Backtesting
Test an idea against real stored ticks instead of resampled bars, using the same bar types and studies the live charts run.
Why tick-level
A backtest is only as honest as its bars
Most retail backtests run on minute bars and assume fills. Both assumptions flatter the result, and the flattering is largest exactly where the edge is thinnest.
Real stored ticks
The same tick history the charts are built from, not a resampled series.
Same bar construction
Range, volume and tick bars are built the same way as live, so a strategy is not tested on a different instrument from the one it trades.
Order-flow inputs
Delta, imbalance and profile values are available to a test, not just OHLC.
Honest cost modelling
Commissions and slippage as explicit inputs, because a per-trade edge smaller than costs is not an edge.
Specifics
At a glance
Before you ask
Questions
How much history will it cover?
Whatever tick history you have recorded and stored locally. Depth of history is a function of how long you have been running the feed, not a bundled dataset.
Will it prove a strategy works?
No backtest does that. It can tell you an idea failed, which is most of the value. Treat a good result as a reason for more scrutiny, not less.
Can I automate a strategy that passes?
That is what the automated trading work is for — see that page for its status.
Early access
Get in before launch
Founding rate, plus a build to try before it goes public.