Backtest Lab

Run your own backtests and optimisations of a strategy, outside the pipeline, with settings you choose.

What the Backtest Lab is

The Backtest Lab runs a backtest or an optimisation of a strategy on your terms, without sending it through the pipeline again. It uses the same engine, data and cost model as the pipeline, so a lab result is comparable with the strategy's own results.

What you can test

You can test the current version of any of your own strategies, strategies you brought in with Import strategy once their port has passed verification, and marketplace copies while your subscription keeps their code unlocked.

Settings you control

Choose the test window, starting capital, fees, leverage and parameter values. For an optimisation, also set the search range of each parameter. Starting capital begins at $100, fees can be up to 0.5%, and the window can start in 2010 at the earliest. A backtest needs at least a day of data, an optimisation at least 90 days.

Price and credits

The price of a job is shown before you start it and fixed when you submit. A backtest costs a few credits that grow with the number of bars; an optimisation costs more because it runs many backtests, depending on how many parameters it searches. Lab jobs need enough credits in your balance. If a job fails before it runs, is cancelled before it starts, or is lost, its credits are refunded in full.

Limits

You can have up to 3 lab jobs active at once, only one of them an optimisation, and start up to 30 jobs per hour.

Reading lab results

Each job reports the Sharpe ratio, total return, maximum drawdown, trade count, win rate and the equity curve. A lab optimisation is your own experiment: it does not change the strategy's pipeline record, its promotion status or its gates.