What causes "Unexpected history count: {history.Count}" in QuantConnect/Lean?

Ensure the backtest date range covers at least 10 trading days before the end date so History(10) can fill.

QuantConnect/Lean · error · RegressionTestException

Unexpected history count: {history.Count}

Error message

Unexpected history count: {history.Count}

What it means

Thrown when a History(symbol, 10) call for either SpxOption or Spx returns fewer or more than 10 daily bars. This verifies the Lean history service correctly retrieves and assembles the requested number of daily-resolution data points from the data store.

Source

Thrown at Algorithm.CSharp/BasicTemplateIndexDailyAlgorithm.cs:83

            AssertIndicators();

            if (Resolution != Resolution.Daily)
            {
                return;
            }

            var openInterest = Securities[SpxOption].Cache.GetAll<OpenInterest>();
            if (openInterest.Single().EndTime != new DateTime(2021, 1, 15, 15, 15, 0))
            {
                throw new ArgumentException($"Unexpected open interest time: {openInterest.Single().EndTime}");
            }

            foreach (var symbol in new[] { SpxOption, Spx })
            {
                var history = History(symbol, 10).ToList();
                if (history.Count != 10)
                {
                    throw new RegressionTestException($"Unexpected history count: {history.Count}");
                }
                if (history.Any(x => x.Time.TimeOfDay != new TimeSpan(8, 30, 0)))
                {
                    throw new RegressionTestException($"Unexpected history data start time");
                }
                if (history.Any(x => x.EndTime.TimeOfDay != new TimeSpan(15, 15, 0)))
                {
                    throw new RegressionTestException($"Unexpected history data end time");
                }
            }
        }

        /// <summary>
        /// This is used by the regression test system to indicate if the open source Lean repository has the required data to run this algorithm.
        /// </summary>
        public override bool CanRunLocally { get; } = true;

        /// <summary>

View on GitHub (pinned to d2c3659f87)

Solutions

Ensure the backtest date range covers at least 10 trading days before the end date so History(10) can fill.
Verify SPX and SpxOption daily data files exist and contain entries for all expected trading days.
Debug-log history.Select(x => x.Time).ToArray() to see exactly which dates are returned.
Check the HistoryProvider implementation for changes to lookback or bar-count resolution.

Example fix

// before
var history = History(symbol, 10).ToList();
if (history.Count != 10)
{
    throw new RegressionTestException($"Unexpected history count: {history.Count}");
}

// after — log missing dates for diagnosis
var history = History(symbol, 10).ToList();
if (history.Count != 10)
{
    var dates = string.Join(", ", history.Select(x => x.Time.ToString("yyyy-MM-dd")));
    throw new RegressionTestException($"Unexpected history count: {history.Count}. Dates: [{dates}]");
}

Defensive patterns

Strategy: validation

Validate before calling

// Validate history count with diagnostic logging
var history = History(symbol, 10).ToList();
if (history.Count != 10)
{
    var dates = history.Select(x => x.Time.ToString("yyyy-MM-dd"));
    Log($"History for {symbol}: count={history.Count}, dates=[{string.Join(", ", dates)}]");
}

Prevention

Ensure backtest window spans enough trading days for the history lookback.
Verify data files exist for every day in the lookback period.
Log history dates when counts mismatch to identify missing days.
Test history requests in isolation before adding assertions.

When it happens

Trigger: Data files for SPX or SPX option are missing days within the 10-bar lookback window, the history request period boundary is off-by-one, the algorithm ran for fewer than 10 days, or the data reader skips corrupt or sparse entries.

Common situations: Regression data set updates that removed specific days, history-provider changes that alter bar aggregation, sparse-data filtering that drops incomplete bars, or backtest window too short to supply 10 history bars.

Related errors

AI-assisted analysis of QuantConnect/Lean@d2c3659f87 (2026-08-13). Data as JSON: /api/errors/68b52c6aaba0a12b. Report an issue: GitHub.