Drawdown Management: Controlling Downside Risk

What you'll learn

• What drawdowns are and how to read an "underwater" chart
• How to calculate drawdown, maximum drawdown, and recovery time step-by-step
• Advanced risk measures: Value at Risk (VaR), Conditional VaR/Expected Shortfall (CVaR/ES), Ulcer Index, and MAR ratio
• Professional techniques to control downside: volatility targeting, stop-losses, CPPI, diversification, and option hedges
• How to set a drawdown budget and align it with your investment horizon
• How to combine rules into a practical risk management plan you can monitor

Concept explanation

A drawdown is the percentage decline from a previous peak in your portfolio value. Imagine climbing a hill: each new highest point you reach is a "peak." If you walk down from that peak, the drop in altitude before you reach a new peak is your "drawdown." In investing, the altitude is your account value. Drawdowns feel painful because they represent losses from money you actually had.

Maximum drawdown (often abbreviated MaxDD) is the worst peak-to-trough decline over a period. It answers, "How bad did it get?" An "underwater chart" shows your drawdown over time as a curve under the zero line; deeper and longer dips mean tougher stretches for both your capital and your emotions.

Drawdowns matter for two reasons. First, they determine survival: the deeper the hole, the harder to climb out. A 50% loss needs a 100% gain to recover. Second, drawdowns shape your behavior. Even profitable strategies can fail if the drawdowns are too big for you to stick with.

Managing drawdowns is not just about picking less volatile assets; it's about designing rules that limit losses, adapt position sizes to changing risk, and diversify across independent sources of return so that no single slump sinks the ship.

Why it matters

Returns and risk go hand-in-hand. Many investors focus on average returns but underestimate the path taken to get them. Two portfolios can have the same average return but very different drawdowns. The one with smaller, more controlled drawdowns is often easier to hold through tough markets and less likely to face forced selling at the worst time.

Professionals measure and manage drawdowns systematically. They use metrics like maximum drawdown, Ulcer Index, Value at Risk (VaR), and Conditional Value at Risk (CVaR, also called Expected Shortfall) to understand potential losses. Then they apply controls such as volatility targeting, stop-losses, Constant Proportion Portfolio Insurance (CPPI), and option-based hedges to keep downside within a predefined "drawdown budget."

For individual investors, adopting even a few of these tools can dramatically improve resilience. Better drawdown control can allow you to stay invested, reduce the odds of panic selling, and protect your long-term compounding.

Calculation method

• Drawdown at time t

  Drawdown_t = (PeakToDate_t - Value_t) / PeakToDate_t

  Where PeakToDate_t is the highest portfolio value observed up to time t.

• Maximum Drawdown

  MaxDD = max over t of Drawdown_t

• Recovery time The number of periods from the trough back to the previous peak value.

• Ulcer Index (penalizes depth and duration of drawdowns)

  UlcerIndex = sqrt( (1/N) * sum_{t=1}^{N} DD_t^2 )

• MAR Ratio (return per unit of maximum drawdown)

  MAR = CAGR / MaxDD

• Historical VaR and CVaR (non-parametric) Suppose you have daily returns over a lookback window.

  VaR_{\alpha} = -\text{Quantile}(\text{returns}, \alpha) CVaR_{\alpha} = -\mathbb{E}[\text{returns} \mid \text{returns} \le \text{Quantile}(\text{returns}, \alpha)]

• Volatility targeting (position sizing by risk)

  Weight = TargetVol / ForecastVol

  Where ForecastVol is an estimate of future volatility (e.g., an exponentially weighted moving standard deviation). Cap weight to avoid leverage beyond your policy.

• CPPI basics

  Cushion = PortfolioValue - Floor RiskAsset = m * Cushion SafeAsset = PortfolioValue - RiskAsset

  Where m is the multiplier (e.g., 2-5), and Floor is the capital you want to protect (often linked to a time horizon or liability).

• Kelly fraction (upper bound for risky sizing, use conservatively) For a simple bet with probability p of winning b units and probability q = 1 - p of losing 1 unit:

  f^* = (b p - q) / b

  In markets with mean excess return μ and variance σ^2, a rough approximation:

  f^* \approx \mu / \sigma^2

  Most investors use a fraction of Kelly (e.g., 0.25×) to limit drawdowns.

Step-by-step example of drawdown calculation:

1. Record daily portfolio values: e.g., 100, 103, 101, 98, 99, 105
2. Compute peak-to-date: 100, 103, 103, 103, 103, 105
3. Drawdown %: (100-100)/100=0%; (103-103)/103=0%; (103-101)/103≈1.94%; (103-98)/103≈4.85%; (103-99)/103≈3.88%; (105-105)/105=0%
4. MaxDD is the largest drawdown: ≈4.85%
5. Recovery time: from the trough at 98 back to the next time value reaches 103; here, recovery occurs at 105 on day 6.

Case study

Assume a $100,000 portfolio invested 60$ thousands) are:

Month 0: 100; 1: 102; 2: 99; 3: 95; 4: 96; 5: 98; 6: 101; 7: 104; 8: 100; 9: 97; 10: 99; 11: 103; 12: 106

1. Peak-to-date and drawdowns

• Peaks: 100, 102, 102, 102, 102, 102, 102, 104, 104, 104, 104, 104, 106
• Drawdowns: 0%, 0%, (102-99)/102≈2.94%, (102-95)/102≈6.86%, (102-96)/102≈5.88%, (102-98)/102≈3.92%, (102-101)/102≈0.98%, 0%, (104-100)/104≈3.85%, (104-97)/104≈6.73%, (104-99)/104≈4.81%, 0%, 0%
• MaxDD ≈ 6.86% at Month 3

2. MAR ratio Assume the 12-month CAGR is: 100 to 106 → CAGR ≈ 6%. MAR ≈ 0.06 / 0.0686 ≈ 0.875

3. VaR and CVaR (historical, monthly) Suppose monthly returns sorted (in %) are: -5.0, -3.8, -3.5, -1.9, -0.5, 0.0, 0.9, 1.0, 1.9, 2.0, 2.9, 3.1.

• 10% VaR is the 10th percentile ≈ -3.8%
• 10% CVaR is the average of worst 10% tail. With 12 points, approximate by averaging the worst two: (-5.0 + -3.8)/2 ≈ -4.4%

4. Volatility targeting Assume forecast monthly volatility of the equity sleeve is 6% and your target portfolio volatility is 2%. If equities are the main driver, scale equity weight by 2%/6%≈0.33. Your 60% equities becomes ~20% temporarily, with 80% in bonds/cash, until volatility normalizes.

5. CPPI Set a Floor at $90,000 and m=3. At Month 3 (Portfolio ≈$95,000), Cushion = 95k - 90k = 5k; RiskAsset = 3 × 5k = 15k; SafeAsset = 95k - 15k = 80k. CPPI automatically de-risks after losses to protect the Floor, and re-risks as the Cushion grows.

Practical applications

• Set a drawdown budget Decide a tolerable MaxDD (e.g., 15%). Use historical backtests and stress tests to check if your allocation keeps MaxDD within that budget. If not, reduce risky exposure, add hedges, or diversify further.

• Volatility targeting for smoother paths Scale risky asset weights by realized or forecast volatility to keep total risk near a target. Add guardrails: caps on leverage and minimum/maximum allocations.

• Stop-loss and trailing-stop rules For individual positions, use closing-price stops or portfolio-level loss triggers. Example: sell or hedge if a position falls 20% from its peak or if the portfolio drawdown exceeds 10%. Combine with cooldown periods to avoid rapid whipsaws.

• Diversification across independent risks Blend assets and strategies whose returns are not tightly correlated: global equities, high-quality bonds, commodities, trend-following, and defensive equity factors. Diversification reduces the probability that everything falls at once.

• Option-based hedging Protective puts, put spreads, or collars can limit downside at known costs. For example, a 6-month 10% out-of-the-money put can cap extreme drawdowns; collars reduce cost by selling a call.

• CPPI or floor-based strategies If you have capital you must protect (e.g., near-term college tuition), set a Floor and apply a multiplier to scale risky assets dynamically.

• Stress testing and scenario analysis Examine portfolio performance under historical crises (e.g., 2008, March 2020) and hypothetical shocks (e.g., -20% equity, +150 bps rates). Check whether losses breach your drawdown budget and adjust.

• Use CVaR alongside MaxDD MaxDD is one event; CVaR summarizes the average of bad outcomes. Target CVaR to manage tail risk and duration of pain.

Common misconceptions

Summary

Glossary

• Drawdown: The percentage decline from a previous peak in portfolio value.
• Maximum Drawdown (MaxDD): The largest peak-to-trough loss over a measurement period.
• Underwater Chart: A graph of drawdown over time, shown as percentages below zero.
• Ulcer Index: A metric that averages the squared drawdowns to penalize depth and duration.
• MAR Ratio: Compound annual growth rate divided by maximum drawdown.
• VaR (Value at Risk): A threshold loss level not expected to be exceeded with a given probability over a set horizon.
• CVaR/Expected Shortfall: The average loss given that losses have exceeded the VaR level.
• Volatility Targeting: Adjusting position sizes so portfolio volatility stays near a chosen target.
• CPPI: A dynamic allocation method that keeps a floor value protected and scales risk with a multiplier.
• Protective Put/Collar: Option strategies to cap downside (and, for collars, potentially cap upside to reduce cost).

Glossary

Drawdown: Loss from a prior peak in portfolio value, expressed as a percentage.

Maximum Drawdown: The worst peak-to-trough decline experienced over a period.

Ulcer Index: Risk metric that averages squared drawdowns, penalizing depth and duration.

MAR Ratio: CAGR divided by maximum drawdown, measuring return per unit of drawdown risk.

Value at Risk (VaR): Loss threshold that will not be exceeded with a specified probability over a set horizon.

Conditional VaR (CVaR): Average loss in the worst tail beyond the VaR threshold.

Volatility Targeting: Position sizing method aiming to keep portfolio volatility near a target.

CPPI: Constant Proportion Portfolio Insurance; dynamic allocation to protect a floor value.

Protective Put: Option giving the right to sell at a strike price, limiting downside.

Collar: Option strategy combining a protective put with a covered call to reduce hedge cost.