Risk Scaling by Setup Quality
Allocating more risk to A+ setups and less to B setups — matching position size to conviction level
Risk scaling means varying your position size based on the quality grade of each setup. Instead of risking the same percentage on every trade, you allocate more to high-conviction setups and less to marginal ones. A typical risk scaling model: A+ setups = 1.0% account risk, A setups = 0.75%, B setups = 0.5%. The logic: A+ setups have the highest win rate and best R:R, so they deserve the most capital exposure. B setups have lower probability and worse R:R — they still have positive expectancy but deserve less exposure. Over 100+ trades, risk scaling produces two benefits: (1) Your winners are disproportionately large because A+ setups win more often AND are sized bigger. (2) Your losers are disproportionately small because B setups lose more often AND are sized smaller. The net effect is a smoother equity curve with shallower drawdowns, even if the total P&L is similar to flat-risk trading. Risk scaling is only meaningful after you have a graded dataset — without data, the grades are subjective and the scaling loses its statistical basis.
✓How to Recognize
- •A+ = 1.0% risk, A = 0.75% risk, B = 0.5% risk (adjust percentages to your account)
- •Winners become disproportionately large (high-grade setups win more AND are sized bigger)
- •Losers become disproportionately small (low-grade setups lose more AND are sized smaller)
- •Requires a graded dataset first — without data, the grades are guesses
⚡How to Avoid
- →Risk scaling without a dataset (you need 50+ graded entries before the grades are meaningful)
- →Sizing up on B setups because "this one feels different" (the system only works if you follow it)
- →Making the risk difference too extreme (2% vs 0.25% creates wild equity swings)
- →Changing the risk model during a drawdown (the model works over 100+ trades, not 10)