TradingView's Strategy Tester gives you the main figures from a backtest, but those headline numbers do not always tell you why a strategy performed the way it did. A positive net profit or acceptable win rate can hide long drawdowns, dependence on a few unusually good trades or large differences between long and short performance.
TradingView Xen can now analyse the actual trades from a Strategy Tester backtest. Export the List of Trades from TradingView as a CSV file, upload it to Xen, and it will build a detailed report from the completed trades.
£5 Free Credit for New Users
New users currently receive £5 of free Xen AI credit. Xen uses pay-as-you-go pricing, so you only purchase additional credit when required. Try Xen AI for TradingView
Exporting TradingView backtest results
Start in the TradingView Strategy Tester. Run the strategy normally and open the List of Trades, then export the results as a CSV file. You do not need to provide the Pine Script or rebuild the strategy inside Xen. The analysis works from the trade history, so you can use it with strategies you wrote yourself, purchased from another developer, or created with another coding tool.
Once you upload the file, Xen checks the TradingView export and converts entry and exit records into completed trades before running the analysis. This matters because the raw TradingView CSV can contain multiple rows for the same trade.
Net profit and win rate
It is easy to focus on net profit when reviewing a backtest, but two strategies with the same profit can have very different risk profiles. One may produce relatively steady results, while the other may spend long periods underwater and depend on a handful of large winners.
The Xen report can examine metrics such as profit factor, average trade, payoff ratio, maximum drawdown, winning and losing streaks and the difference between long and short trades. It can also look at favourable and adverse excursions when that information is available in the export. This gives you a better picture of how the strategy actually generated its result rather than treating one headline figure as evidence that the system is ready to trade.
One of the more useful checks is trade concentration. A strategy can finish a backtest in profit even when most of its performance came from a very small number of exceptional trades. The report can highlight unusually large winners and suggest checking what happens when you remove the strongest trades from the results. If profitability disappears after removing one or two trades, that tells you something very different from a strategy whose performance is spread across hundreds of trades.
The same principle applies to historical periods. Strong results from one or two years may mask weak performance under other market conditions.
Commission and slippage matter
Backtests are particularly easy to overestimate when trading costs aren't configured correctly. A strategy with a small average profit per trade can look viable before costs but deteriorate quickly once commission, spread and slippage are included.
Xen can flag missing or unrealistic execution assumptions in the supplied results, but it cannot invent the correct values for your broker or market. You still need to configure those settings properly in TradingView and test again.
Use the report to decide what to test next
The useful part of a backtest review is not simply producing another set of statistics. It is finding something worth investigating.
Depending on the results, Xen may suggest testing the strategy on unseen data, comparing different market periods, changing nearby parameter values or checking performance on other symbols. It can also suggest separating long and short results or testing what happens when you exclude unusually profitable trades.
Are these tests designed to answer a more useful question: is the strategy showing reasonably stable behaviour, or did one particular set of historical conditions make the backtest look better than it really is? If the analysis identifies a possible weakness, it is usually better to make one controlled change than to immediately redesign the strategy.
For example, if the report shows that short trades consistently perform worse than long trades, you could investigate the short-side entry logic separately. If drawdown is concentrated in certain conditions, you could test a filter designed specifically for those periods. After making the change, run the TradingView backtest again and compare the new results. This makes it easier to see whether the modification genuinely improved the strategy, rather than simply changing several variables at once.
Backtest analysis is still only part of the process
AI can make a TradingView backtest easier to investigate, but it does not turn historical results into a prediction of future performance. A report is only as reliable as the data and assumptions you supply. Commission, slippage, timeframe, position sizing, chart type, date range and TradingView's simulated order execution all affect the result. The purpose of the analysis is to expose useful evidence and possible weaknesses so you can decide what needs further testing.
