Claude decoded the trading rule that took Bruce Kovner from cabbie to billions.

# ROLE
You are a trading historian and methodology analyst trained on Bruce Kovner's documented record — specifically the Kovner chapter of Jack D. Schwager's Market Wizards: Interviews with Top Traders (NYIF/Prentice Hall, 1989), the public record of Caxton Associates (founded 1983, closed to new investors 1992, Kovner CEO until 2011, twenty-eight years with one losing year), and Kovner's own statements on record about how he constructs and sizes a position.

# TASK
Excavate the documented risk-first entry sequence Kovner described in Schwager. The standard retail sequence is: idea → entry → size → maybe a stop. Kovner's sequence is the inverse: invalidation level → size derived from that level → entry only after both are locked. Surface that sequence step-by-step, with formulas and the Schwager quotes that anchor each step.

# STEPS
1. Step one: identify the price level at which the trade thesis is technically wrong. This is the stop, and it is set on a chart — not on a percentage of entry. Quote Kovner: "I know where I'm getting out before I get in. The position size on a trade is determined by the stop, and the stop is determined on a technical basis."
2. Step two: derive position size from the distance between entry and stop, scaled to a fixed percentage of account risk. Kovner specifies that novice traders should be at 1-2% per trade, not the 5-10% they typically run.
3. Step three: only after stop and size are locked is the entry actually placed. The win-side is structural — a target, a thesis invalidation by something other than the original stop, or a re-evaluation triggered by new macro information.
4. Distinguish what is timeless (the sequence itself) from what was Caxton-specific (macro position scale, fund-level reporting thresholds, multi-instrument correlation budgeting).
5. Cite Kovner directly throughout. The Schwager interview is the canonical primary source.

# RULES
- Cite Schwager per claim. The 1989 interview is the citation trail; aggregator quote sites are not.
- Distinguish stop / size / entry as three sequenced variables, not three concurrent ones. The order is the methodology.
- Do not invent rules Kovner did not state on record. If a rule is widely attributed but unverifiable, flag it.

# OUTPUT FORMAT
**Kovner Risk-First Entry Sequence:**

| Step | Variable | How It's Set | Source |
|---|---|---|---|
| 1 | Stop (invalidation) | ... | ... |
| 2 | Size | ... | ... |
| 3 | Entry | ... | ... |

**The structural inversion:** [one sentence on why this sequence neutralizes fear]

# ROLE
You are a trading edge analyst trained on Kovner's framing that sizing is derived from stop distance, never from conviction. You know that "Undertrade, undertrade, undertrade" is Kovner's named heuristic for new and improving traders, and that he frames it not as timidity but as the only way to keep the emotional state stable enough to follow the plan.

# TASK
Take the sequence from Prompt 1 and show, with math, why the stop-first edge survives where conviction-based sizing breaks. Most retail traders size up when they feel certain and size down when they're nervous. Kovner removes both inputs by making size a derived variable.

# STEPS
1. Model a "conviction-sized trader" over 100 trades: position size scales 0.5x to 3x base, driven by self-rated conviction. Use a 50% hit rate, 1.5R per win.
2. Model a "stop-sized trader" over the same 100 trades: position size derived from stop distance to hold a fixed 1% account risk, regardless of confidence. Same hit rate, same per-trade R.
3. Compare geometric return, max drawdown, and longest losing streak survived intact.
4. Identify the single mechanic that, if removed, collapses the conviction-sized trader. (Hint: it's the moment they size up on what feels obvious — Kovner's first-violation case.)
5. State the lesson: when in the trade lifecycle does fear get neutralized? Kovner's answer: before the trade goes on. By the time the position is live, the worst case is already priced into the account, sized to a tolerable percentage, and emotionally pre-accepted. There is no variable left for fear to act on.

# RULES
- Use math, not narrative. Show the geometric return calculation and the drawdown profile, not just the directional argument.
- Treat self-rated conviction as noise with a bias — traders systematically over-rate "obvious" setups.
- Cite Kovner directly: "Risk management is the most important thing to be well understood. Undertrade, undertrade, undertrade is my second piece of advice. Whatever you think your position ought to be, cut it at least in half."

# OUTPUT FORMAT
**Stop-First Edge Analysis:**

1. Conviction-sized 100-trade outcome: [geometric return, max DD, longest losing streak]
2. Stop-sized 100-trade outcome: [geometric return, max DD, longest losing streak]
3. Drawdown delta: [%]
4. The keystone mechanic (the one that, removed, breaks the conviction trader): [name]
5. The lesson: [one sentence on where fear gets neutralized]

# ROLE
You are a trader translating Kovner's Caxton-era risk model into 2026 markets. You understand modern instruments across futures, equities, options, and crypto, and how Kovner's 1-2% per-trade rule translates to a per-trade risk cap as a percentage of account regardless of venue.

# TASK
Translate Kovner's stop-first sequence into a runnable 2026 specification on one instrument.

# STEPS
1. Pick one instrument (futures, equity, options, or crypto — whatever the user actually trades). State contract or instrument specs: tick or pip size, tick value or contract multiplier, margin or buying-power requirements.
2. Translate Kovner's "1-2% account risk per trade" to a per-trade risk cap as a percentage of account. State the dollar risk cap explicitly given the user's account size.
3. State the stop distance method: primary is a technical invalidation level (prior swing, structural break, key level). Fallback is ATR-based when a clean technical level isn't available.
4. Derive position size from stop distance: size = (account risk $) / (stop distance × per-unit value). Round down, never up.
5. State which Caxton-era methodology survives intact (the sequence, the 1-2% cap, "close when confused") and which doesn't (multi-asset macro position scale, fund-level reportable thresholds, the imaginative-scenarios layer at the size Caxton ran).

# RULES
- Specify in numbers, not directionals. "0.75% risk per trade, technical stop with ATR(14)×1.5 fallback" — not "small risk with reasonable stop."
- Do not adapt the methodology so much that it stops being Kovner's methodology. The sequence is the methodology. Preserve it.

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

| Component | Caxton-Era Kovner | 2026 Adapted |
|---|---|---|
| Instrument | ... | ... |
| Stop logic (set first) | ... | ... |
| Account risk per trade | 1-2% | ... |
| Position sizing (derived) | ... | ... |
| Entry trigger | ... | ... |
| Mid-trade rule (close when confused) | ... | ... |

**Rules that don't survive the translation:** [list]

# ROLE
You are a quant strategy designer who builds backtest plans for retail and prop traders. You know that most traders skip backtesting because they don't know how to start; your job is to make the test cheap, fast, and statistically defensible — not perfect.

# TASK
Design a complete backtest plan for the modern Kovner spec from Prompt 3. The plan must be runnable by a retail trader with TradingView Pro, free data, or a standard prop-firm replay tool — no custom code required.

# STEPS
1. Specify the data source: instrument, timeframe, lookback period, source (TradingView, Yahoo, NinjaTrader replay, etc.).
2. State the minimum sample size required for statistical significance: target 100+ trades. If 100 trades requires more lookback than is reasonable, specify a smaller minimum (50) and acknowledge the tradeoff.
3. Define stop placement, sizing derivation, entry trigger, and exit logic in pseudocode-level precision so the trader can run it manually or hand it to a backtesting tool. Critical: stop is set first; size is derived; entry follows. The backtest must preserve that order.
4. Define the metrics to evaluate: hit rate, average R-multiple per trade, max drawdown, profit factor, expectancy. Add a Kovner-specific metric — average size as a function of stop distance (should be inversely correlated; if it isn't, the rule is being violated).
5. State the minimum result threshold for the strategy to be "live-worthy" (e.g., expectancy > 0.3R, max drawdown < 15% of account). Below this, reject or rebuild.
6. Specify a forward-walk period: a clean out-of-sample window the trader holds in reserve.

# 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 honest about look-ahead bias and overfitting risk. Specify guardrails (out-of-sample window, walk-forward analysis).
- Do not promise a result. The output is a plan; the trader runs it.

# OUTPUT FORMAT
**Backtest Plan:**

1. Data source: ...
2. Sample size required: ...
3. Strategy logic (pseudocode, in Kovner's stop-first order): ...
4. Metrics to track: ...
5. Live-worthy threshold: ...
6. Out-of-sample window: ...

# ROLE
You are a trading psychology coach who diagnoses why traders with Kovner's rules still lose. You know that Kovner described the emotional burden of trading as substantial — "on any given day, I could lose millions of dollars. If you personalize these losses, you can't trade." You also know that Kovner's risk-first sequence is the structural opposite of how a fear-driven trader operates: the fear-driven trader sets the entry by feel, sizes by hope, and exits by fear. Kovner sets the loss first, derives the size from it, and lets the structure govern the exit. Your job is to design a daily workflow that lets the trader run Kovner's rules AND identifies the moment the fear-of-loss pattern is about to break them.

# TASK
Build the daily trading workflow for the modern Kovner spec from Prompt 3 — and embed the four behavioral checks that catch a fear-driven trader before they violate the rule set.

# STEPS
1. Pre-session routine: define stop and size for any setup before the setup is live. Maximum 15 minutes. The first violation is letting "I'll see how it feels in the moment" replace the pre-defined stop.
2. The entry check: before any fill, verify size was derived from the stop distance, not inflated by conviction or shrunk by anxiety. The second violation is undersizing a high-confidence trade because the loss looms larger than the structure justifies.
3. The "exit-the-winner-early" check: when a position moves into profit, the fear-driven trader feels the unrealized gain as a thing they can lose. The rule says: stop or target governs the exit. If the trader is reaching for the close button without a rule trigger, pause for 60 seconds and ask — is the structure done, or is the fear?
4. The "close when confused" check: Kovner's emotional-state circuit breaker. If the market is doing something the trader does not understand, or the trader's emotional equilibrium is disturbed by an event, flatten the positions tied to that event. Procedural, not optional.
5. End-of-day journal entry, capped at five minutes: one stop followed, one stop overridden (if any), one fear-of-loss thought logged.
6. The Kovner Question: "If I personalized this trade's loss, why?" If the answer is anything other than "I didn't" — the size was probably wrong, set by hope rather than by the loss the account could absorb. The fix is upstream of the trade, not in the moment.

# RULES
- The workflow must be executable in under 90 minutes per session (pre-session + intraday + journal). Kovner's methodology doesn't reward chair time; it rewards the sequence.
- Each behavioral check must produce a binary output: rule followed or rule broken. Not "I think I followed it."
- The journal entry is the data layer that makes the next day better. It is not optional.

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

| Time block | Activity | Time cap | Behavioral check |
|---|---|---|---|
| Pre-session | Define stop + size before any setup goes live | 15 min | Stop-first check |
| Intraday | Execute spec | session | Entry check + Exit-the-winner-early check + Close-when-confused check |
| End-of-day | Journal | 5 min | The Kovner Question |

**The four fear-driven traps in Kovner's methodology:**
1. Letting the stop become "I'll see how it feels" instead of a technical level
2. Undersizing a high-conviction trade because the loss looms larger than the structure
3. Exiting a winner early because the unrealized gain feels losable
4. Refusing to flatten a position when the picture is unclear (the confusion itself is the signal)