Day trading backtest: Auto-entry on the initial trend signal

Discussion in 'ETFs' started by ninZa.co, Aug 23, 2025 at 4:54 AM.

  1. ninZa.co

    ninZa.co Sponsor

    Aloha traders!

    $5,470 profit in just 5 days… and nearly doubled in 20.

    We’ve recently backtested a fully automated strategy with a single focus:
    Jump in at the very first signal of a trend, then let automation handle everything else.

    It’s a clean, straightforward approach – and the results were better than expected.

    [​IMG]

    Inside, we’ll show you the exact setup so you can start running it right away.

    No babysitting charts, no second-guessing – just smooth, hands-off execution.

    If you’ve been searching for a day trading method that captures the earliest momentum for bigger profit potential…

    This strategy deserves a closer look.
     
  2. MarkBrown

    MarkBrown

    suggestions for improvement

    profit factor is everything a good benchmark is 1.75

    $5470 profit with a $3320 drawdown is a 60.69% drawdown - 8% is a good benchmark

    56 trades total divided by 14 trades a day means your testing period was only 5 days

    a system should be tested with as much history as possible not 5 days
     
    TrailerParkTed and SunTrader like this.
  3. Absolute minimum statistically is 100 events. Everything works until it doesn't
     
    MarkBrown likes this.
  4. Businessman

    Businessman

    100 events is not enough either.

    More like 10 years worth. And even that might not be enough..
     
    Last edited: Aug 23, 2025 at 9:35 AM
  5. Businessman

    Businessman


    Percent drawdown is meaning less over such a small sample size.

    Metrics like the MAR ratio are calculated over a year.

    Even then the user sets the target drawdown % by how much how much money they have allocated to the system.
     
    Last edited: Aug 23, 2025 at 9:55 AM
  6. MarkBrown

    MarkBrown

    ok mr business man now systems expert lol
     
  7. SunTrader

    SunTrader

    So @ 10 trades per day, 250 trading days per year X 10 years = 25,0000 fricken' trades? :confused:

    It's a day trading system, that is unnecessary overkill. Yeah I know markets change (up/down/sideways) over time. Still.
     
    TrailerParkTed and MarkBrown like this.
  8. MarkBrown

    MarkBrown

    yea the commission will eat you alive
     
  9. MarkBrown

    MarkBrown

    its not a matter of days at all it's a matter of data points "how many data points" that you test on no matter if it's tick data or daily data it's still always read above.

    mr "businessman" know it (all) very little
     
  10. Businessman

    Businessman

    Really you want the back test to cover slow and fast markets, bull and bear markets, all the extremes.

    If I trade things like stock index futures, I would want to test in extreme market conditions like 2008 and really slow market where the VIX fella below 10 for a while, which hasn't happened in a while.

    You could go and just pick out those periods in isolation, but you might as well just gather 20 years of data and see how the system did.