Trading costs
The simulation models entry and exit fees, configured per-contract charges and slippage. Standard fees apply by default; an eligible cached account percentage-fee snapshot can be used when available.
Historical simulation
Backtesting with more context.
Watch historical markets replay. Inspect the results. Understand the assumptions before you rely on them.
More than a closing price.
Citrine Edge's dedicated Coinbase historical market-data archive supports tests with recorded trade ticks, one-second best bid and ask information, and reconstructed one-second price paths.
That gives the simulation more execution context than candle open, high, low and close values alone. The visual replay uses aggregated candles; it is not a tick-for-tick recording of fills.
See the worked bid/ask execution example to separate quote-side prices, added slippage and fees without counting the spread twice.
- Test BTC and ETH within one portfolio
- Select completed dates, up to 90 days when coverage permits
- Review net PnL, equity, trades and drawdown
- Save, filter, favorite and export detailed results
Understand the result
Costs count. Assumptions count, too.
Margin context
Sizing uses recorded overnight margin information. Where needed, a prior eligible observation for the same product may be carried forward, not information from a later date.
Funding coverage
Eligible recorded funding is used where available. Qualified estimates captured from live market observations can also be used and are identified as estimates. Each run reports its coverage; missing funding data does not prevent a backtest.
Inspect the details
Results and exports let you examine a run beyond its headline return. Compare periods and assumptions rather than selecting only favorable outcomes.
How to read your results
Historical and simulated results are hypothetical. They do not establish future performance or the result you would have received from real orders.
- Archive coverage and warmup requirements constrain the usable test period. No fixed amount of history is guaranteed.
- Fee snapshots are not a reconstruction of every historical account charge. Live fees, funding, margin requirements and execution may differ.
- Queue position, liquidation behavior and exchange interruptions are not fully reproduced. Funding estimates are not final exchange settlement records; coverage can be incomplete for selected dates.
- Maximum drawdown measures the largest decline from an earlier account high at recorded trade exits and the end of the run. It can occur while equity remains above the starting balance, and is not the largest single losing trade or every intratrade equity swing.
Paper Mode uses standard simulation fees. Shadow Mode captures account-based funds and percentage-fee context for a simulated session. Neither places real orders.
Research before committing funds.
Backtesting and replay are included in Explore. No subscription is required, and standard backtesting does not require an API key.