From Idea to Strategy: How to Describe a Trade Setup in Plain English
A strategy starts as a sentence. "I want to buy oversold tech stocks when the broader market is healthy." If you can say that out loud and have someone else understand you, you are 80% of the way to a testable rule. The hard part is being precise about what each word means.
Strategies Are Just Rules and Conditions
Every strategy, no matter how complex, breaks down into the same four pieces: an asset universe, an entry condition, an exit condition, and a risk envelope. If you can fill in those four blanks, you have a strategy. If you cannot, you have a feeling.
The 4-Part Template
- Universe — which symbols this strategy is allowed to trade
- Entry — what has to be true before you buy
- Exit — what has to be true before you sell, including profit-take and stop-loss
- Risk — how much per trade, how many concurrent positions, when to stop trading
Notice that "prediction" is not on the list. You are not predicting the future. You are defining a rule and letting the data tell you whether it works.
Examples: A Stock Setup
Universe: AAPL, MSFT, NVDA, GOOGL, META.
Entry: 14-period RSI under 30 and SPY trading above its 50-day moving average.
Exit: RSI above 60, or a 5% trailing stop, whichever fires first.
Risk: 1% of equity per position, four concurrent positions max, regular trading hours only.
And a Crypto Setup
Universe: BTC-USD, ETH-USD, SOL-USD.
Entry: price at or below 95% of the 7-day high, with volume above its 20-period average.
Exit: take 50% off at +8%, trail the rest with a 4% stop.
Risk: 1.5% of equity per position, three concurrent positions max, 24/7.
What Vague Language Hides
"Buy when it looks oversold" is not a rule. It is a wish. "Looks oversold" hides at least three decisions: which indicator, which lookback, which threshold. Vague rules backtest beautifully because you, the human, are filling in the gaps after the fact. That is exactly the bias an honest backtest needs to remove.
If you cannot write your rule down without ambiguity, your idea is not finished yet. Going through the work of writing it usually exposes whether the idea was real.
From Sentence to Backtest
Once your four blanks are filled, the rule is testable. Run it, look at the equity curve, see whether it makes sense. If it does not, the answer is almost never to add another condition. The answer is usually that the original idea was not as good as it sounded.
Honest strategies are short. Two or three conditions, applied consistently, beat ten conditions tuned to a specific quarter every time.