Our Strategy
To measure our screener’s baseline quality, we trade every qualifying setup that triggers each day with our own money. We also built advanced simulation machinery to rigorously replay this exact strategy over the past 20 years. Here’s how it would have performed.
Median of 25 historical simulations · After modeled costs
per year, compounded
Oct 6, 2006 – Oct 8, 2026
Provisional historical results. Not live performance or a forecast.
About these results
The line shows the middle result across 25 simulations at each date. Each period starts with cash and no open positions. Every simulation chooses its own trades using the same strategy rules, portfolio limits and modeled costs.
Our live account trades this same method, starting August 27 and carrying its own positions forward. Historical replay approximates intraday fills and order timing. This median line summarizes simulations; it is not an individual account’s trading history.
- Lowest simulation
- 21.0% per year
- Highest simulation
- 28.2% per year
The spread describes these simulations, not the full range of possible future returns.
Provisional historical simulation. Median of 25 fresh-cash simulations, with liquidity- and order-size-adjusted entry and exit costs. Open holdings valued at daily closes. Some source histories and older timeout exclusions remain under review. Through 2026-10-08. Not live performance or a forecast.
How we trade
We trade what the screener finds
Our own money, fixed rules, no hand-picking.
We trade the setups our screener ranks. It buys breakouts, and only sells short while the S&P 500 is below its 200-day average. Every night the screener drops the breakout patterns without a real record behind them, so what is left is what gets traded.
The same rules on every trade
Each trade risks 0.65% of the account, with a stop at the last swing low.
The buy order waits at the breakout line. If price jumps over the line overnight we buy at the open, and skip it if the jump is more than 5%. Once a trade reaches its target, the stop is cancelled and we hold until it closes below its 10-day average.
20 years, replayed
26.4% a year in a provisional simulation, against 11.4% for the S&P 500.
54% of 1,998 backtested trades won, and the average win was 1.51× the average loss, from 2006 to 2026 with modeled trading costs. This is a historical simulation, not live results, and not a promise of future returns. Its deepest drop along the way was 26.7%. Every trade in it started as a setup on the screener.