Stack, don't scatter
No single indicator is an edge. A robust setup stacks a few independent reads that agree: a trend filter (e.g., price above the 200-day), a location (a support level or breakout), and a trigger (a candle or momentum cross). Confluence beats any one signal.
Write the setup down as rules before you trade it — entry, the exact invalidation (stop), and the target. If you can't state where you're wrong, you don't have a setup, you have a hope.
Plan the exit first
Define the stop from structure or ATR (Module 5.2), then the target, then check the reward:risk. A setup risking ₹10 to make ₹30 (3:1) can be wrong more than half the time and still profit. If the reward:risk is below ~1.5:1, most traders skip it.
A setup checklist
A tradeable setup answers all of these before you click:
- Trend filter — am I with the higher-timeframe trend (e.g. above a rising 200-day)?
- Location — is price at a level that matters (support, breakout, MA)?
- Trigger — what specific event gets me in (a candle, a break, a momentum cross)?
- Invalidation — exactly where am I wrong (the stop)?
- Target and reward:risk — is the payoff worth the risk (≥ ~1.5:1)?
Think in expectancy
Win-rate alone is a vanity metric. Expectancy = (win% × average win) − (loss% × average loss). A setup that wins just 45% of the time but makes 3× what it risks is strongly profitable: 0.45×3 − 0.55×1 = +0.8R per trade. Your edge lives in cutting losers fast and letting winners pay — the entry is the least important part.