What you'll learn
- Write entry rules that are unambiguous
- Design exit rules for both losing and winning trades
- Use filters to avoid conditions the strategy is not built for
Entries
An entry rule names the trigger precisely: a candle close, a break of a level by a set number of pips, a limit order at a specific retracement. Market versus limit is part of the rule. Anticipation entries fill better and fail more; confirmation entries fill worse and fail less. Pick one per setup and keep it.
Exits
- The stop: where the idea is wrong, beyond the noise, set before entry, never widened.
- The target: a fixed level, a multiple of the stop, or the next structural level.
- Trailing: a rule for moving the stop as the trade works, such as behind each new swing low.
- Time: closing a trade that has not moved in a set number of bars.
- Partial exits: taking some at a first target and trailing the rest, if the strategy defines it.
Filters
Filters keep the strategy out of conditions it was not built for: trading only with the higher timeframe trend, only during the London and New York sessions, not within an hour of high-impact news, not when ATR is below a threshold, not on the day before a central bank decision. Each filter reduces the number of trades and, if chosen well, raises the quality. Filters are where most of the improvement in a young strategy comes from.
Testing rules one at a time
Change one rule, run the test, compare. Change two and you cannot tell which one helped. The backtesting lesson explains the process; the point here is that rules are hypotheses and the record is the evidence.
Key takeaways
- Entries are precise triggers; confirmation costs price and saves failures.
- Exits include the stop, the target, trailing, time and partial rules.
- Filters remove bad conditions and are the main source of early improvement.
- Change one rule at a time.
Knowledge check
Trading forex, CFDs and other leveraged products carries a high risk of losing money. This lesson is general education, not advice. Risk disclosure.