Backtesting: test any strategy on history

Every strategy is tested before you deploy it. Change the setup by asking.

July 25, 2026Conviction team

Every strategy on Conviction is tested on past prices before you save or deploy it. You see what it would have done, trade by trade.

What you can do

  • Get a backtest automatically. It runs as soon as a strategy is built. The window is chosen for you, long enough to mean something.
  • Pick your own dates. "Test it over 2022."
  • Change the amount. "Try it with $5,000."
  • Turn guardrails on or off and rerun to see the difference.
  • Run several setups side by side and compare them.
  • Test on past events. "Test it on past wars." "Test it on past rate-hike cycles." The strategy is run over earlier episodes of the same kind, each scored for how much it resembles today.
  • Backtest after every edit, before you save the change.

What you see

  • Return, number of trades, win rate, profit factor and the worst drop.
  • An equity curve across the whole window.
  • Every trade with its date, price and reason.
  • The same rule on five similar stocks, so you can tell a real pattern from a coincidence.
  • Threshold suggestions. Whether a nearby setting, such as RSI 35 instead of 30, would have held up better.
  • A reason when there are no trades, such as a guardrail blocking every entry.

No hindsight

A backtest only uses what was public at the time. Earnings count from their release, filings from their filing date, inflation figures from the day they were published.

Every market has its own test

  • Options replay each contract's historical bid and ask, and count commissions.
  • Crypto runs seven days a week on your exchange's prices, with its fees.
  • Futures run on each product's own session, with rolls and margin.
  • Staged buys are tested purchase by purchase, and the result shows the average return per round.

Sample strategies

  • Buy $500 of NVDA when the daily RSI drops under 30. Sell at +12% or after 10 trading days. Test it over 2022.
  • Each month, hold the five S&P 500 stocks with the best six-month return, equal amounts. Test it on past rate-hike cycles.
  • Buy $1,200 of GLD and $800 of TLT when the 10-year Treasury yield falls below 4%. Test it on past wars.
  • Buy SOL when it closes above its 50-day average. Sell on a 10% trailing stop. Test it over the last five years.
  • Buy $1,000 of AAPL when the daily RSI drops under 30. Sell at +10%. Compare it with an RSI threshold of 35.
  • Buy $200 of QQQ when it closes above its 50-day average. Sell on a 5% trailing stop. Run it with and without my guardrails.

Got an idea? We’ll trade it.

Get Started for Free

A backtest shows how rules would have performed on past prices. It does not predict future results.

Cover photo: CC0, via Wikimedia Commons