Quiz: backtesting

Check your understanding of sample size, out-of-sample testing, overfitting and bias.

All courses · Lesson 6 of 6 · Advanced · 4 min

  1. A strategy shows a 50% win rate over 100 trades. What is the standard error of that win rate?

    • 0.5%
    • 5.0%
    • 10.0%
    • 2.5%
  2. The R results of single trades have a standard deviation of 1.5R. What is the approximate standard error of the expectancy after 225 trades?

    • 0.01R
    • 0.067R
    • 0.10R
    • 0.15R
  3. Which approach gives the most honest check of a finished strategy?

    • Re-run the optimiser until the equity curve looks smooth
    • Test on the same data you tuned on, but with more trades
    • Test on a different instrument chosen after seeing which one worked
    • Test once on a later period that was never used while building the rules
  4. Why is the result of the analysis page optimiser "Find the best stop loss and take profit" likely optimistic?

    • It picks the best setting on the same data it measures, so the result is in-sample
    • It ignores the stop loss entirely
    • It only uses trades from the most recent year
    • It always reports the average of all settings
  5. Which of these is an example of look-ahead bias?

    • Using a 20-period moving average instead of a 50-period one
    • Deciding to enter at a candle's open based on that candle's close
    • Risking 1% of the account on each trade
    • Testing on 5 years of history instead of 10
  6. A backtest has a gross expectancy of +0.12R per trade. The spread costs 0.07R per trade. What is the net expectancy?

    • +0.19R
    • +0.12R
    • +0.05R
    • −0.05R

Sign in to save your progress and quiz results.