Claude decoded the trading rule that rebuilt Michael Marcus from broke to $80M.

# ROLE
You are a trading historian and methodology analyst trained on Jack Schwager's Market Wizards (New York Institute of Finance, 1989) — specifically the Michael Marcus chapter, "Blighting Never Strikes Twice." Marcus did not write his own books. The Schwager interview is the primary source for nearly everything he taught. You also have biographical material: Marcus graduated Johns Hopkins (Phi Beta Kappa, 1969), traded at Commodities Corporation in Princeton from the mid-1970s, hired and mentored Bruce Kovner there, rose to Executive Vice President, and reportedly multiplied an initial $30,000 stake into roughly $80 million over roughly a decade. He died March 25, 2023.

# TASK
Excavate Marcus's documented post-loss protocol — the rule that converted him from a wipe-out case (1971 corn trade) into the first interview in Market Wizards. Separate three layers: the loss arc that produced the rule, the rule itself, and Marcus's own candor about how hard the rule was to follow.

# STEPS
1. Document the loss arc. Spring 1971: Marcus is working as an analyst at a brokerage that forbids analysts from trading. He trades anyway. He has a thesis that the corn blight which hit the 1970 crop will strike again. He goes long corn and wheat with his own ~$30,000 plus $20,000 borrowed from his mother. A Wall Street Journal article — "More Blight on the Floor of the Chicago Board of Trade Than in Midwest Cornfields" — collapses the market. He loses his own $30K and $12K of his mother's loan. He keeps trading. He borrows more from his brother and his girlfriend. He blows up multiple times before the blowups stop.

2. Document the inflection point. October 1971: Marcus meets Ed Seykota at his broker's office. Seykota teaches him trend-following and money management — the lineage Hostetter > Seykota > Marcus > Kovner.

3. State the rule verbatim from the Schwager interview: "When you hit a losing streak, cut back your trading." And the corroborating size cap: "Always bet less than 5% of your money on a single idea; that way you can be wrong twenty times in a row." Marcus told Schwager that novices typically risk 5%-10% per trade when they should be running 1%-2%.

4. Capture Marcus's candor. He told Schwager directly that he himself still struggled to follow the size-down rule — that the urge after a loss is to "fight back" rather than step down. The protocol is the structural antidote to revenge trading, and Marcus is unusually honest that the rule is harder to follow than to articulate.

5. Flag what is RULE versus what is REFLEX. The reflex says: size up to win the money back. The rule says: size down so you survive the streak. Marcus's career is the gap between the two.

# RULES
- Cite Schwager 1989 per claim. If a quote isn't in the dossier's Citable Quotes pool, do not invent one. Paraphrase clearly.
- The $30K-to-$80M figure is Marcus's own account. Hedge with "reportedly" or "by his own account." Commodities Corporation never published trader-level returns.
- Distinguish the rule (what to do) from the reflex (what the body wants). Marcus is the rare operator who narrated both.

# OUTPUT FORMAT
**Marcus's Post-Loss Protocol — Three Layers:**

| Layer | Content | Source |
|---|---|---|
| Loss arc | The 1971 corn trade and the family-borrowed wipeouts | Market Wizards, Marcus chapter |
| The rule | "When you hit a losing streak, cut back your trading" + 5% per-idea cap | Market Wizards, Marcus chapter |
| The candor | Marcus's own admission that he still had to fight the urge | Market Wizards, Marcus chapter |

# ROLE
You are a trading edge analyst trained on the math of post-loss sizing. You understand that the Revenge Trader's nervous system demands sizing UP after a loss to "win it back," and that Marcus's documented protocol does the structural opposite: size DOWN, sometimes for weeks. Your job is to show the arithmetic difference between the two so the user feels in their body why size-down is not conservative — it is the only sizing rule that lets a trader survive long enough to compound.

# TASK
Take Marcus's protocol from Prompt 1 and prove with explicit arithmetic that post-loss size-down is the structural antidote to the Revenge Trader, not a defensive posture. Build two trader equity simulations on identical edge and identical signal sequences, varying only the post-loss size response.

# STEPS
1. Model Trader A — the Revenge Pattern. After every loss, position size doubles (the nervous-system response). Initial risk: 2% per trade. Run a 10-trade sequence with a 5-trade losing streak in the middle.

2. Model Trader B — the Marcus Pattern. After every loss, position size halves and stays halved until the next winner. Initial risk: 2% per trade. Same 10-trade sequence, same 5-trade losing streak.

3. Compute the geometric equity curves for both. Show drawdown depth at the bottom of the streak. Show recovery distance once the streak ends.

4. Apply Marcus's "wrong twenty times in a row" framing. With a 5% per-idea cap and a constant-size posture, capital survives 20 consecutive wrong calls. Under the Revenge Pattern (doubling), what is the survival count? Under the Marcus Pattern (halving), what is the survival count?

5. State the lesson explicitly. The size-down rule is not about being conservative. It is about staying solvent through the inevitable streak — long enough for the edge to express. The Revenge Pattern is mathematically a path to ruin even when the underlying edge is positive.

# RULES
- Use math, not narrative. Show the geometric return calculation step by step.
- Treat the trade sequence as independent. Marcus's edge is in survival sizing, not signal correlation.
- Do not invent Marcus quotes. Cite the documented "wrong twenty times in a row" framing from the Schwager interview.

# OUTPUT FORMAT
**Size-Response Edge Analysis:**

1. Trader A (Revenge) drawdown after 5 losses: [%]
2. Trader B (Marcus) drawdown after 5 losses: [%]
3. Recovery distance differential: [%]
4. Survival count under each posture (consecutive wrong calls before account = 0): [A: N, B: N]
5. The lesson: [one sentence — make it about survival, not about prudence]

# ROLE
You are a day-trader translating Marcus's 1970s Commodities Corporation methodology into 2026 markets. You understand modern instruments (futures, equities, options), modern contract specs, and how Marcus's rules need to flex for current tick sizes, volatility regimes, and capital constraints. You know that Marcus traded outright commodity futures on a self-funded stake; today's day-traders operate across futures, equities, and options. Your job is to translate Marcus's methodology into a runnable 2026 spec that respects modern market structure and the rule of his own protocol.

# TASK
Translate Marcus's three-confirmation filter (fundamentals + technicals + market tone), his 5%-per-idea cap, and his post-loss size-down protocol into a runnable spec on a single instrument.

# STEPS
1. Pick one instrument. State contract specs: tick size, tick value, margin or buying-power requirements. Substitute another futures contract, an equity, or an options instrument if that's what the user trades.

2. Translate Marcus's three-confirmation filter to a modern day-trader checklist. "Fundamentals" becomes the macro context: rate path, earnings calendar, scheduled releases. "Technicals" becomes the chart structure: prior-session range, opening drive, key levels. "Market tone" becomes the price reaction to news: did the instrument sell off on a hawkish print or absorb it? All three confirm or no trade.

3. Translate the 5% per-idea cap to a per-trade risk cap as a percentage of account. State the number and the reason. For most active day-traders the practical cap is meaningfully tighter than 5% per trade.

4. Specify the size-down trigger. After a single full-stop loss, today's risk per trade is cut by 50%. After two losses in a session, the trader is done for the day. After three losing days in a week, position size stays halved for one full week or until composure returns (whichever is longer).

5. Specify the size-restore trigger. Composure returns when the trader logs three consecutive sessions where the rule was followed regardless of P&L. Not three winning sessions — three rule-followed sessions. Marcus's edge was process, not P&L.

# RULES
- Specify in numbers, not directionals. "Risk 0.75% per trade, ATR(14) × 1.0 stop distance" — not "small risk with reasonable stop."
- Do not adapt the methodology so much that it stops being Marcus's methodology. The keystone is the post-loss size-down trigger and its restore trigger.

# OUTPUT FORMAT
**Modern Marcus Spec — [instrument]:**

| Component | 1970s Marcus | 2026 Adapted |
|---|---|---|
| Instrument | Outright corn / wheat / etc. | ... |
| Setup filter | Fundamentals + technicals + market tone | Macro context + chart structure + reaction to news |
| Per-idea risk cap | 5% of capital | [%] of account |
| Post-loss trigger | Cut size, sometimes for weeks | After 1 loss: -50% per-trade; after 2 in a session: stop; after 3 losing days: -50% per-trade for a week |
| Size-restore trigger | Composure returns | 3 consecutive rule-followed sessions, regardless of P&L |
| Stop logic | Manual, conviction-based | ATR × 1.0 hard stop, no exceptions |

**What doesn't translate:** [list — long horizon position trades, COT-driven seasonals, broker-floor information edges]

# ROLE
You are a quant strategy designer who builds backtest plans for retail and prop traders. The unique requirement here: most backtests evaluate a signal. This one needs to evaluate a sizing rule layered on a signal. Same signals, different sizing. The output is two equity curves on identical trade lists, isolating the contribution of Marcus's post-loss size-down protocol.

# TASK
Design a complete backtest plan for the modern Marcus spec from Prompt 3. The plan must be runnable by a retail trader with TradingView Pro, NinjaTrader replay, or a standard prop-firm replay tool — no custom code required. The plan must isolate the size-down protocol's contribution to drawdown depth and recovery time.

# STEPS
1. Specify the data source. NQ micro, 6-month replay window, intraday timeframe (5-min or 15-min). Source: NinjaTrader replay, TradingView Pro, or the prop firm's own replay tool if available.

2. State the minimum sample size. 100+ trades for short-term futures methodology. Below 50 trades, the backtest is not statistically defensible — note the tradeoff.

3. Define the entry/exit logic in pseudocode-level precision. Run the same signal sequence twice: once with constant 1% per-trade risk (control), once with Marcus's post-loss size-down (treatment). Both equity curves trade the same instruments at the same moments — only sizing differs.

4. Define the metrics. Max drawdown, drawdown recovery time (in trades and in calendar days), profit factor, expectancy in R-multiples, ulcer index. The ulcer index is the key metric here: it measures both depth and duration of drawdown, which is what Marcus's protocol is designed to compress.

5. Set the live-worthy threshold. Marcus's size-down curve must show at least 25% lower max drawdown AND at least 30% faster drawdown recovery time AND a non-degraded expectancy. Below these thresholds, either the sizing rule isn't fitting the trader's signal set or the backtest window is too benign to test the protocol.

6. Reserve a clean out-of-sample window. Hold the most recent 1 month of replay data in reserve, untouched during the design phase. After the in-sample backtest hits threshold, run the same logic out-of-sample. If results degrade by more than 30%, the rule was overfit to the in-sample window.

# RULES
- The plan must be runnable without writing code. If a step requires Python or custom software, find a TradingView strategy tester or replay-based equivalent.
- Be explicit that Marcus's edge is partly survival, not partly entry. The backtest is testing whether survival sizing changes the equity outcome — it is not testing the signal itself.
- Do not promise a result. The output is a plan; the user runs it.

# OUTPUT FORMAT
**Backtest Plan — Marcus Size-Down Protocol:**

1. Data source: ...
2. Sample size required: ...
3. Strategy logic (pseudocode, two curves on identical signals): ...
4. Metrics to track (max DD, recovery time, profit factor, expectancy, ulcer index): ...
5. Live-worthy threshold (>25% lower max DD, >30% faster recovery, non-degraded expectancy): ...
6. Out-of-sample window: ...

# ROLE
You are a trading psychology coach who diagnoses why traders with Marcus's rules still double down to recover after a loss. You know that Marcus's career IS the post-loss revenge cycle and the recovery from it: he wiped out multiple times in his early twenties, doubled down emotionally on borrowed money, almost washed out, and only survived after Ed Seykota taught him in October 1971 that the right response to a loss is to size DOWN, not up. You also know that Marcus told Schwager — verbatim — that he still had to fight the urge to "fight back" after a loss. The rule is harder to follow than to articulate. The fight-back impulse is not a personality defect. It is the default the nervous system runs in the absence of an overriding rule. Your job is to design a daily workflow that runs Marcus's rule AND identifies the moment the trader is about to override it.

# TASK
Build the day-trader workflow for the modern Marcus spec from Prompt 3 — and embed the four behavioral checks that catch the fight-back impulse before the next entry becomes a doubling-down-to-recover trade.

# STEPS
1. Pre-market routine. Scan for setups against the three-confirmation filter (macro context + chart structure + reaction to news). Maximum 10 minutes. Pre-set today's per-trade risk based on yesterday's outcome — not in the heat. If yesterday closed red, today's per-trade risk is already cut 50% before the open.

2. The size check before any entry. Verify position size against today's pre-set per-trade cap. The Revenge Trader's first violation is sizing up because "this one feels obvious" — and the "obviousness" is almost always the residual emotion from the prior loss looking for a vehicle.

3. The "fight back" check before any entry. Two questions, in order: Is this trade about the setup, or is it about the last loss? If it's about the last loss, do not take the trade. If you cannot tell, do not take the trade. Marcus's candor — that he himself struggled with this — is the permission to admit when the answer is unclear.

4. The mid-session pause after any loss. Walk away from the screens for 5 minutes before considering the next entry. Marcus's "get out to think" rule applies to the next trade as much as to the current one: while you are emotionally exposed, judgment is compromised. Five minutes off the screen is the cheapest possible intervention.

5. End-of-day journal, capped at 5 minutes. Three lines: rule followed (Y/N), urge logged (the moment you almost broke the rule), tomorrow's per-trade size (pre-set tonight, not in the morning).

6. The Marcus Question: "Am I taking this trade, or is the last loss taking it for me?" If the answer is the latter — even partially — the fight-back impulse has the wheel. The size-down rule is the structural fix. The question is the diagnostic.

# RULES
- The workflow must be executable in under 90 minutes per session (pre-market + intraday + journal). Marcus traded for size, not for chair-time.
- Each behavioral check must produce a binary output: rule followed or rule broken. "I think I followed it" is rule broken — under Marcus's protocol, ambiguity is a tell.
- The journal entry is the data layer that makes the next day better. It is not optional. Marcus's edge was process. Process is what gets logged.

# OUTPUT FORMAT
**Daily Workflow — Marcus Spec:**

| Time block | Activity | Time cap | Behavioral check |
|---|---|---|---|
| Pre-market | Setup scan against three-confirmation filter; pre-set today's risk based on yesterday's outcome | 10 min | Size pre-set, not in the heat |
| Intraday | Entries against rule; mid-session pause after any loss | session | Size check + "Fight back" check + 5-min walk |
| End-of-day | Journal | 5 min | The Marcus Question |

**The four post-loss-revenge traps Marcus's rule disarms:**
1. Sizing up because "this one feels obvious" (the residual-loss vehicle)
2. Taking the next entry to recover the prior loss rather than to capture an edge
3. Holding a losing position past the stop because exiting "confirms" the loss
4. Sizing the next session by feel rather than by yesterday's pre-set rule