• Recent
  • Categories
  • 0 Unread 0
  • Popular
  • Tags
  • More
    • Guidelines
    • Disclaimer
  • Users
Skins
  • Light
  • Default
  • Cerulean
  • Cosmo
  • Flatly
  • Journal
  • Litera
  • Lumen
  • Lux
  • Materia
  • Minty
  • Morph
  • Pulse
  • Quartz
  • Sandstone
  • Simplex
  • Sketchy
  • Spacelab
  • United
  • Yeti
  • Zephyr
  • Dark
  • Cyborg
  • Darkly
  • Slate
  • Solar
  • Superhero
  • Vapor
Collapse
Definedge Forum Logo

Inconsistent Backtest Results

Scheduled Pinned Locked Moved Momentify
@prashantshah@askdefinedge@definedge
7 Posts 4 Posters 205 Views
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
Reply
  • Reply as topic
Log in to reply
This topic has been deleted. Only users with topic management privileges can see it.
  • G Offline
    G Offline
    G Heble
    wrote on last edited by Gautam Heble-1756230857752
    #1

    I'm running into a consistency issue with my backtesting that's making it hard to trust the results.

    I've been backtesting a weekly strategy, tracking all my parameters (index, backtest period, filters, market trend filter, etc.) in a spreadsheet along with results, screenshots, and CSV file export.

    The problem is when I rerun the exact same backtest a few days or weeks later using identical parameters, I'm getting completely different results.

    This is a problem because if backtest results re not reproducible, then I cannot rely on them.

    For example, I tested a specific parameters set a couple of weeks ago and got the below results.
    2025-10-18 13_32_48-.jpg

    When I ran the same backtest parameters again today, I got a completely different result!
    54df918b-5b54-4c0b-a199-d7b2f5c89c6c-image.png

    R 1 Reply Last reply
    0
  • A Offline
    A Offline
    Alpha Trader
    wrote on last edited by
    #2

    This happens due to split,bonus and other corporate actions, if you see for longer horizon it won't affect much

    recently I bought 9 shares as per system, but after split, I done backtesting and it showed 10 shares

    so minor differnce could be there, but in long run it average out

    1 Reply Last reply
    0
  • G Offline
    G Offline
    G Heble
    wrote on last edited by
    #3

    I doubt thats it, because these backtests were executed within a span of 2 weeks. So I can expect minor differences, but over a backtest period of 9 years, the huge difference is concerning.

    1 Reply Last reply
    0
  • R Offline
    R Offline
    Rajendra Manke

    Pro User

    replied to G Heble on last edited by
    #4

    @G Heble
    Try to post images of all the screens used in the back testing for better understanding. A small variation in a parameter can give different results.

    G 1 Reply Last reply
    0
  • G Offline
    G Offline
    G Heble
    replied to Rajendra Manke on last edited by
    #5

    @Rajendra Manke At this point I don't have screenshots of the criteria, just my notes on each and both the criteria notes have the same parameter values.

    1 Reply Last reply
    0
  • Y Offline
    Y Offline
    Yogesh Kumar
    wrote on last edited by
    #6

    I have also observed the difference between back testing and real portfolio through momentify. Will post with images.

    G 1 Reply Last reply
    1
  • G Offline
    G Offline
    G Heble
    replied to Yogesh Kumar on last edited by
    #7

    @Yogesh Kumar Yes, I saw your post as well, thanks for testing it against your real portfolio.

    I am now struggling to get convinced to invest in the platform more until there's clarity around this aspect and the support team has not been responding.

    1 Reply Last reply
    0


Recent Posts

  • G
    Inconsistent Backtest Results
  • Y
    Wrong enty exit and differences in Momentify Portfolio and RZone backtest
  • B
    Creation of Own index
  • D
    Swing Trading
  • N
    🚀 Nifty Outlook — Trend Intact. Retest Normal. Opportunity Alive.
  • Login

  • Login or register to search.
  • Blog
  • Youtube Videos
  • Refresh
  • First post
    Last post
0
  • Recent
  • Categories
  • Unread 0
  • Popular
  • Tags
  • More
    • Guidelines
    • Disclaimer
  • Users
  • Login

  • Login or register to search.
  • Blog
  • Youtube Videos
  • Refresh