Skip to Content
Core WorkflowsResults & Analysis

Results and analysis

When a backtest completes, the Backtest Panel shows the report for that strategy version, data selection, period, and simulation configuration.

Report status

Reports move through PENDING, EXECUTING, COMPLETED, or FAILED.

  • Expand a pending or executing report to see its current status.
  • A completed report exposes its summary, metrics, charts, and—when available—trade activity.
  • A failed report shows the execution error when one is available.

Execution summary

The summary separates the inputs from the results:

  • Execution Details — selected symbol and frequency chips, plus execution time.
  • Simulation Parameters — initial cash, leverage, commission, and Buy On Close.
  • Strategy Parameters — the values submitted for the strategy.
  • Detailed Output — expandable runtime output and error output when present.

The requested date range is part of the run configuration. Use the report and its data selection together when comparing runs.

Performance metrics

Completed reports can show these metric groups.

Performance

  • Final Portfolio Value
  • Realized PnL
  • Total Return
  • CAGR
  • Annualized Sharpe Ratio
  • Annualized Volatility
  • Maximum Drawdown
  • Std Dev of Returns

Risk-Adjusted

  • Calmar Ratio
  • Recovery Factor
  • Profit Factor
  • Payoff Ratio
  • Expectancy
  • Average Trade

Trading activity

  • Number of Trades
  • Win Rate
  • Average Bars Held

Benchmark comparison (S&P 500)

  • Alpha
  • Beta
  • Sortino Ratio
  • Information Ratio

Metrics can be unavailable when the run does not have enough data or trades for a calculation. Treat all values as historical simulation results, not forecasts.

Performance charts

The Performance Analysis section has two tabs:

  • Account Balance — portfolio value through the run.
  • Drawdown Chart — drawdown percentage through the run.

Use these charts to see both growth and periods of decline. A strong return without understanding drawdown can give an incomplete picture.

Trading charts

The trading chart displays candlesticks for the selected symbol, with available indicators and trade markers. Chart settings let you control the view, and the chart can be expanded into its own window.

When a report contains multiple symbols, use the symbol tabs to inspect each data source. If timeline data is unavailable, the chart shows an empty or no-data state instead of inventing values.

Trade log

The Trade Log appears only when the report contains executions. It shows:

  • Total executions
  • Round-trip trade count when available
  • Wins and losses when available
  • Per-symbol tabs for multi-symbol runs

Use the log with the chart markers to understand how the strategy produced its metrics. An empty log does not necessarily mean the report failed; it can mean the rules generated no executions in the selected period.

How to compare runs

When comparing two reports, first align:

  • Strategy version and parameter values
  • Symbols and bar frequency
  • Date range
  • Initial cash, commission, leverage, and Buy On Close

Then compare total return with drawdown, volatility, trade count, win rate, and risk-adjusted metrics. A result with a higher return is not automatically better if it uses more risk or far fewer trades.

Optimize parameter choices

Use Optimize in the Backtest Panel when you want to compare a grid of one or two numeric strategy parameters. The optimizer lets you choose the data, common frequency, date range, simulation settings, parameter grids, and objective metrics. Optimization runs are queued separately from ordinary backtests and can consume a free optimization entitlement or credits.

Review the optimization results before selecting a parameter set for a new backtest. See Optimization best practices.

Research workflow

  1. Start with a simple strategy and a clearly defined period.
  2. Inspect the report summary and the account-balance and drawdown charts.
  3. Use the Trade Log and symbol charts to explain individual outcomes.
  4. Change one assumption at a time and run another backtest.
  5. Use optimization for a bounded parameter sweep, not as a substitute for understanding the rules.

Backtests are sensitive to the selected data, period, costs, and execution assumptions. Record those inputs when you save a conclusion.

Next steps

Last updated on