour on the live account and the EA was trading the opening spike I'd always avoided. The first two weeks looked like a different strategy.

The second thing I'd put on the list is the trading hours of the symbol. Some brokers close certain pairs for a short break every day, and my EA had a trailing stop that couldn't modify the order during the break, which gave it a stale stop for twenty minutes.

For the first live weeks I keep a simple side-by-side file that records, for each entry, the demo fill price and the live fill price taken at the same minute. After about fifty trades the average difference is a number I can plug straight into the backtest as realistic extra cost.
PTScalper
Site Admin
Posts: 7821
Joined: Mon Jul 20, 2026 1:28 pm

Re: Demo to live: a checklist before attaching an EA to real money

Post by PTScalper »

FXS London Scalper wrote: Tue Nov 03, 2026 4:19 pm EAs moved from demo to live very often do worse at first. A checklist catches the usual reasons before money is at stake.
  • Same broker and the same account type on demo and live? Demo servers can have different spreads and execution.
  • Same symbol names, digits and contract sizes? Suffixes like .r or .pro break symbol checks.
  • Commission included in the backtest and demo? Some demo accounts charge none.
  • Stop level and freeze level identical? Read them from the live specification.
  • Lot size calculated from live balance and the live account's minimum step?
  • Magic number unique, so the EA doesn't touch manual trades or other EAs?
  • Behaviour after a restart: does it recognise open positions?
Then run live at minimum size for a few weeks and compare each live trade with the demo trade at the same moment. Differences in fills and timing will show up quickly.

A worked example: a strategy making 0.3 pips per trade on demo with no commission becomes negative with $7 round-turn commission (0.7 pips). The checklist catches that before the first live trade.

What's on your checklist?
Hi,

you are right, from my point of view it is really important to start with demo account deployment of your EA strategy after like backtesting.
Because you never know and it is better to make optimalisation and potential bug fixing before real hard earn money are at risk.

I personaly do it like this:

1) I got an idea for trading EA.
2) With AI i will write that EA code like in MQL5 or Csharp for ctrader.
3) Make basic back testing on 6 months, 2 years and 10 years period to get like basic output, if that idea is strong enought and proven in time, not only
in some short period of time.
4) Create backtesting on combinations of combinations of variables, which i can change (up to 10 000 combinations) with M1 charts.
5) Top 20 combinations i will back test with tick by tick data.
6) Make final check up with several different AI models.
7) Deploy in on cloud instance and run it for 1 week on demo account with extended logging from EA.
8) Make analysis of logs with AI.
9) Fix potencial issues, errors, bugs.
10) Run it for next 2-4 weeks on small live account.
11) Fixing all potential issue and make it as good as possible.
12) Start to run in on normal size live trading account.
13) If i like the output i will create another 2-3 options of that EA with different account size and risk setup.
14) Once a week i check all posible statistics, logs etc.
Preserve your own money. Scale with the market's money. Exponential growth is the ultimate key.
Post Reply