ARK Cathie Playbook

Generated from LEDGER.json by ledger-stats · as of 2026-08-17 local. Do not hand-edit; prose narrates, this block is the record.

Open positions

Who I Am (External Persona)

I am ARK Invest — Cathie Wood's disruptive innovation shop. I don't talk in this system. I'm mirrored: every trade ARK publishes for their flagship innovation fund (ARKK), I replicate here at scaled position sizing.

This persona is not an internal voice. It is a signal source. What ARK buys and sells is the signal. My P&L tells you whether their high-conviction disruptive innovation thesis is working in this regime.

Mission Statement (as copy-trader)

Mirror ARK Invest's flagship ARKK fund trades in a $100K paper book. Scale each ARK trade to our book based on ARK's own position sizing — if ARK adds 0.3% of ARKK to a ticker today, we add 0.3% of our $100K ($300) to the same ticker. This preserves the portfolio-weight semantics of their trades.

Methodology (ARK's, not ours)

ARK's core thesis, as documented in their published research:

1. Disruptive innovation is underpriced by the market relative to 5-year TAM projections
2. Concentrated conviction beats diversification — top 10 positions often > 50% of fund
3. Sell winners into strength — if a position rips, they trim and rotate into lower-conviction names that have pulled back
4. Buy weakness aggressively — when conviction names drop, they add. Often into -50%+ drawdowns
5. Themes over sectors — genomic revolution, AI, blockchain, robotics, space. Not SaaS vs hardware.

Entry Rules (as mirrored)

1. ARK published a BUY today for ARKK → we BUY the same ticker at (etf_percent / 100) * $100K (our book-weighted equivalent).
2. If we already hold the ticker, we ADD the scaled amount.
3. If the ticker is not in our watchlist, we still mirror the trade (ARK picks names before we do).
4. Must have OHLC data available — skip if the ticker isn't fetchable.
5. Minimum trade size: $50. Below that, skip (noise level).

Exit Rules (as mirrored)

1. ARK published a SELL for ARKK → we SELL proportionally. If they sold 20% of their position, we sell 20% of ours.
2. If ARK fully exits, we fully exit.
3. No independent stop-losses. We follow ARK's discipline, not ours. The whole point is to measure whether their methodology beats our research.

Position Sizing

Scaled to ARK's own weighting of ARKK. If ARK's trade represents 0.5% of ARKK that day, we allocate 0.5% of $100K = $500 to the same trade.

What I DON'T Do

• No independent thesis. I'm mirroring, not researching.
• No RSI, no technicals, no "maybe ARK is wrong here." ARK publishes, we mirror.
• No multi-fund aggregation for v1. ARKK only. ARKG/ARKW/ARKQ/ARKF/ARKX can be added later as separate ledgers if useful.
• No same-session trade compression. If ARK bought and sold the same ticker in one week, we record both.

The Core Truth

Running ARK as a copy-trader persona answers two questions simultaneously:

1. Is ARK's style working right now? Bull markets for innovation: ARK rips. Rate-hike cycles: ARK gets crushed. Having their performance measured inside our book tells us which regime we're actually in.

2. Is OUR research beating ARK's? If claude-trader + claude-momentum + yolo combined can't beat a passive mirror of ARK, maybe we should just copy ARK.

This is the same "honest measurement" philosophy as DCA Larry, but at the style level instead of the passive baseline level.

Data Source

• Primary: arkfunds.io v2 API (/etf/trades?symbol=ARKK) — unofficial but reliable mirror of ARK's daily CSV publications
• Import cadence: daily (or weekly at full scan time)
• Import tool: deno task ark-import

Current Regime Read

To be filled in after first import.

Performance Notes

To be populated after first import.