TradingView Bar Replay for Indicator Validation

Start an indicator replay and record each decision

Add the indicator to a standard candlestick chart, then open Bar Replay from the top toolbar and choose a starting bar. Use Forward for one step or Play for automatic playback. Indicators and drawings can remain on the chart when you leave replay. The procedure below is a proposed audit workflow, not a report of measured trading results.

• Write down the exchange-qualified symbol, chart interval, session, timezone, script version, and inputs before starting.
• Choose the test dates and entry rules before viewing their outcomes. Leave enough earlier bars for the indicator to initialize.
• Advance one bar at a time near a signal. Record its first observable timestamp and the timestamp where its marker is plotted.
• Capture the chart before moving on. After the session, reload and compare the same timestamps for changed values or drawings.
• Keep rejected setups and ambiguous cases in the log. If you revise a rule, give it a new version and test a separate period.

Bar Replay, Replay Trading, and Strategy Tester answer different questions

Replay Trading adds simulated orders and performance reports to a replay session. TradingView documents market positions, limit, stop and stop-limit orders, bracket controls, and capital and commission settings. You can export its Performance and List of trades reports. This is a separate mode from Paper Trading; keep a copy of the session records rather than relying on them as a permanent journal.

• A plotted indicator arrow is not an order. Automated testing requires a strategy script with explicit order rules.
• Bar Magnifier uses lower-timeframe data for strategy fill simulation. It does not turn ordinary Bar Replay into a complete live tick history or remove lookahead mistakes.
• For a manual close-confirmed entry rule, state whether you assume a next-bar open fill. Record costs separately; there is no universal slippage allowance that fits every market.

Check the available replay history

TradingView reports that daily replay uses available daily history. For intraday time-based data, Essential starts with six months at one minute and Plus with one year; their windows scale with the interval in minutes. Premium and professional plans access the stored time-based history for the symbol, which can still be shorter than the requested period. Use the earliest available replay date to check your actual dataset.

• The current data-availability page lists Tick Replay on Ultimate for up to the most recent seven days. That is distinct from ordinary time-based bar stepping.
• TradingView help pages are not fully aligned on tick support: the general setup article still excludes tick-based charts. Check the dedicated data page and the Replay options available to your account.
• A missing earlier date can be a data or plan limit. Changing indicator code cannot create missing market history.

Inspect repainting mechanisms, not just the finished chart

Repainting includes differences between historical and realtime calculations; it is broader than deliberate use of future data. A delayed pivot can be valid if its confirmation delay is disclosed. Ordinary bar stepping cannot reconstruct every transient value that existed inside a historical candle, so a quiet replay is not a certificate of non-repainting behavior.

• The confirmed HTF value is intentionally delayed. Using close without the offset together with lookahead_on leaks future HTF values into historical bars.
• Using close[1] with lookahead_off has different historical update timing; it is not an interchangeable version of the documented confirmed-HTF recipe.
• This example addresses one data request. Other parts of a script can still use changing data or draw into the past.

//@version=6
indicator("Confirmed Daily Close for Replay", overlay = true)

if not timeframe.isintraday
    runtime.error("Use an intraday time-based chart below one day.")

// The expression offset and lookahead setting work together.
// This uses the previous completed daily bar, not the developing daily close.
previousDailyClose = request.security(syminfo.tickerid, "1D", close[1], lookahead = barmerge.lookahead_on)
plot(previousDailyClose, "Previous confirmed daily close", color = color.teal, linewidth = 2)

Use a signal marker that has one clear confirmation rule

This small EMA example places a marker only when a crossover is present on a confirmed chart bar. It has no simulated entries, stops, targets, or profitability claim. In replay, inspect the markers; after leaving replay, create an alert separately to check live delivery. Historical bars are confirmed, so the confirmation gate alone cannot demonstrate how an open live candle behaved.

//@version=6
indicator("Replay Signal Observation", overlay = true)

fastLength = input.int(9, "Fast EMA length", minval = 1)
slowLength = input.int(21, "Slow EMA length", minval = 1)
fastEma = ta.ema(close, fastLength)
slowEma = ta.ema(close, slowLength)

crossUp = ta.crossover(fastEma, slowEma)
crossDown = ta.crossunder(fastEma, slowEma)
buySignal = crossUp and barstate.isconfirmed
sellSignal = crossDown and barstate.isconfirmed

plot(fastEma, "Fast EMA", color = color.blue)
plot(slowEma, "Slow EMA", color = color.orange)
plotshape(buySignal, title = "Confirmed up cross", style = shape.triangleup, location = location.belowbar, color = color.teal, text = "UP", size = size.small)
plotshape(sellSignal, title = "Confirmed down cross", style = shape.triangledown, location = location.abovebar, color = color.maroon, text = "DN", size = size.small)

// These declare choices for a separately created live alert.
// Advancing replay does not send historical alert notifications.
alertcondition(buySignal, "Confirmed up cross", "Confirmed EMA up cross on {{ticker}} {{interval}}")
alertcondition(sellSignal, "Confirmed down cross", "Confirmed EMA down cross on {{ticker}} {{interval}}")

Keep an observation log before calculating performance

Record signal availability separately from the trade outcome. The fields below are a blank recording specification, not actual trades. Use timestamps as the main reference because bar_index can change when the loaded history changes. When a bar touches both a target and a stop, mark the order sequence as unresolved unless the available data and your fill model resolve it.

• Use a spreadsheet or journal for the observations. Pineify Trading Journal supports manual trade logging and CSV import; keep replay trades clearly identified as simulations.
• If you use a TradingView export, check its columns against the journal import format. This page does not establish a direct Replay Trading import integration.
• Turn specific mismatches into code requests: for example, plot a pivot at confirmation time or replace an unconfirmed daily filter with the previous daily value.
• Evaluate revised rules on a separate period. A fixed count such as 50 or 100 trades cannot, by itself, establish statistical reliability.