i'm not talking about the duration in time, but the duration in bars.

I remember watching a video about a Neural Network trying to predict the next day close value. Of course it did a terrible job at it (because it's impossible to predict next day value based on past values) but I remember that results were worse the farther away it tried to predict (Day D + 2, Day D + 3, etc)

so my question is : is it harder to predict the outcome of a trade the longer the duration (in bars) ?
do we have to take that into consideration when designing trading systems ?

what is your experience ?

thanks

Jeff

Sorry for being this guy but...

If you are trying to predict where the Market is going while trading then you are doing it wrong.

Prediciting means you are bringing expectactions with you going into a trade, having expectations in a mostly random Market is futile.

The only thing you need to ask yourself is..

Does the setup fill all my requirements to take the trade?

Yes? take the trade.
No? skip the setup.

Wether the Market after taking a trade goes into your favourable direction or not has nothing to do with the setup.

Thats why its typically said that the income of a trade doesnt determine if a trade was good or bad. As long you have followed your trading plan a trade is a good trade.

So to answer the question: both on HTF and LTF its impossible to predict the Market.

PumbaPLS wrote: Fri Jun 03, 2022 10:32 pm Sorry for being this guy but...

If you are trying to predict where the Market is going while trading then you are doing it wrong.

Prediciting means you are bringing expectactions with you going into a trade, having expectations in a mostly random Market is futile.

The only thing you need to ask yourself is..

Does the setup fill all my requirements to take the trade?

Yes? take the trade.
No? skip the setup.

Wether the Market after taking a trade goes into your favourable direction or not has nothing to do with the setup.

Thats why its typically said that the income of a trade doesnt determine if a trade was good or bad. As long you have followed your trading plan a trade is a good trade.

So to answer the question: both on HTF and LTF its impossible to predict the Market.

well sorry for being this guy but, any successful trading system predicts and has an edge over the market, an edge that is quantifiable. that's how you make money, you successfully predict the direction of a trend, let say at 58% and then close it in profit more often than not.
the question is if your system has more short trades than long trade will it be intrinsically more profitable ? (i'm not talking about bulls and bears of course, but the duration of the trade in bars)

Each trader opens any deal only when he has a forecast.
Without a forecast, even one transaction will not open, because how will you open it when you do not have a forecast for further price movement?
The difficulty of forecasting is always the same for all types of transactions, at shorter distances and at longer ones.
There is no difference in timeframes how your forecasts will develop, at any distances and timeframes they will always be the same in execution. For this is your trading system, your views on the bars and your expectations and forecasts.
On minute or daily timeframes, your predictions will always be the same.
The only difference is
that on the minute charts you will have a negative mathematical expectation, on the daily charts you will have a positive mathematical expectation.
As a result, those who try to open transactions within the day will remain at a loss, those who open on the medium-term trends of the weekly and daily charts will eventually make a profit.
This will happen due to the laws of mathematics, because such is the mathematics of the market and money.

ionone wrote: Sat Jun 04, 2022 12:04 am well sorry for being this guy but, any successful trading system predicts and has an edge over the market, an edge that is quantifiable. that's how you make money, you successfully predict the direction of a trend, let say at 58% and then close it in profit more often than not.
the question is if your system has more short trades than long trade will it be intrinsically more profitable ? (i'm not talking about bulls and bears of course, but the duration of the trade in bars)

I missunderstood your intial asked question then.

if you take a very simple virtual asset that can only go up and down by one increment each step.
at the first step, you have a 33% chance to predict the outcome ( -1, 0, +1)
at the second step, it drops to 20 % ( -2, -1, 0, 1, 2)
and the third "bar" it's now 14.28 %

and so forth

so I think the shorter trades the easier it is to predict

am i wrong thinking that way ?

sure my example is simplistic. But when you place a trade, you enter a game of prediction where you try to predict the outcome of a price movement.
so I think i'm not totally wrong here

Jeff

I view price as fractal and that it moves the same on all time frames, just at different speeds. With that in mind I always just look for areas of liquidity I feel the markets will reach for and assess current price position in the range to that point. As such it wouldn't represent a "harder trade" per se, other than the various difficulties people face with patience on HTF vs LTF trading.

as another image : if you have a shooting target that is at a distance of 1 meter or one that is at 100 meters, which one will be the easier to shoot?

Also let's set your TP to 1 tick (shortest distance possible), what winrate will that be ? (i'm not saying it can be a profitable system)
I think something like 99.99% (99.99% of the time, price will go up 0.1 pip from the opening price)

so shortest trades seems to be easier to predict. Doesn't mean it's a system by itself (like a R:R with a TP two times the risk doesn't change the odds in your favour by itself). Other factors come in term.

Jeff