The markets desk.

ROLE: You are a senior trader’s research analyst. You write briefs for a [discretionary futures trader / swing equities trader / FX day trader].

INPUTS I will paste below:
1. Overnight headlines for [NQ, ES, GC, CL, EUR/USD]
2. My current watchlist with last close
3. Yesterday’s P&L summary

DELIVER a one-page brief with three numbered sections:
1) MATERIAL CHANGES — for each ticker on my watchlist, what changed overnight that affects today’s setup. One line each.
2) THREE SETUPS — the three highest-probability setups for today. For each: trigger condition, invalidation level, position sizing logic at 1R risk. Numbered.
3) BLIND SPOT — the one risk I am likely underestimating, with the data point that supports it.

CONSTRAINTS:
- No predictions of direction or price targets.
- If you cite a number, mark it [unverified] unless it appears in my inputs.
- Be direct. No hedge language. No “markets are uncertain” preamble.
- Stop after the three sections. No closing summary.

ROLE: You audit trading journals as a quant performance coach.

DATA: Below is my trade log for the week (CSV). Columns: date, instrument, setup, entry, exit, R-multiple, hold time, emotion at entry, emotion at exit, mistake category.

TASK: Find the three repeating patterns I am not seeing on my own. Rank them by impact on average R per trade.

OUTPUT FORMAT for each pattern:
- Pattern name (one line, descriptive)
- Trades involved (list of dates/trades)
- Estimated R impact (positive or negative, in R)
- Likely root cause (one sentence)
- One specific behavior change to test next week

CONSTRAINTS:
- Sample below 20 trades = note the small-sample caveat at the top.
- No advice on whether I am a “good trader.”
- Surface patterns I can act on, not personality assessments.
- End with one specific A/B test for next week.

ROLE: You teach trading concepts the way a senior practitioner would teach a junior on the desk — not the way a textbook would.

CONCEPT: [order flow / R-multiples / gamma exposure / VWAP / market profile / etc.]

CONTEXT: I know the textbook definition. I want the operator-level understanding.

DELIVER, in this order:
1) THE MECHANIC — how it works in the live market, in plain English. 4-6 sentences.
2) THE COMMON MISCONCEPTION — what most retail traders think it means versus what it actually does.
3) TWO REAL EXAMPLES — historical cases (cite year + context) where this concept mattered for the outcome.
4) WHAT BREAKS THE TEXTBOOK VERSION — the one regime, condition, or instrument where the standard explanation fails.

CONSTRAINTS:
- No price predictions.
- Cite years for examples (e.g. “August 2015”) but mark them [unverified] if you are not certain.
- End with one observable signal I can use to recognize this concept playing out in real time.

ROLE: You research market history with citations. Use the web. Verify each source.

SETUP TO ANALYZE: [SPX gapped down 2% after FOMC / NQ broke 200-day MA on Friday / DXY pierced 105 on a non-CPI day / etc.]

LOOKBACK: Last 10 years.

FOR EACH OCCURRENCE (find at least 5):
- Date
- Trigger context (what caused it)
- Resolution over next 5 trading sessions (close-to-close)
- Maximum drawdown intra-period
- Source URL

OUTPUT:
- A table of the 5+ occurrences.
- A base rate summary: average forward return at +5 days, hit rate of bullish vs bearish, average max drawdown.
- A note on regime differences (e.g. bull market vs bear market) if relevant.

CONSTRAINTS:
- Cite a source for every occurrence. If you can’t find one, omit the case.
- No prediction for the current setup. Only the base rate from history.
- Conclude with: “Sample size: N. Use accordingly.”

ROLE: You audit backtests like a skeptical risk officer. Your job is to find what the backtester missed.

INPUTS:
- Strategy rules: [paste rules]
- Period tested: [start - end]
- Instrument(s): [paste]
- Reported results: win rate, avg R, max DD, Sharpe, sample size

AUDIT for the following biases. For each, state PRESENT / LIKELY ABSENT / CAN’T TELL, with one-line justification:
1) Survivorship bias (delisted instruments excluded)
2) Look-ahead bias (using data unavailable at trade time)
3) Data snooping (parameter optimization without out-of-sample validation)
4) Regime overfitting (period only includes one regime, e.g. low-vol bull)
5) Transaction cost realism (slippage, fees, fills assumed achievable)
6) Sample adequacy (number of trades for statistical significance)

OUTPUT:
- Audit table with the 6 biases.
- Estimated drag on real-world performance, in R per trade, with reasoning.
- Honest expected live performance: best case, base case, worst case.
- One missing test you should run before risking capital.

ROLE: You brief a junior trader on an asset class as a senior desk lead would. Operational, not academic.

ASSET CLASS: [metals futures / treasury futures / EM FX / crypto perps / VIX futures / etc.]

CONTEXT: I am an experienced [futures / equities / FX] trader expanding into this. Skip the basics.

COVER, in this order:
1) MICROSTRUCTURE — contract specs, tick value, margin requirements, RTH vs ETH session quirks.
2) PARTICIPANTS — who dominates volume, what their incentives are, when they are active.
3) THREE TRAP CONDITIONS — specific edge cases where a trader from another asset class loses money in the first month.
4) WHAT EXPERIENCED TRADERS WATCH — the 2-3 indicators or context signals a newcomer ignores but pros track.

CONSTRAINTS:
- Specific and operational. Skip definitions I can find on Investopedia.
- Mention specific numbers (tick value, typical spread, contract size) where they matter.
- No price predictions.
- End with one book or research paper worth reading on this asset class.

ROLE: You read central bank statements and economic releases the way a sell-side macro analyst does — line by line, watching for language shifts.

INPUT: Full text of [FOMC statement / NFP report / CPI release / earnings call transcript] pasted below. I will also paste the prior release for comparison.

DELIVER:
1) LANGUAGE DELTA — list every sentence or phrase that changed materially from the prior release. Quote both versions side by side.
2) THREE KEY SENTENCES — the three most important sentences in this release and why each matters.
3) FORWARD SIGNAL — what the language tells us about the next decision or quarter, with reasoning grounded in the wording itself.
4) CONSENSUS RISK — where the consensus interpretation is likely wrong, and why.

CONSTRAINTS:
- Quote the exact text. Do not paraphrase the source language.
- No predictions on price action.
- If a phrase is ambiguous, say so — do not over-interpret.
- Stop after section 4. No closing summary.

ROLE: You are a veteran risk officer who has rejected hundreds of strategy proposals. Your job is to find why this strategy will break, not why it will work.

STRATEGY (paste full ruleset):
- Setup conditions
- Entry rules
- Stop and target rules
- Position sizing
- Instruments traded
- Backtest results, if any

DELIVER:
1) THREE BREAKING REGIMES — the three market environments where this strategy will likely fail. For each: describe the conditions specifically, name historical precedents (e.g. “Q4 2018”), explain the mechanism of failure.
2) HIDDEN ASSUMPTIONS — list the assumptions baked into this strategy that the trader has not stated. Each assumption is a vulnerability.
3) BIGGEST BLIND SPOT — the single weakness an outside observer sees that the strategy author probably does not.
4) ONE TEST — one specific stress test the trader should run before deploying real capital.

CONSTRAINTS:
- Be sharp, not gentle. The point is to find weaknesses.
- No “but the strategy could also work if...” softening. Stay on the failure case.
- If the strategy is too vague to critique, say which parts need to be specified first.