Risk Management: The Math Behind Survival
Position sizing, stop losses, and portfolio construction — how professionals think about risk before returns.
The most common reason individual investors underperform isn't bad picks — it's bad position sizing. Too much in high-conviction ideas that turn out wrong, and the drawdown destroys recovery potential.
The Kelly Criterion
f* = (bp - q) / b
Where b = odds, p = win probability, q = loss probability. Most professionals use Half-Kelly or less because your estimate of p is uncertain and full Kelly leads to massive drawdowns.
Practical position sizing
- Core (highest conviction): 4–6% of portfolio, 5–8 positions
- Satellite (moderate): 2–3%, 10–15 positions
- Watch (speculative): 0.5–1%, 5–10 positions
Stop losses: fundamental vs. price-based
Fixed percentage stops (-15%) are crude but effective at limiting tail risk. Better approach: fundamental stops. Sell when the thesis breaks (ROIC drops from 25% to 10%), not when the price drops on a bad market week.
Our pick agent's hybrid system
- Hard stop: Alpha < -15% vs S&P 500
- Soft triggers: Score decay (>10 points), momentum reversal, dead money (90 days, no alpha)
- Classification-aware: Compounder picks get wider stops than momentum picks
Correlation matters
20 tech stocks isn't diversified — it's concentrated. True diversification: 3–4 sectors, mixed factor exposures, some defensive positions, and ETFs for international/fixed income.
Remember: a 50% loss requires a 100% gain to recover. Risk management isn't optional — it's the foundation everything else is built on.