Best ratio smaller TF / bigger TF

Discussion in 'Trading' started by Sekiyo, Oct 19, 2024.

Best ratio Smaller TF / Bigger TF

  1. 1/2

    1 vote(s)
    20.0%
  2. 1/3

    3 vote(s)
    60.0%
  3. 1/4

    1 vote(s)
    20.0%
  4. 1/5

    0 vote(s)
    0.0%
  1. Sekiyo

    Sekiyo

    What do you think is the best ratio regarding the Smaller TimeFrame (Entry / Exit) to the Bigger TimeFrame (Context, Trend, S&R).

    I know the 1min / 5min is heavily used but the 5min ATR isn’t 5x the 1min ATR …

    I am especially talking about Range, Tick charts and Co.

    I used 1/2 but found 1/4 to be more interesting ?

    Voting for 1/4

    Please don’t tell me 3.14 :banghead:
    Or maybe I should try :D
     
    Last edited: Oct 19, 2024
    birdman and HawaiianIceberg like this.

  2. No - of course not: the relationship between timeframe and volatility is a square-root one.

    The 4-minute ATR is double the 1-minute ATR because two squared is four. Roughly double, anyway - if it's a liquid market without sudden news or other complications.

    The 4-hour ATR is double the 1-hour ATR in the same way, and that's a better example. ;)
     
    MACD, HawaiianIceberg and Sekiyo like this.
  3. Sekiyo

    Sekiyo

    Very interesting. Thanks for sharing.
    So the ratio btw 1min and 5min is … ~2.24
     
    Probability likes this.
  4. At one time (when I used to trade from much faster charts than I would, now!), I used to trade a highly liquid instrument using 12-minute charts for directional bias and 2-minute charts for the entries.

    The difference in ATR between the two charts was always a factor of just under 2.5, as you'd expect, because 12/2 = 6, and the square root of 6 is just under 2.5.
     
    HawaiianIceberg and Sekiyo like this.
  5. Sekiyo

    Sekiyo

    Traders ain’t Psycho for nothing :p

    So if you had to vote … you’d pick 1/2.

    Thanks for sharing your experience.
    Refreshing compared to chatGPT!
     
    Probability likes this.
  6. The time variable might be misleading you, that is a flaw of most indicators. We tend to separate data in chunks of time (X) simply because it is easy to understand.

    There are other points of view that remove time from the data series.

    https://pippenguin.net/trading/learn-trading/tick-chart-in-trading/

    That will give you a much closer view to volatility. It is pure action grouped in number of ticks. Then a candle might take hours or minutes to complete depending on the activity. Time is random.
     
    Sekiyo and Probability like this.

  7. In my opinion, almost everyone trading from tick charts is (knowingly or unknowingly - but very often the latter) using them as an approximate substitute for volume-bars/candles, and would actually be better off just trading from those instead.

    I no longer use timed bars at all (but the approximate “time-equivalents” of the charts I trade from now are quite a bit slower than in the example I mentioned above).

    Personally, I strongly prefer range-bars and/or volume-bars.

    These both seem to me to be such an improved approach, compared with timed bars, that I don’t really understand why more people aren’t using them.

    My (pretty strong) impression, anyway, is that almost nobody who switches from timed bars to volume-bars and/or range-bars ever switches back. :)
     
  8. The never ending sentence that we see everywhere about retail trading being a pool of losers, is mostly because the wrong tools are promoted.

    Time series provide a lot more signals, which leads to more executions, ending in more commissions for brokers.

    People do not use other indicators because they don't know they exist.

    We see a lot of indicators with a time axis, but you have to digg and look specifically for the ones that don't use time as a variable.
     
    Probability likes this.
  9. I had the best luck with a 10 / 60 minute timeframe, before I got tired of daytrading. Shorter timeframes always put in patterns that would get suddenly obliterated by something else that was forming in the bigger picture.
     
    Sekiyo likes this.
  10. oshjdf

    oshjdf

    Not necessarily true. I went back from data-based (tick, volume, range) to time-based. My stats and execution are better on time-based. It all depends on the methodology of the trading system. The best thing using time-based is more robust. If you have great result from either data-based or time-based, who cares.
     
    #10     Oct 20, 2024