How it works
From an idea to a verdict, in four steps
No code. You choose a rule from a menu, the lab runs the fair test in a few minutes, and you get a verdict in plain words. If it holds up, you can watch it trade on paper.
Build a rule
Choose when to buy (a moving-average crossover, a breakout to new highs, or a short dip in an uptrend), when to sell (the signal reverses, or after a number of days), the stop that protects each trade, and an optional profit target. Then pick the stocks it is built on: the Dow 30, or 20 other large caps.
Run the fair test
The lab replays the rule on daily prices from 2016, with a fee and slippage on every trade. Then it asks four questions, and the rule must pass all of them:
- Unseen years. Built on 2016–2021, judged on 2022 onwards. It must earn more than zero per trade after costs, by more than its own uncertainty.
- Unseen stocks. The other set of stocks, never used to build it, must confirm it.
- Against holding. In at least half the stocks, it must do at least as well as simply buying and holding, on risk-adjusted return.
- Against luck. Its average trade must beat 95% of runs where the same trades, of the same length, are taken on random dates.
Read the verdict and the traps
One sentence first: Passed, Does not hold up or Too few trades. Then the reasons, and the traps found on the way: a rule fitted to its years, a high win rate that still loses money, profits that come from one stock, a rule that is rarely in the market. Every number comes with the range it could really be in, not just the flattering middle.
Run it on a paper account
Connect your own free Alpaca paper account and turn on the autopilot. It follows your rule every day, places its orders with their stops, and lets you pause, close a position or stop at any time. The portfolio page shows what Alpaca reports: positions, orders and the account's value. Paper money only.
What the test will not do
It will not tell you what to buy, predict next year, or turn a rule that failed into one that works. A rule that passes has survived a strict test on the past; it can still lose money in the future. That is why the next step is paper trading, not your savings.