How to Tell if a TradingView Backtest Is Reliable: 7 Checks
A TradingView backtest is reliable when it survives checks the Strategy Tester does not run for you: realistic fees and slippage, enough trades, no look-ahead, a period the strategy has never seen, and a second engine that reproduces the same trades. The number in the Strategy Tester tab is an exact calculation on assumptions; reliability depends on the assumptions.
Updated 17 October 2026. Research on past data, not investment advice.
The checklist in one table
| # | Question | Where to look in TradingView | Warning sign |
|---|---|---|---|
| 1 | Are fees declared? | strategy(commission_type=…, commission_value=…) or the Properties tab |
Commission at 0 |
| 2 | Is slippage declared? | slippage parameter (in ticks) |
Default value: 0 ticks |
| 3 | Are there enough trades? | List of trades | Fewer than about thirty trades |
| 4 | Does the script read the future? | request.security, lookahead, execution settings |
lookahead_on, a warning banner in the report |
| 5 | Was the test period used for tuning? | Your tuning history | Parameters chosen on the whole displayed period |
| 6 | Does it beat buy-and-hold? | "Buy & hold return" in the report | Positive profit, but below buy-and-hold |
| 7 | Does another engine reproduce the trades? | Exported list of trades | Nobody replayed it |
1 and 2. Are fees and slippage the ones you will pay?
TradingView's Pine documentation states that slippage is expressed as a fixed number of ticks and that its default value is 0. A backtest without costs assumes perfect fills.
The effect is measurable. In our RSI-below-30 study on BTC (67,861 one-hour Binance candles, 2019-2026), the best gross cell was +0.027% per signal; after 0.20% round-trip fees it became −0.173%. How to set costs: fees and slippage in the TradingView Strategy Tester.
3. Are there enough trades to conclude?
A result built on 7 trades says almost nothing. The Lab's judge requires at least 30 independent events in the test period before it can return "holds". That threshold eliminates famous rules: Faber's 10-month rule produced only 11 trades and 1 independent event over 2012-2026 in our 1,306-variant test. It was not wrong; it was unjudgeable.
4. Does the script read the future?
request.securitywithlookahead_onmisused: the script reads a higher-timeframe close before it exists. The Lab's Pine exports explicitly writelookahead_off.- Execution settings. TradingView's documentation warns that changing script execution settings "can cause repainting", and calls misleading any behaviour that assumes knowing a bar's high at its opening tick; the report then shows a warning banner.
process_orders_on_close = true: the order fills at the close of the bar that created it, instead of the next open (the default). Sometimes intended, but it is an assumption to own.
5. Was the displayed period used for tuning?
If you picked your parameters while looking at the result over the whole period, the displayed result is a training score, not a test. Our GPU found a strategy at +172.7% on its training period that made −37% on unseen data (post-mortem). Step by step: test a Pine strategy out of sample for free.
6. Does the strategy beat buy-and-hold?
Making 40% while the asset made 90% is not a head start. The TradingView report shows the buy-and-hold return: always compare. Across 9 markets, none of our 25 strategies beat buy-and-hold in a bull market over one year (World Arena).
7. Does a second engine reproduce the same trades?
A backtest becomes a measurement when another engine finds the same trades. On 13 October 2026 we pasted Lab Pine exports into a real TradingView account (Basic plan) and compared every trade: 258 of 258, 97 of 97 and 26 of 26 identical, final capital identical to the cent (full method). The same check found a defect on our side: a stop order that crashed the Pine script in TradingView, since fixed (62 of 62 trades identical).
Reproducible does not mean profitable: the 258-trade export lost 42.45%. The reproducibility check comes before the verdict.
Run the checklist on your strategy
- Test your Pine strategy for free with the Quick Check: script compatibility and a buy-and-hold comparison, fees included.
- The free Lab for Windows: sealed period, counted trials, local verdict, Pine export with a receipt.
- Check the trades in TradingView: paste the Pine export into the editor and compare the trade list. Open TradingView (partner link); a free account is enough.
FAQ
Is the TradingView Strategy Tester reliable?
The calculation is exact for the declared assumptions. Our exports replayed in TradingView gave the same trades as our engine to the cent. The reliability of the result depends on fees, slippage, look-ahead and the chosen period, all set by the author.
How many trades does a backtest need?
There is no universal threshold. The Lab's judge requires at least 30 independent events in the test period; below that it returns "inconclusive".
Is a backtest that beats buy-and-hold reliable?
Not necessarily: if it was tuned on the same period, it is a training score. Only a period set aside before any tuning gives a test.
Do I need a paid plan to check a backtest?
Not for the checks in this list: our exports were verified on a Basic account. Paid plans give more history and Deep Backtesting (see which TradingView plan for backtesting).
Sources
- TradingView Pine Script documentation, "Strategies" page (slippage in ticks, default 0;
process_orders_on_close; warnings on execution settings), accessed 17 October 2026. - Strategy Arena, trade-by-trade reproducibility, 13 October 2026.
- Strategy Arena, RSI below 30, fees included, 13 October 2026.
- Strategy Arena, 1,306 variants judged, 13 October 2026.
Disclosure: Strategy Arena takes part in the TradingView partner program. If you create a new account through our links and then subscribe on the web, we may earn a commission at no extra cost to you. It changes no number and no verdict in this article.
Research on past data, not investment advice. Past results do not guarantee future results.
⚠️ Disclaimer — This article is for informational and educational purposes only. It does not constitute investment advice or a buy/sell recommendation. Past performance does not guarantee future results. Strategy Arena is an educational simulator with virtual capital. Always do your own research before making investment decisions.