Product documentation

1. Check data coverage

The Data page shows which market is available, its supported timeframes, the covered date range, and data-quality state. Research currently covers Binance BTCUSDT USDⓈ-M perpetual futures. Days whose data had to be reconstructed after an audited gap in the source trades are labeled as such, never presented as clean.

2. Explore historical data

Explore charts prepared candles with selectable features: price, volume, buy/sell volume, trade counts, trade size, delta, VWAP, cumulative volume delta (CVD), and rolling statistics such as a volume percentile rank. Click any bar to inspect its exact values.

Test a condition shows how often a rule such as “volume percentile rank above 90” held historically, and you can carry it straight into the strategy builder.

3. Build a strategy

Strategies are defined with rules, not code. The builder walks through:

• Market & test window — market, timeframe, long or short, and the historical date range.
• Entry rules — one or more feature conditions, combined with all/any, plus the execution model that decides when signals are filled.
• Exit rules — maximum holding period, take profit, stop loss, and optional feature-based exits. Whichever fires first ends the trade.
• Position & costs — fixed or percent-of-equity sizing, starting equity, slippage, and entry/exit fees.

The strategy is validated before it runs; an unsupported combination is rejected, not guessed at.

4. Run a backtest

A backtest replays the strategy bar by bar over the selected window, using only data available at each decision. Runs execute in the background; Backtests lists every run with its status and headline results.

5. Interpret results

• Overview — total return, net PnL, max drawdown, trade count, win rate, and profit factor, with price/trade and equity/drawdown charts. Additional statistics, including Sharpe and Sortino, are one click away. Small samples (fewer than 30 trades) are flagged so ratios aren't over-read.
• Trades — every trade with its entry and exit time, price, fees, PnL, and exit reason, plus skipped signals and why they were skipped.
• Details — the strategy definition, execution assumptions, and the data, feature, and engine versions needed to reproduce the result.

6. Activate monitoring

A completed backtest that uses the scheduled daily execution model (and no take profit or stop loss) can be activated as an Active Strategy. After each UTC day's data is published, the strategy is evaluated on the completed daily bar and any entry or exit signal is recorded. Users can optionally enable informational SMS alerts — see the SMS opt-in flow and SMS terms.

Active Strategies never place brokerage orders and do not replay historical trades. Signals are research output; trading decisions remain yours.