Risk Management

The fundamental rule of professional risk management: never risk more than 1–2% of total capital in a single trade. This is not a conservative suggestion — it is the mathematical boundary that makes the difference between a recoverable drawdown and ruin. Why this rule? It is about the survival of the account across sequences of losses, which are inevitable. With 2% risk per trade: • 10 consecutive losing trades = −20% of account (recoverable — you need +25% to return to break-even). • 20 consecutive losing trades = −40% of account (difficult but possible — you need +67%). With 10% risk per trade: • 10 consecutive losing trades = −100% of account (GAME OVER, mathematically — actually you hit margin call far earlier). The math of recovery (asymmetric and brutal): • Lose 20% → you need +25% to break even. • Lose 50% → you need +100% to break even. • Lose 80% → you need +400% to break even. • Lose 90% → you need +900% to break even. The deeper the hole, the steeper the climb. This is why avoiding large losses is mathematically more important than seeking large gains — the compounding of losses is asymmetric. The Kelly criterion connection: the optimal fraction to bet is proportional to your edge divided by the odds. For a system with a realistic 40–50% win rate and 1:2 R/R, the full-Kelly fraction is around 5–8% — but professional desks use a fraction of Kelly (quarter-Kelly, ~1.5–2%) to drastically reduce volatility and drawdown depth while keeping the geometric growth positive. The 1–2% rule IS quarter-Kelly for a typical retail edge. It is not arbitrary; it is the mathematically justified size for surviving variance while compounding a real edge. The psychological layer: small risk per trade also removes the emotional spike of each outcome. At 1%, a loss is a rounding error in the day; at 10%, a loss is a crisis. You cannot trade well in a crisis state — the cortisol destroys decision quality. Risk size is, in part, a tool for keeping yourself in a cognitive state where you can keep executing your plan.