← Back to Learn
I BeginnerWeek 1 • Lesson 2Duration: 40 min

RSK Risk Management Fundamentals

You can't compound returns from a blown account

Learning Objectives

  • •Understand why risk management matters more than signal quality
  • •Use the R-multiple framework to normalize trade outcomes
  • •See the math behind why drawdowns are so hard to recover from
  • •Learn how adaptive risk scaling protects capital in real time

Explain Like I'm 5

Imagine a jar with 100 marbles — 59 green (winners) and 41 red (losers). If you bet your entire jar on each draw, one red marble wipes you out. But if you bet just 1 marble at a time, you steadily grow your collection. Risk management is about how many marbles you bet each round — small enough to survive bad streaks, large enough for wins to add up.

Think of It This Way

Risk management is the seatbelt for your portfolio. It doesn't make the car faster, but it keeps you alive when you crash. Every trader hits drawdowns. The difference between those who survive and those who blow up is their risk infrastructure.

1Why This Comes Before Everything Else

Here's something most people get backwards: a strategy with 55% win rate and bad risk management will blow up faster than a strategy with 45% win rate and excellent risk management. The reason is simple math. Losses and recovery aren't symmetric: • Lose 10% → need 11.1% to get back • Lose 25% → need 33.3% • Lose 50% → need 100% • Lose 75% → need 300% This gets ugly fast. A 50% drawdown — which plenty of retail traders experience — means you need to double your remaining capital just to break even. The math is clear: keeping your capital alive during bad stretches is what makes long-term compounding possible. Prop firms like FTMO cap drawdowns at 5% daily and 10% total. They've seen thousands of traders. They know that survival equals opportunity. Reference: Hull, J.C. "Risk Management and Financial Institutions." Wiley, 4th Edition.

The Asymmetry of Drawdown Recovery

2The R-Multiple Framework

R-multiples give you a universal ruler for trade outcomes, regardless of instrument, position size, or account value. 1R = the initial risk on a trade. It's the dollar distance between your entry and stop-loss, times your position size. Some examples: • Stop loss is 50 pips, risking 100→1R=100 → 1R =100 • Trade makes $200 → result = +2.0R • Trade hits stop → result = −1.0R • Exit early at $50 loss → result = −0.5R Why this matters: normalization. A +2R trade on EUR/USD is directly comparable to a +2R trade on gold or the S&P 500. You can analyze performance across instruments without converting everything to dollars. When a system reports +533.9R over 4,505 trades, that metric scales linearly with account size. Whether the account is 10Kor10K or1M, the R-multiple sequence is identical — only the dollar value of 1R changes. Reference: Tharp, V.K. "Trade Your Way to Financial Freedom." McGraw-Hill.

3Adaptive Risk Scaling

Smart risk management doesn't use a fixed position size. It adapts based on how things are going. The logic is straightforward: as losses pile up and drawdown deepens, reduce size to protect what's left. As equity recovers, gradually restore full sizing. A well-designed protocol works across zones: • Normal (0–4% DD): Full sizing — everything is within expected range. • Caution (4–6% DD): Reduced sizing — drawdown is elevated, capital preservation comes first. • Danger (6–8% DD): Minimum sizing — only the best signals are worth taking. • Critical (8–10% DD): Emergency sizing — survival is the only goal. This is counterintuitive. Most traders increase risk after losses, trying to make it back fast. That's revenge trading, and the data is clear: it accelerates ruin. Scaling down during drawdowns and up during good runs is the mathematically correct approach. Monte Carlo simulations with 20,000+ iterations confirm that properly calibrated adaptive risk keeps breach probability well below 5% even under worst-case scenarios.

Adaptive Risk Scaling vs Fixed Risk — Survival Analysis

Key Formulas

Risk of Ruin (Simplified)

p is edge probability, B is bankroll in risk units, R is risk per trade. As B/R grows (smaller risk per trade), ruin probability drops fast. This is why risking 0.30% of capital per trade makes ruin effectively impossible.

Drawdown Recovery Requirement

A 50% drawdown needs 100% gain to recover. A 10% drawdown only needs 11.1%. This asymmetry is why max drawdown control matters more than almost anything else in risk management.

Hands-On Code

Monte Carlo Drawdown Simulation

python
import numpy as np

def simulate_max_drawdown(win_rate: float, avg_win: float,
                           avg_loss: float, risk_pct: float,
                           n_trades: int, n_sims: int = 10000):
    """
    Monte Carlo simulation of maximum drawdown distribution.
    
    Parameters:
        win_rate: probability of winning trade
        avg_win: average winning trade in R-multiples
        avg_loss: average losing trade in R-multiples
        risk_pct: fraction of equity risked per trade
        n_trades: number of trades to simulate
        n_sims: number of Monte Carlo iterations
    """
    max_dds = np.zeros(n_sims)
    
    for i in range(n_sims):
        equity = 100.0
        peak = 100.0
        max_dd = 0.0
        
        for _ in range(n_trades):
            if np.random.random() < win_rate:
                equity += equity * risk_pct * avg_win
            else:
                equity -= equity * risk_pct * avg_loss
            
            peak = max(peak, equity)
            dd = (peak - equity) / peak
            max_dd = max(max_dd, dd)
        
        max_dds[i] = max_dd * 100  # percentage
    
    return {
        'median': np.median(max_dds),
        'p95': np.percentile(max_dds, 95),
        'p99': np.percentile(max_dds, 99),
        'breach_prob': np.mean(max_dds > 10) * 100,
    }

# Conservative risk parameters
result = simulate_max_drawdown(
    win_rate=0.592, avg_win=1.65, avg_loss=0.95,
    risk_pct=0.003, n_trades=4500
)

print(f"Median Max DD:     {result['median']:.2f}%")
print(f"95th Percentile:   {result['p95']:.2f}%")
print(f"99th Percentile:   {result['p99']:.2f}%")
print(f"10% Breach Prob:   {result['breach_prob']:.2f}%")

This runs thousands of possible equity paths and finds the worst drawdown in each. By checking the 95th and 99th percentiles, you can see whether a given risk level keeps drawdown within your limits. At 0.30% risk per trade, breach probability typically lands well below 1%.

Knowledge Check

Q1.If your account drops by 50%, how much must you gain to recover to the original balance?

Q2.What does "1R" represent in the R-multiple framework?

Q3.What is the correct response when a trading system enters a deeper drawdown?

Assignment

Using the Monte Carlo simulation code, test risk_pct values of 0.5%, 1.0%, 2.0%, and 3.0%. At what risk level does the 99th percentile max drawdown first cross 10%? Build a table of risk level vs. drawdown percentiles vs. breach probability. Write 300 words explaining why smaller sizes produce much better survival rates.

← Previous LessonNext Lesson →