Skip to content

Backtests and results

A backtest replays one saved Strategy version over historical data and reports the trades and account path produced by its rules. Use it to inspect behavior under stated assumptions, compare versions, and find what needs more testing.

On this page

Prepare the version

Open the Strategy and confirm its instruments, bar type, rules, and source on Setup. On Backtest, choose the date window and trading-cost assumptions shown for the run. These settings belong to the evidence you are about to read, so choose values that match the market and holding period you intend to study.

Run the pre-check first. It verifies that the version can be evaluated and that the requested data is usable. If Mavori identifies missing or partial coverage, resolve the named issue or narrow the test before treating the result as representative.

Run and read the report

  1. Select the version

    Use the version list on the Backtest tab. The report always belongs to the selected version.

  2. Set the test

    Review the date window, instruments, cadence, and cost assumptions.

  3. Run the backtest

    Start the run and return when the result is ready; the Strategy keeps the completed report.

  4. Read behavior before return

    Start with coverage, trade count, max drawdown, and the equity curve, then read total return, win rate, benchmark comparison, and individual trades.

Interpret the result

The first evidence to review
ResultWhat to look for
CoverageWhether the requested instruments and period had enough usable data.
Trade countWhether the result rests on a meaningful number of decisions.
Max drawdownThe largest decline the tested account experienced.
Equity curveWhether gains were broad or depended on a short favorable stretch.
Costs and assumptionsWhether fills and friction are plausible for the strategy.
Trade listWhich entries, exits, symbols, and periods drove the summary.
The first evidence to review

Choose the next step

If the run exposes a rule problem, return to Strategies and versions, fork the version, and change one idea at a time. If the behavior remains understandable across meaningful tests, review the target account, capital, and Risk Controls before creating a Strategy Deployment.

Last updated .