Five strategies, one ledger.

Five rule-based personas each run a $100k paper portfolio on the S&P 500 using free end-of-day prices. They rebalance on their own logic, accrue a track record with real risk metrics, and once a week a local LLM grades each one's decisions. No real money — just an honest answer to “what would this strategy have done?”

// the raceCumulative return since inception

Each line is a persona's paper portfolio, rebased to its starting capital. Hover a name to focus it.

Loading curves…

// leaderboardHow the personas stack up

#	Persona	Return	Sharpe	Vol	Max DD	Beta	Holds
Loading leaderboard…

Click a persona to inspect its book below.

// the book—

NAV

—

Holdings

Ticker	Weight	Value	P/L

Recent trades

Date	Side	Ticker	Qty	Price	Why

// under the hoodHow it works

Free EOD data. A daily job pulls end-of-day prices for the whole S&P 500 (Yahoo via yfinance, Stooq fallback) into Postgres. No paid feed.

Deterministic personas. Each strategy is pure, price-based logic — momentum, mean-reversion, low-vol, equal-weight and buy-and-hold. Reproducible, no AI in the trade loop.

Weekly AI review. A local LLM (Ollama) runs on separate hardware once a week and grades each persona's decisions — process and risk, not just return.

Read-only here. This page only reads the database; all the heavy lifting happens off Vercel on a scheduled worker.

Five strategies, one ledger.

// the raceCumulative return since inception

// leaderboardHow the personas stack up

// the book—

NAV

Holdings

Recent trades

// the tapeNews on the book

// under the hoodHow it works