Question 1

Does the backtester include trading fees and slippage?

Accepted Answer

Yes, and it is on by default. Every run is reported net of commission, slippage and (for crypto perpetuals) funding, because a strategy taking 200 round trips a year at 0.04% per side burns about 16% of capital annually - enough to turn a reported winner into a real-world loser. You pick a venue fee schedule or type your own numbers, and the result panel shows profit before costs, each cost line, and a warning when costs alone flipped the outcome. The parameter optimizer is charged the same costs, so it cannot rank a high-frequency variant first just because fees were ignored.

Question 2

Can I check whether my optimized parameters still work out of sample?

Accepted Answer

Yes - walk-forward validation is built in and free. A parameter grid reports the best of up to 60 variants measured on one window, and it picks that variant because it scored highest on exactly that data, so the headline is an upper bound rather than an estimate. Walk-forward splits your date range into 2-6 anchored folds, selects parameters on the earlier bars of each fold, then scores them on the bars the selection never saw. You get a per-fold pass/fail table, the in-sample versus out-of-sample average, how much of the edge survived, and the parameter set the folds actually agreed on - which is what is offered as the recommendation, never the single luckiest fold. When no fold produced enough out-of-sample trades, it says so instead of inventing a verdict.

Question 3

Can I backtest a volume condition or a custom RSI level?

Accepted Answer

Yes, both, without writing code. RSI thresholds are yours to set per condition rather than fixed at 30/70, and you can pick a level test ("RSI below 25") or a cross ("RSI crosses up through 25") - different signals that behave very differently, since a level stays true for every bar of an oversold stretch while a cross fires once. Volume adds two more: above its rolling average, or a spike of N times that average, with the lookback window yours to choose. The average is built from bars strictly BEFORE the one being tested, so a spike never inflates the baseline it has to beat. Conditions combine with AND, so "RSI crosses up through 30 AND volume is twice its 20-bar average" is a single strategy. Every one of these parameters is also sweepable by the optimizer and checkable by walk-forward validation. If a market's candles carry no volume - spot forex has no central exchange reporting it - the backtester says so instead of returning a run with no trades.

Question 4

Is TraderAI's strategy backtester free?

Accepted Answer

Yes. You build a strategy visually (entry/exit conditions from RSI, moving averages, MACD, Bollinger Bands and more), run it on real historical candles for crypto, forex and US stocks, and get win rate, profit factor, max drawdown, Sharpe and Sortino - all on the free plan. The builder is open without an account; running a backtest just needs a free sign-up.

Question 5

Can I backtest a strategy without coding?

Accepted Answer

Yes - two no-code paths. Describe the strategy in plain language (English or Vietnamese), like "buy when RSI drops below 30 and price is above the 200 MA, sell at 2R", and the AI converts it into a runnable strategy. Or paste a TradingView Pine Script and the AI converts it into the backtester's format. You can still fine-tune every condition visually afterwards.

Question 6

What is the strategy optimizer and is it free?

Accepted Answer

The optimizer grid-searches one or two strategy parameters (for example the RSI threshold and the stop-loss size) across up to 60 combinations on the same historical data, then ranks every run and lets you apply the best settings back into the builder with one click. It's free and nothing is stored - it's an exploration tool, so treat the best result as a hypothesis to validate, not a guarantee.

Question 7

Does a good backtest guarantee future profits?

Accepted Answer

No. A backtest shows how a rule set behaved on past data; markets change, and over-optimized parameters often fail forward. That's why TraderAI shows profit factor, max drawdown, Sharpe and Sortino alongside win rate, and why the optimizer is capped at two parameters - to make over-fitting harder. Validate any promising strategy on a different period or symbol before risking real money.

Question 8

Can I backtest RSI divergence with a stop at the previous swing low?

Accepted Answer

Yes. Bullish and bearish RSI divergence are built-in entry signals, detected on confirmed swing pivots: a swing is only known once a set number of candles has closed beside it, so the signal fires on that confirmation candle and never uses future prices. The stop-loss can be placed beyond the previous swing low (long) or high (short) of a chosen window, and the target set as a multiple of that risk, such as 2R. If price has already moved through the swing level, the trade is skipped rather than taken without a stop, and the number of skipped signals is reported.

Question 9

Can I get an alert or run a forward test from a strategy I backtested?

Accepted Answer

Yes. From a completed backtest you can create a Smart Alert that evaluates the strategy's own entry rule on every closed candle and notifies you in-app, by email or Telegram when it fires, or start a forward test that freezes the strategy and replays it on each new closed candle as paper trading. Neither ever places an order. Forward-test results are recomputed from closed candles each time you open them rather than stored, so they cannot drift from the market data they describe, and every trade shown was opened after the strategy was frozen.

Question 10

Does signal-weighted (smart) DCA beat plain DCA?

Accepted Answer

Not reliably, and the Smart DCA simulator is built to show that honestly. It buys plain DCA and a version that scales each buy by RSI or by the distance from a moving average on the same scheduled days, at the same next-session fill prices and with the same fees, then compares return on invested capital, average cost and money-weighted return. When the multipliers make the result worse, it says so. You can then get a weekly or monthly reminder of this week's multiplier; you place the order yourself - TraderAI never trades for you.

Related tools

Strategy Backtester

Strategy Backtester