Build a Simple Strategy From Scratch in Seven Steps
You don't need a complicated system to start. A simple, clear strategy you understand is better than a complex one you can't follow. Here's a step-by-step process.
Step 1: Choose your market and session. One pair, one window.
Step 2: Define your bias method. For example, higher-timeframe structure on the 4H.
Step 3: Define your location rule. For instance, only buy in discount and sell in premium.
Step 4: Define your trigger. A liquidity sweep followed by a structure shift on a lower timeframe.
Step 5: Define risk rules. Stop placement, risk per trade, maximum trades per day, daily loss limit.
Step 6: Define exits. Target at a specific liquidity level, partial profit rules, break-even rule.
Step 7: Write it as a checklist. Yes or no questions, so there's no ambiguity.
Then test it:
1. Backtest at least 100 setups on replay.
2. Calculate win rate, average R, and drawdown.
3. Forward test on demo for several weeks.
4. Start small live if results and discipline hold.
Common mistakes:
- Adding too many conditions
- Changing rules after a few losses
- Using vague language like "strong move"
Remember that simple rules are easier to execute under pressure.
A strategy isn't a secret formula. It's a set of clear decisions you make in advance, so that you don't need to improvise when emotions run high.
Build a Simple Strategy From Scratch in Seven Steps
Build a Simple Strategy From Scratch in Seven Steps
It’s Fairman 
Re: Build a Simple Strategy From Scratch in Seven Steps
One more thing: when you get to the testing step, write the rules down before you open any charts. Otherwise you'll end up tweaking them to fit whatever you're looking at, and the strategy you test isn't the one you'll actually trade. A good check is to give your written rules to someone else and ask them to find the trades on a chart. If they get the same trades you would, the rules are clear enough. If they keep asking whether something counts, the rules need tightening. Vague rules are the biggest reason simple strategies fail, because every decision becomes a judgment call, and judgment changes with your mood.
It’s Fairman 
Re: Build a Simple Strategy From Scratch in Seven Steps
At step one, people tend to choose too many markets. A simple strategy on one or two pairs is easier to test and learn. Once it works there, you can test whether it carries over to others. Each pair has its own habits, typical range and news schedule, and learning them takes time. EURUSD in the London session is a sensible first choice for most people in Lagos, with tight spreads and active hours from 8am. After 100 trades with clear results, try the same rules on GBPUSD in backtesting before trading it live, since the wider moves may need different stops.
It’s Fairman 
-
FXS Prop Desk
- Posts: 1689
- Joined: Wed Oct 07, 2026 9:52 am
Re: Build a Simple Strategy From Scratch in Seven Steps
Decide in advance how many trades are needed before evaluating the strategy, for example 50 in demo. That prevents giving up too early or changing rules after a few losses.
Reviewing the strategy after the test period, and making only small changes based on data, keeps it improving.
Writing the rules in a single document, with examples of valid and invalid setups, also helps apply them consistently.
Sharing the rules with another trader for feedback can reveal unclear points before testing begins.
Reviewing the strategy after the test period, and making only small changes based on data, keeps it improving.
Writing the rules in a single document, with examples of valid and invalid setups, also helps apply them consistently.
Sharing the rules with another trader for feedback can reveal unclear points before testing begins.
FXS Official Team | The Idea Lab & Strategy
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Build a Simple Strategy From Scratch in Seven Steps
Step 4 is where most simple strategies quietly stop being simple. A liquidity sweep needs a number attached before it can go on a yes or no checklist, for example price trading at least 2 pips beyond the Asian low and closing back inside within three M5 candles.
The same goes for the structure shift. Deciding whether a wick break counts or only a candle close is the kind of detail that changes the trade list completely when two people test the same rules.
On sample size, 100 setups is a sensible floor, but it is worth knowing how wide the uncertainty still is. With a true win rate of 50 percent, a 100 trade test can easily come out anywhere from about 40 to 60 percent just by chance.
That is why I would treat the 50 demo trades mentioned above as a checkpoint for execution rather than a verdict on the edge. Fifty trades can tell you whether you follow the rules, but not much about whether the rules make money.
Would you count skipped valid setups as part of the test? Missing three good trades out of ten says as much about whether a strategy fits your life as the win rate does.
The same goes for the structure shift. Deciding whether a wick break counts or only a candle close is the kind of detail that changes the trade list completely when two people test the same rules.
On sample size, 100 setups is a sensible floor, but it is worth knowing how wide the uncertainty still is. With a true win rate of 50 percent, a 100 trade test can easily come out anywhere from about 40 to 60 percent just by chance.
That is why I would treat the 50 demo trades mentioned above as a checkpoint for execution rather than a verdict on the edge. Fifty trades can tell you whether you follow the rules, but not much about whether the rules make money.
Would you count skipped valid setups as part of the test? Missing three good trades out of ten says as much about whether a strategy fits your life as the win rate does.
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Build a Simple Strategy From Scratch in Seven Steps
Step 6 is the one I would keep deliberately crude for the first test. Partials and a break even rule each change results, and adding them from the start hides which rule is doing the work.
Say 100 trades use a 1R stop and a 2R target, with 40 winners and 60 losers for plus 20R. Now add a move to break even once a trade reaches 1R.
Suppose 15 of those winners dipped back to entry after reaching 1R, so they now scratch at zero. If 10 of the losers had also touched 1R before turning, they scratch too, leaving 25 winners for plus 50R, 50 losers for minus 50R and a flat result.
The equity curve looks smoother with fewer full losses, yet the 20R is gone. Real numbers could go the other way, which is why the plain version gets tested first.
Fairman's test of handing the rules to someone else is a useful check on the exits as well. Two people often agree on the entry and then manage the trade differently once it is 1R up.
Does anyone keep a fixed target through the whole first sample? Break even seems to be the rule most often added untested.
Say 100 trades use a 1R stop and a 2R target, with 40 winners and 60 losers for plus 20R. Now add a move to break even once a trade reaches 1R.
Suppose 15 of those winners dipped back to entry after reaching 1R, so they now scratch at zero. If 10 of the losers had also touched 1R before turning, they scratch too, leaving 25 winners for plus 50R, 50 losers for minus 50R and a flat result.
The equity curve looks smoother with fewer full losses, yet the 20R is gone. Real numbers could go the other way, which is why the plain version gets tested first.
Fairman's test of handing the rules to someone else is a useful check on the exits as well. Two people often agree on the entry and then manage the trade differently once it is 1R up.
Does anyone keep a fixed target through the whole first sample? Break even seems to be the rule most often added untested.