How replay works
The mental model behind replay — what each step carries, what speed changes, and why fills stay precise.
StrategyTune's replay is built around one idea: show you the market the way it actually arrived, not a sanitized version of it. Every price you see is a real recorded quote, played back in the order it happened.
The mental model
Replay advances in small time steps — sub-second steps at 1×. Each step carries:
- The latest bid and ask quote.
- The forming bar's OHLC so far — so the chart's live candle updates on essentially every step, not only when a bar closes.
Between two candle closes you therefore see many price updates — each one moves the price forward, repaints the forming bar, and updates your unrealized P&L, so the chart looks the way it would have looked in real time.
Speed coarsens the picture, never the fills. At higher speeds each step covers more market time, so you see fewer of the recorded quotes — but nothing is synthesized or interpolated. Pending orders — limits, stops, take-profits, stop-losses — are always evaluated against the full-resolution recorded data, so a stop fills at the same price at 50,000× as at 1×.
Order behavior — market vs. pending, and what triggers fills on each instrument type — is covered in Trading & P&L.
One quirk worth knowing: on stock-type instruments the candles are built from recorded trades, so the price can flicker without the candle changing — a quote arrived, but no trade printed. On quoted instruments like forex, every step updates the forming candle.
Manual and automated backtesting
Everything above applies to both kinds of backtest, because both run over the same recorded market data. Manual replay runs in your browser as you watch — you drive the clock and place the trades; automated backtesting runs your Strategy script over the same data on StrategyTune's cloud — see Automated backtesting. The fill rules are identical in both modes — see Trading & P&L.
Something missing or wrong? Email support@strategytune.com.