forex-scalping.com

Turn a Story Into Rules You Could Hand to a Stranger

A trader describes their method: "I buy when price dips to support in an uptrend and then bounces." It sounds like a strategy. Test it by asking whether a stranger could trade it identically.

Probe each phrase. "Uptrend" on which timeframe, and defined how? "Dips to support" means touching a level, coming within some distance of it, or closing at it? "Support" is which levels, drawn by whom? "Bounces" means what, a candle color, a break of the previous high, a 10-pip rise?

Now rewrite it with numbers. Uptrend: on the four-hour chart, price closes above a rising 50-period average. Support: a zone made from the last two swing lows on the one-hour chart, with the zone defined as the lows plus 5 pips. Dip: price trades into the zone. Bounce: a one-hour candle closes back above the zone's top and above its own open. Entry: buy stop 1 pip above that candle's high. Stop: 5 pips below the zone's lowest point. Target: previous swing high, with a minimum reward of 1.5 times risk, else skip.

Count what changed. The story had five vague terms. The rule set has seven checkable conditions, each answerable with yes or no.

Why this matters: only the second version can be backtested, and only the second version can be followed the same way on a bad day.

Your turn: take your own favorite one-sentence strategy and rewrite it until nothing in it requires a feeling.

Turning vague ideas into precise rules is the first step in testing any strategy. It also helps with consistency, since the same setup is traded the same way each time. It is surprising how much changes when a method is written with numbers.

The rewrite shows how much is hidden in a single sentence. Five vague terms become seven checkable conditions, each with a yes or no answer. That's the difference between a method that can only be followed by the person who invented it, on a good day, and one that anyone could follow the same way. It's also the difference between a method that can be tested and one that can't. A backtest of the original story would require the tester to make judgement calls on every trade, which inevitably introduces bias toward the trades that worked.

Mine used to be: I buy pullbacks on EUR/USD in the London session when it is trending up. I could not explain why I skipped some days and took others, so I wrote rules. The window is 08:00 to 10:30 London time. Trend is up when the five minute close is above a 50 period EMA and that EMA is higher than six bars ago. A pullback means price touches the 20 EMA on the one minute chart. The trigger is a candle that closes back above it, with a buy stop half a pip above its high.

The stop goes one pip below the pullback low, and if that makes the risk more than six pips I skip. Target is 1.5 times the risk. After two losses in the window I am done for the day.

When I backtested the written version on about 120 setups, I found three of my old favorites had been trades that broke my own conditions, and they were net negative. Uncomfortable, but worth it.

The minimum reward rule, 1.5 times risk or skip, is a useful filter. It means the trader only takes setups where the target is far enough from the entry to justify the stop. Without it, the method might take trades with poor ratios simply because the conditions were met. Counting how many setups this rule filters out, and comparing their outcomes with the ones taken, would show whether the filter improves results. Sometimes the filtered trades would have been profitable, and sometimes they would have been losers. Only the data can tell which.

The stranger test is a good standard for any strategy. If two people given the same rules and the same chart would take different trades, the rules aren't yet precise enough. A practical way to apply it is to give your written rules to a trading friend and ask them to mark the signals on a chart without discussing it. Comparing their marks with yours reveals where the rules are still ambiguous. Each disagreement points to a term that needs a clearer definition. A few rounds of this usually produces a much tighter set of rules.

The point about following rules the same way on a bad day is perhaps the most important. Vague rules allow the trader to interpret them differently depending on mood, recent results, or what they want to see. Precise rules leave less room for that. After a run of losses, a trader with vague rules might see support where there isn't any, or skip a valid setup out of fear. Has anyone here rewritten a strategy into precise rules like this? I'd be interested in how many conditions you ended up with and whether your results became more consistent.