Advanced Risk Frameworks for Equity Books

Expert Track capstone (instruments → book risk). Builds on Risk 101, Kelly sizing, ATR sizing, correlation & portfolio risk, and advanced instruments in options, index futures, and leveraged ETFs. Hub: stock courses.

Single-Trade Risk Is Not Book Risk

Most traders eventually learn to size one stock idea: risk 0.5–1% of equity to a stop, respect ATR, avoid all-in bets. That is necessary and still insufficient. Accounts die from portfolio geometry — five “independent” 1% risks that are actually one tech-factor bet, options + shares + a 3× ETF stacked on the same thesis, a short book that gaps on a squeeze, or a month of winners followed by a single unhedged event that erases a year. Professional equity desks do not ask only “how much on this ticker?” They ask “what is the distribution of outcomes for the whole book, under stress, with kill switches that fire without negotiation?”

This course is an operator’s map of advanced equity risk: VaR and expected shortfall as intuition (not a PhD derivation), historical and hypothetical stress tests, risk budgets by sleeve and factor, drawdown protocols, hard kill switches, concentration and liquidity constraints, and how to integrate instruments you already met in Track 5 without accidental leverage. Use the risk calculator, Kelly calculator, and win rate calculator for line-level discipline; this course governs the system above the line.

1. From Risk 101 to a Risk Framework

Risk 101 teaches loss caps, R-multiples, and not risking ruin on one idea. Kelly teaches that edge and variance jointly determine growth-optimal size — and that full Kelly is usually too aggressive. ATR sizing links stop distance to volatility. Correlation course teaches that diversification is a claim, not a default. A framework ties those tools into standing rules:

• Limits: max loss per trade, per day, per week, per month; max gross/net exposure; max single-name and sector weight.
• Measurement: how you estimate loss under normal and stress conditions (simple scenario P&L, VaR-style summaries, factor proxies).
• Governance: who can override a limit (ideally: no one mid-drawdown, including you).
• Response: pre-committed actions at each breach level (reduce 25%, flatten directional, halt new risk).

Without governance, measurement is theater. Without measurement, limits are vibes. Without response, both are diary entries after the blowup. Broker house rules and margin (post-June 2026: no classic $25k PDT floor as current law — still real intraday margin and house BP; see broker course) are external constraints; your framework should be stricter than the broker’s maximum leverage.

2. VaR Intuition (Without Worshipping the Number)

Value at Risk (VaR) answers a question of the form: “Over horizon H, at confidence level α, what loss amount L is such that the probability of losing more than L is about 1−α?” Example language: “1-day 95% VaR of $8,000” means that under the model, on about 95% of days the loss should be smaller than $8,000 — and on about 5% of days it can be worse (how much worse is not VaR’s job).

Common estimation families (conceptual):

• Historical simulation: reprice today’s book through past return days; read the loss quantile.
• Parametric / variance-covariance: assume returns roughly elliptical; VaR scales with volatility and positions (crude for options and gaps).
• Monte Carlo: simulate paths under a model; useful when nonlinear instruments dominate.

What VaR is good for: a single comparable number across days; a dashboard signal that risk rose when you added names or vol expanded; communication (“book risk doubled this week”).

What VaR is terrible for: telling you the size of the left tail on the bad 5% days; assuming correlations stay stable in crises (they do not — Course 37); treating a calm sample period as destiny; applying linear VaR blindly to short options, leveraged ETFs with path dependency, or binary event risk (earnings, FDA, gap opens from gap trading).

Worked intuition example. Equity $200,000. Long-only book of liquid large-caps, recent 1-day portfolio vol estimate 1.0% of equity. A crude normal-ish 95% 1-day VaR scale is on the order of 1.65 × σ × equity ≈ 1.65 × 0.01 × 200,000 ≈ $3,300. If you then add a concentrated small-cap biotech at 15% of equity ahead of binary data, realized gaps can be multiples of that “VaR” in a single open. The number did not “fail math” — you asked a quiet-history model a violent-event question. Always pair VaR with event inventory and stress.

3. Expected Shortfall (CVaR) — Caring About How Bad “Bad” Is

Expected shortfall (ES), also called CVaR in many practitioner settings, asks: given that we are already in the loss tail beyond VaR, what is the average loss in that tail? ES is more sensitive to extreme outcomes than VaR alone. If two books share the same 95% VaR but one has a long left tail of rare disasters, ES will usually flag the dangerous book.

For a retail or small prop equity book you may not compute formal ES daily. You can still adopt the mindset: when you write a limit, ask “what is my average loss on the worst 5% of days in a realistic stress sample?” If that answer is account-threatening, size is wrong even if “typical day VaR” looks fine. Short premium structures, naked shorts (short selling), and leveraged products fatten that conditional tail.

4. Stress Testing: Historical, Hypothetical, Reverse

Stress tests reprice the book under explicit shocks. Three useful flavors:

1. Historical: apply returns from a known bad window (e.g. a violent index week, a sector crash, a rate shock period) to today’s holdings and betas.
2. Hypothetical: “SPX −8%, NDX −12%, my top sector −15%, VIX +20 points, credit spreads wider” — then estimate P&L including options convexity if any.
3. Reverse stress: “What combination of moves would lose 10% of equity in a day?” If the answer is a plausible Monday open, you are overexposed.

Worked stress sketch. Book: $150k equity. Gross long $180k (1.2×), short stock $30k, net long $150k. Approximate portfolio beta 1.1 to S&P. Stress: index −6% in one day, shorts squeeze +4% against you on the short names. Rough long-side loss ≈ 180k × 1.1 × 6% is the wrong formula — better: net beta exposure × index move, plus residual. Simpler operator method: map each position’s expected % under the scenario, sum dollars. Suppose longs mark −7% average (−$12.6k on $180k), shorts mark against you −$1.2k, options hedges help +$2k. Net ≈ −$11.8k ≈ −7.9% equity. If your monthly kill zone is −8%, one stress day is already a framework event — reduce gross before the event, not after. Translate scenarios with the percentage change calculator and book them like trades with the P&L calculator.

Index futures hedges (Course 48) change stress results; so do put hedges and collars from options courses. Always re-run stress after adding a “hedge” that is actually a correlated long in disguise (e.g. long stock + long 3× bull ETF).

5. Risk Budgets: Sleeves, Factors, and Themes

A risk budget allocates scarce loss capacity across activities. Example for a $100k account:

Factor budgets matter as much as sleeve budgets. Cap: single name 10% of equity at risk-notional, single sector 30% of gross, “same catalyst theme” 20%, short book 40% of long gross, options short-premium margin at risk as if stressed. When five swing names are all AI infrastructure, your correlation matrix from Course 37 is not five coins — it is one crowded trade with five tickers. Relative strength and sector leadership help you see the theme; risk budgets force you to size it.

6. Drawdown Protocols and Kill Switches

A drawdown protocol is a ladder of automatic risk reductions as equity declines from a high-water mark (or from month-start, if you prefer calendar governance). A kill switch is a non-negotiable full or near-full risk-off when a hard line is hit.

Example ladder (illustrative — write your own numbers cold):

• −3% from month open: no new experimental risk; review open theses.
• −5%: cut gross exposure 30%; day-trade sleeve off.
• −8%: cut gross 50%; only predefined core holds with tight invalidation.
• −12% (kill): flatten all discretionary risk for a cooling period (e.g. 5 sessions); journal root cause before re-entry.

Daily loss limits (e.g. −1.5% equity → stop trading) prevent revenge spirals that advanced skill cannot fix mid-session. These are behavioral risk controls as much as statistical ones. Pair with process from common mistakes and structural invalidation from market structure / S/R.

Kill switches fail when they are soft (“I’ll reduce if it feels bad”) or when leverage products and options make “flat” unclear. Define flat: no stock, no short, no short premium, no leveraged ETF overnight, futures flat or hedge-only as written policy.

7. Concentration, Liquidity, and Gap Risk

Advanced risk is not only volatility — it is exit feasibility. A position that is 2% risk to a stop on paper but 40% of average daily volume cannot be exited at that stop in stress. Cap position size as a fraction of ADV; widen assumptions for opens and news. Illiquid names and tight floats behave like squeeze fuel when short (Course 47) and like trapdoors when long into bad news.

Gap risk means stops are orders, not promises. Hard catalysts (earnings, binary events) require either reduced size, hedges, or no position. If you hold through events, pre-define the max loss as if the stop is missed by a wide open — that dollar number is your true R, not the tidy intraday stop. ATR and structure still inform placement; they do not create continuous trading.

8. Multi-Instrument Books: Stacking Risk by Accident

Track 5 instruments interact. Classic failure modes:

• Long stock + long calls + long 3× bull ETF on same theme → triple beta.
• Short stock + short puts → disaster if squeeze and assignment path align.
• “Hedge” with inverse ETF held multi-day without understanding daily reset (Course 49).
• Short ES as hedge while adding net-long high-beta names faster than the hedge.
• Iron condors + directional long book that both lose when vol and trend expand the wrong way.

Maintain a delta / beta inventory (even a spreadsheet): approximate equity beta dollars long, short, options delta-equivalent, futures notional. Caps apply to the inventory total, not to each ticket in isolation. Recompute when you roll options or change futures hedges.

9. Measurement Cadence and Dashboards

Minimum professional cadence for a serious equity book:

• Pre-market: event calendar, open risk $, gross/net, top concentrations, kill-ladder status.
• Intraday: P&L vs daily loss limit; do not negotiate with the limit.
• EOD: update high-water / month P&L; check sleeve budgets; note correlation surprises.
• Weekly: stress one historical and one hypothetical scenario; review factor drift.
• Monthly: expectancy and win-rate honesty (break-even, win rate tools); re-validate Kelly-fraction caps on large samples only.

If you cannot measure open $ risk in under two minutes, you cannot manage a complex book. Simplify until you can — fewer names beats a “sophisticated” mess.

10. Failure Modes of “Advanced” Risk Programs

• False precision: three-decimal VaR while position sizes are gut feel.
• Limit tourism: raising limits after a win streak; never lowering after calm markets.
• Hedge theater: small puts that soothe psychology but do not move book P&L in stress.
• Backtest comfort: models fit on a bull sample with suppressed vol of vol.
• Override culture: “just this once” through a kill switch — the most expensive sentence in trading.
• Ignoring financing and margin: overnight futures margin, options assignment, hard-to-borrow recalls.
• Confusing accounting P&L with risk: unrealized winners funding larger losers.

Advanced frameworks exist to constrain future-you under stress. If the framework is optional, it is not a framework.

11. Implementation Checklist (Ship This Week)

1. Write max loss per trade, day, week, month as % of equity.
2. Write max single-name, sector, and theme weights.
3. Define sleeves and dollar risk budgets per sleeve.
4. List open positions with $ risk to invalidation (sum them).
5. Build one historical and one hypothetical stress; record estimated % equity loss.
6. Write drawdown ladder + kill switch with cooling period.
7. Inventory multi-instrument stacking (stock/options/futures/levered ETF).
8. Define daily stop-trading rule and stick it on the monitor.
9. Journal one risk-rule breach from the past month — what override cost you.
10. Schedule weekly 20-minute risk review on the calendar.

Tools for This Course

• Risk Calculator — line-level $R that rolls up into sleeve budgets.
• Kelly Calculator · Win Rate · Break-Even — validate that your edge survives after risk caps.
• P&L Calculator · Percentage Change · SL/TP — stress math and invalidation planning.
• Stock Courses Hub — Track 6 continues with statements, valuation, and portfolio construction.