Chapter 18: Advanced Risk Management & Portfolio Construction

Turn single-trade sizing into a robust portfolio that survives drawdowns, scales predictably, and protects your edge.

This chapter moves you from “size one trade” (RFPP) to “size a trading business.” You’ll learn how to allocate risk across multiple strategies, control portfolio-level drawdown, use volatility-aware sizing, account for correlations, and build scaling rules that protect capital when you grow. Includes practical templates, step-by-step math, and an audit checklist you can use today.

Why portfolio-level risk matters

Trading performance is not just “how good each trade is” — it’s how those trades interact. Multiple strategies that look profitable in isolation can produce bad portfolio outcomes when they’re correlated (all lose in the same market shock).

Portfolio risk management:

• Keeps maximum drawdown within survivable limits.

• Makes scaling predictable (you know when to increase/decrease risk).

• Reduces tail risk through diversification and volatility-aware sizing.

• Turns a collection of strategies into a repeatable business.

Core concepts (short definitions)

• Risk budget: the % of account equity you allocate to each strategy.

• Volatility parity: size strategies so each contributes a similar dollar volatility (not equal capital).

• Correlation: how strategies move together (−1 to +1). Low/negative correlation helps reduce portfolio variance.

• Drawdown gate: hard thresholds that pause scaling or force review.

• Risk-of-ruin: the probability of blowing up your account given your edge and sizing — keep it tiny.

Step 1 — Set your portfolio risk profile (decide limits)

Before any allocation, pick these non-negotiables:

1. Max acceptable drawdown (MA D) — the largest % drop you will tolerate before changing business tactics. Typical: 10–25% depending on appetite.

2. Target annual return (TAR) — realistic goal for the year (e.g., 12%).

3. Risk budget (RB) — % of equity you’re willing to risk per month across all strategies (example: 3% monthly).

4. RFPP single-trade limit — keep single-trade risk small relative to RB (example: if RB = 3%/month and you expect 30 trades/month, single risk ≤ 0.1%–0.25%).

Write these down. These anchors guide allocation and scaling.

Step 2 — Risk-budget the portfolio (practical template)

Allocate your total risk budget across strategy buckets, not by equity.

Example plan (you can copy this layout):

• Account equity: $50,000

• Max monthly risk budget (RB): 3% of equity

• RB in dollars = 0.03 × $50,000 = compute digit-by-digit:

  • 50,000 × 0.03 = 50,000 × (3 ÷ 100)

  • 50,000 × 3 = 150,000

  • 150,000 ÷ 100 = 1,500 → $1,500 monthly risk budget

Suppose three strategy buckets:

• Scalping (high-frequency): 30% of RB → 0.30 × $1,500 = compute: 1,500 × 0.30 = 1,500 × (3 ÷ 10) = (1,500 × 3) ÷ 10 = 4,500 ÷ 10 = $450

• Swing (trend/swing): 50% of RB → 0.50 × $1,500 = 1,500 × 0.5 = $750

• Carry / Long-term: 20% of RB → 0.20 × $1,500 = 1,500 × 0.2 = $300

Now you have dollar risk limits per strategy per month: Scalping $450, Swing $750, Carry $300.

Why risk-dollar first? It prevents size creep when one strategy wins and you double down without rebalancing.

Step 3 — Convert strategy risk budget to per-trade sizing (RFPP applied)

Take the Scalping bucket as example.

Assumptions:

• Scalping expects ~90 trades/month.

• Bucket monthly risk = $450.

• Max risk per trade = Bucket / expected trades = 450 ÷ 90 = step-by-step:

  • 450 ÷ 90 = divide numerator and denominator by 90; 90 × 5 = 450, so result = 5 → $5 risk per trade.

If your RFPP single-trade % target is 0.1% of account ($50,000 × 0.001 = $50), you see $5 is far smaller than 0.1% — which is fine for scalping which uses many trades.

For swing:

• Swing bucket $750, expected trades 15/month → 750 ÷ 15 = step-by-step:

  • 750 ÷ 15 = (75 ÷ 15) × 10 = 5 × 10 = $50 per trade.

That $50 per trade equals 0.1% of the $50,000 account (good alignment with RFPP).

Rule: ensure single-trade risk per bucket respects your global RFPP ceiling.

Step 4 — Volatility parity sizing (balance risk contribution)

Equal dollar risk budget is useful, but strategies differ in volatility. Volatility parity sizes each strategy so its dollar volatility (σ × weight) is equalized.

How to do volatility parity (two-step):

1. Estimate each strategy’s annualized volatility (σ) from historical returns or typical P/L swings.

2. Compute weights inversely proportional to σ.

   
   

Example (hypothetical vol numbers annualized):

• Scalping σ = 12% (0.12)

• Swing σ = 25% (0.25)

• Carry σ = 8% (0.08)

Compute inverse vol:

• 1/σ_scalp = 1 ÷ 0.12 = step-by-step:

  • 1 ÷ 0.12 = 1 ÷ (12 ÷ 100) = (1 × 100) ÷ 12 = 100 ÷ 12 = 8.333333... → 8.3333333

• 1/σ_swing = 1 ÷ 0.25 = 4.0 → 4.0

• 1/σ_carry = 1 ÷ 0.08 = 12.5 → 12.5

Sum = 8.3333333 + 4.0 + 12.5 = compute:

• 8.3333333 + 4.0 = 12.3333333

• 12.3333333 + 12.5 = 24.8333333 → 24.8333333

Weights = (1/σ_i) ÷ sum:

• w_scalp = 8.3333333 ÷ 24.8333333 = step-by-step:

  • 8.3333333 ÷ 24.8333333 ≈ 0.3356 → 33.56%

• w_swing = 4.0 ÷ 24.8333333 ≈ 0.1610 → 16.10%

• w_carry = 12.5 ÷ 24.8333333 ≈ 0.5034 → 50.34%

Interpretation: under volatility parity, Carry gets ~50% risk weight because it’s least volatile; Swing gets least weight.

Now apply to dollar risk budget ($1,500):

• Carry: 0.5034 × 1,500 = compute: 1,500 × 0.5034 = (1,500 × 5034) ÷ 10,000? Simpler: 1,500 × 0.5 = 750; 1,500 × 0.0034 = 5.1; sum = 750 + 5.1 = $755.10 (round to $755)

• Scalping: 0.3356 × 1,500 = 1,500 × 0.3356 = 1,500 × 0.3 = 450; remaining 0.0356 × 1,500 = 53.4; sum = 450 + 53.4 = $503.40 (round $503)

• Swing: 0.1610 × 1,500 = 1,500 × 0.1610 = 1,500 × 0.16 = 240; plus 0.001 × 1,500 = 1.5; sum = 241.5 → $242.

Compare with equal-risk split earlier (Scalp $450, Swing $750, Carry $300). Volatility parity shifts more budget to Carry because it’s less volatile — this reduces portfolio variance.

Use case: volatility parity avoids overweighting high-volatility strategies that would blow up the portfolio more often.

Step 5 — Account for correlations (real portfolio variance)

Portfolio variance for two strategies (simplified) is:

σp2=w12σ12+w22σ22+2w1w2σ1σ2ρ12\sigma_p^2 = w_1^2\sigma_1^2 + w_2^2\sigma_2^2 + 2 w_1 w_2 \sigma_1 \sigma_2 \rho_{12}σp2​=w12​σ12​+w22​σ22​+2w1​w2​σ1​σ2​ρ12​

Let’s compute a simple two-strategy example (numbers illustrative):

• Strategy A (Swing): weight w1 = 0.6, σ1 = 0.25

• Strategy B (Scalp): weight w2 = 0.4, σ2 = 0.12

• Correlation ρ = 0.2

Compute each term step-by-step:

1. w1^2 σ1^2 = (0.6^2) × (0.25^2)

   • 0.6^2 = 0.36

   • 0.25^2 = 0.0625

   • Multiply: 0.36 × 0.0625 = step-by-step:

     • 0.36 × 0.06 = 0.0216

     • 0.36 × 0.0025 = 0.0009

     • Sum = 0.0216 + 0.0009 = 0.0225

2. w2^2 σ2^2 = (0.4^2) × (0.12^2)

   • 0.4^2 = 0.16

   • 0.12^2 = 0.0144

   • Multiply: 0.16 × 0.0144 = step-by-step:

     • 0.16 × 0.01 = 0.0016

     • 0.16 × 0.0044 = 0.000704

     • Sum = 0.0016 + 0.000704 = 0.002304

3. 2 w1 w2 σ1 σ2 ρ

   • 2 × 0.6 × 0.4 × 0.25 × 0.12 × 0.2

   • Compute step-by-step:

     • 2 × 0.6 = 1.2

     • 1.2 × 0.4 = 0.48

     • 0.48 × 0.25 = 0.12

     • 0.12 × 0.12 = 0.0144

     • 0.0144 × 0.2 = 0.00288 → 0.00288

Now σ_p^2 = 0.0225 + 0.002304 + 0.00288 = compute:

• 0.0225 + 0.002304 = 0.024804

• 0.024804 + 0.00288 = 0.027684 → σ_p^2 = 0.027684

Portfolio σ_p = sqrt(0.027684) = approx 0.1664 → 16.64% annualized volatility

If the correlation were higher (ρ = 0.8), the covariance term becomes larger and σ_p increases — that’s why low correlation is valuable.

Takeaway: always estimate pairwise correlations between strategy returns. If two strategies are highly correlated, treat them as one bucket for risk budgeting.

Step 6 — Drawdown gates & risk-of-ruin control

Set hard stop levels at portfolio and strategy levels:

• Soft gate: e.g., if portfolio drawdown > 60% of MA D (e.g., MA D = 20% → soft gate 12%), reduce new risk allocations by 50% and perform review.

• Hard gate: if drawdown > MA D (20%), stop scaling, reduce live risk to 50%, and move to demo for failing strategies.

Risk-of-ruin: Keep risk per trade and monthly risk such that the probability of account destruction is tiny. Simple rule of thumb: keep worst-case number of consecutive losses (N) manageable. If your strategy’s historical max consecutive losing streak is 10 trades, and you risk 2% each, odds of ruin are high. Lower risk per trade or increase diversification.

Step 7 — Hedging & overlays (practical FX hedging)

Hedging should be tactical, not comfy.

• Pair hedge: if you hold long EUR/USD and short EUR/GBP to neutralize EUR directional risk, you’ve reduced currency exposure but increased cross-currency exposure. Use hedges when macro tail risk is high (central bank events).

• Correlation hedge: use an instrument that historically moves negatively to your basket (e.g., gold vs risk-on FX pairs) as an overlay rather than per-trade hedges.

• Cost-aware: hedges cost spreads/commissions; use them only when the reduction in portfolio VaR justifies the cost.

Example: if you plan a safe-mode hedge during a central bank week, size the hedge to reduce net delta by X% for Y days, then unwind.

Step 8 — Scaling rules (how to grow safely)

Scaling must be rule-based.

1. Performance gates: only increase risk after N consecutive months meeting targets. Example gate: 3 months with ≥ target return and max drawdown < 50% of MA D.

2. Step increases: when allowed, increase risk budget by a small fraction (e.g., 10–20% of current RB), not by doubling.

3. Liquidity check: increase size only if average trade size × new lot size still fits liquidity (ESP for exotics).

4. Execution check: re-verify slippage at larger sizes with 10–20 incremental test trades.

5. Stop-loss reuse: whenever risk increases, keep stop methodology identical; only increase lot size — never reduce stop precision.

Example scaling sequence:

• Month 0: RB = 3%

• After 3 months passing gates → RB += 20% → RB = 3% × 1.2 = compute: 3 × 1.2 = 3.6 → 3.6% total risk budget.

Step 9 — Operational controls & audit checklist

A regular audit turns discipline into habit.

Monthly audit checklist

•  Confirm entity & account used for live funds.

•  Export trade history and compute: expectancy, profit factor, max drawdown.

•  Update volatility inputs (σ) for each strategy (last 12-month window).

•  Recompute correlations between strategies (last 6–12 months).

•  Run portfolio variance formula and update projected MA D given current RB.

•  Confirm all live sizes match risk-budgeted sizes (no manual overrides >5%).

•  Verify withdrawal process and cash buffer (at least 1 month's RB in cash).

•  Confirm data logging & backups (trade logs, chat logs with brokers).

Quarterly governance

• Stress test worst 1-in-100 scenario (historical or synthetic shock).

• Recalculate risk-of-ruin for current sizing and adjust RB if needed.

• Review broker relationships and update contingency plan (backup brokers, withdraw methods).

Step 10 — Practical templates & what to implement now

Implement these three simple templates in a spreadsheet — manually or with your trading journal software:

1. Risk Budget Table

   • Columns: Strategy | Annualized Vol (σ) | Alloc % (vol parity) | Dollar RB | Expected Trades/month | $ risk / trade | RFPP cap check (Y/N)

2. Correlation Matrix (pairwise)

   • Rows/cols: strategies; values: ρ (last 12 months). Use for portfolio variance calc.

3. Scaling Gate Log

   • Date | Metric checked (Expectancy, PF, Drawdown) | Pass/Fail | RB change | Notes

Common mistakes & fixes

• Mistake: budgeting by capital instead of risk → result: oversized positions in volatile strategies.

Fix: budget by $ risk (RB) first, then convert to lots.

• Mistake: ignoring correlation changes during stress (correlations spike). Fix: run stress tests and use conservative correlation estimates during crises (assume ρ→0.6+ for majority of pairs).

• Mistake: scaling after short bursts of luck.

  Fix: require multi-month consistency (≥3 months) and use small incremental increases.

Final rules to add to your trading plan

1. Always risk-budget at the portfolio level before adjusting single-trade sizes.

2. Use volatility parity if strategies have materially different σ.

3. Regularly measure correlations — rebalance risk when correlations rise.

4. Hard stop: pause scaling if drawdown > MA D.

5. Small, consistent scaling steps beat large, emotional increases.

One-page checklist (implement in 30 minutes)

•  Write MA D and RB on a sticky note.

•  Build the Risk Budget Table for your strategies (or tell me to generate the spreadsheet).

•  Compute per-trade risk for each bucket given expected trade count.

•  Run a one-minute correlation check (last 3 months) — flag any ρ > 0.7.

•  Set Monthly Audit calendar reminder.

Portfolio construction turns trading into a business. Sizing by dollar risk, accounting for volatility and correlation, and enforcing hard drawdown gates will prevent catastrophic failure and let you scale with confidence.