Runtime Errors

Runtime errors happen after your code passes validation but fails during execution. The backtest stops and the error is reported with a stack trace.

TypeError: Cannot read properties of undefined

TypeError: Cannot read properties of undefined (reading '14')

Common cause: Accessing an indicator that was not declared in init().

function init(ctx) {
  ctx.indicator('sma', 'SMA', { period: 20 });
}

function onBar(ctx, i) {
  // Wrong — 'ema' was never declared
  const val = ctx.ind.ema[i];  // TypeError: ctx.ind.ema is undefined

  // Correct — use the name from init()
  const val = ctx.ind.sma[i];
}

ReferenceError: variable is not defined

ReferenceError: threshold is not defined

Common cause: Using a parameter name directly instead of ctx.p.name.

// Wrong — bare variable
if (ctx.ind.rsi[i] < threshold) { }

// Correct — access via ctx.p
if (ctx.ind.rsi[i] < ctx.p.threshold) { }

NaN Results

Your strategy runs but produces no trades or strange values.

Common cause: Using an indicator before it has enough data.

function onBar(ctx, i) {
  // An SMA(20) is NaN for the first 19 bars
  // Always guard against NaN
  if (q.isNaN(ctx.ind.sma[i])) return;

  // Now safe to use
  if (ctx.series.close[i] > ctx.ind.sma[i]) {
    ctx.order.market('ASSET', 1);
  }
}

No Trades Generated

Your strategy runs without errors but produces zero trades.

Checklist:

1. Are conditions too strict? — Print intermediate values or simplify the logic
2. Is the guard too broad? — If you return early on NaN, make sure you only check the indicators that matter
3. Wrong signal direction? — In LONG_ONLY mode, short signals are ignored
4. Already in position? — Check if you're trying to enter when already holding
5. Date range too short? — Ensure enough bars for indicators to warm up

Dry Run Validation

When you save a strategy, the platform runs a dry run — a short execution against a small sample of candles. If the dry run crashes, you will see the runtime error before submitting a full backtest. Fix the error shown in the dry run output.

Debugging Tips

• Start with a minimal strategy and add complexity gradually
• Use ctx.state to track intermediate values you can inspect in trade metadata
• Add signal metadata: ctx.order.market('ASSET', 1, { signal: 'buy', rsi: ctx.ind.rsi[i] })
• Test with a narrow date range first to iterate quickly