Live Track Record — Since April 14, 2026

An honest
ledger of a quiet
algorithm.

A fully automated, ergodicity-filtered trading system, reporting every closed position in real time. No backtests, no curve fits, no hindsight. The numbers on this page are recomputed from the database on each load.

Portfolio Value

—

from an initial $100,000 paper balance

Cumulative Return

—

since inception · paper trading

Win Rate

—

— closed positions

Time-Growth Rate g

—

ergodic, per-trade geometric mean

The curve, without ornament.

Daily marks on the book balance. Hover the line for a reading.

Period return

—

Since Apr 14, 2026. Starting capital $100,000. Marks are end-of-session.

Max drawdown—

Best day—

Avg P&L / trade—

II.

On ergodicity, briefly.

After Ole Peters, Nature Physics, 2019.

“

The gambler who wins on average may still be ruined in time. We optimise the trajectory, not the ensemble.

— House rule · Ventury Alpha

Every signal must clear one gate before it reaches the market: the expected log-return of the bet must be strictly positive. This filter rejects trades that look profitable on average but compound to zero, or worse, over a finite life. Position sizes follow a half-Kelly rule, capped at 15% of capital.

III.

Closed positions, the book.

Every trade, in reverse chronological order. No selection, no trimming.

Loading open positions…

— positions on record Ergodic filter · ON Half-Kelly sizing Realtime · Supabase

№AssetSide Entry Exit P&L % Trajectory Closed

Loading book…

IV.

How the book is made.

Six components, from signal to settlement.

Quantitative engine

A GARCH(1,1) volatility regime detector paired with a hidden Markov model for market-state classification. Momentum is extracted via a Kalman filter.

ii.

Signal generation

RSI divergence and momentum breakouts. Every candidate is validated by a 50-agent Monte Carlo “swarm gate” before it is ever written to an order.

iii.

Ergodic filter

Each signal must satisfy g = p·log(1+w) + (1−p)·log(1−ℓ) > 0. Trades positive on average but destructive in time are rejected outright.

iv.

Half-Kelly sizing

Position size f = 0.5 · (p·w − (1−p)·ℓ) / w, with a hard 15% capital ceiling. We aim for trajectory growth, not variance.

Execution

Orders submitted at U.S. market open through the Alpaca API, from a dedicated VPS. A 24/7 monitor posts stop-loss and take-profit events to a private Telegram.

vi.

Transparency

All trades are written to Supabase at the moment of closure. The time-growth rate g is recomputed from the full ledger. No backtests are displayed on this page.

An honestledger of a quietalgorithm.

An honest
ledger of a quiet
algorithm.