A 14-day demo plan for testing any free EA (we use our own as the example)

Most traders test a free EA like this: download it, attach it to a chart, watch it for two hours, get bored, and either delete it or, worse, put it on a live account. Neither decision is based on evidence. Here is a simple 14-day demo plan that works for any EA, free or paid. We will use our own free robot, Wonder 8, as the concrete example, but the plan itself is generic: you can run it on any EA you find on the Market.

Day 0: set up the demo properly

Open a demo account with the broker you actually plan to use later, not a random one. Conditions differ: spread, commission, stop level, swap. Then:

• Attach the EA to the exact symbol and timeframe the seller specifies. For Wonder 8 that is USDJPY on H1.
• Keep default settings. You are testing the product as published, not your own variant.
• Make sure the platform runs 24/5: a VPS or a PC that never sleeps. An EA that misses sessions gives you useless data.
• Note the starting balance and the date. This is your baseline.

Days 1 to 3: does it behave as described?

In the first days you are not judging profit. You are checking behavior. Every time a position opens, verify one thing above all: does it have a stop loss and a take profit attached from the start? Wonder 8 opens every trade with both, at a fixed 0.01 lot. If an EA you are testing opens naked positions, or starts adding positions to a loser, you have already learned what you needed to know.

Days 4 to 7: watch the clock, not the balance

Write down the time of every trade. Most rule-based EAs have a rhythm: a breakout system like Wonder 8 tends to act when the market actually moves, not at random hours. Compare what you see with the backtest the seller published. You are looking for coherence: similar trade frequency, similar hold times, similar behavior. A demo that looks nothing like the published backtest is a red flag, whatever the profit says.

The first weekend: an underrated test

Check the account on Saturday. Are positions held over the weekend? Is that consistent with what the seller describes? Gaps at Monday open are where hidden risk lives. Note whether your EA carries exposure into them and how large the stop distance is if it does.

Days 8 to 14: consistency and stress

The second week is about repetition. Keep the same daily log:

• Each trade: open time, close time, stop loss and take profit present, result.
• The spread at the moment of entry, especially around news releases.
• Anything unexpected: modified stops, skipped signals, error messages in the Experts log.

If a high-impact news day falls in your window, even better. You want to see how the EA behaves when the spread widens: does it keep its rules, or does it do something the manual never mentioned?

Day 14: the decision

After two weeks you will not know if the EA is profitable. Two weeks is far too small a sample for that, and anyone who claims otherwise is selling you noise. What you will know is whether the EA is honest: rules respected, stops always in place, behavior coherent with the published tests, no surprises in the log. That is the filter. Profitability is a question for months of forward data and for backtests with a real sample size, and remember that backtest results and demo results do not guarantee future performance.

If you want to run this plan on something concrete this week, Wonder 8 is free, trades USDJPY H1 with a fixed 0.01 lot, and opens every position with a stop loss and a take profit. It is available for MT5 and MT4. Put it on a demo, start the log, and judge it with your own eyes.