Slipping in the rain past the drips - page 3

 
Konstantin Yartsev:

It all depends on the size of the deposit and the profit plan. ....

How can you even plan for profit in forex. OK, I can understand dividends on purchased stocks, but profits from speculation on forex ....

 
Aleksey Mavrin:

I hope you understand that it depends entirely on your planned Stops, or rather your planned series of Stops to endure), i.e. on the TC? And this has direct relevance to maxDD in the next thread ;))

No one knows the future series of Stops and the loss from it.

It can only be checked very, very conventionally on history (in the tester). And only with a 100% quality history. In the MT5 tester (I don't even take MT4 - there is no quotes from your broker there - only "average" quotes from MKL) - it is only 1 previous month. The rest of the quote, the so called "99% history quality" has very little to do with reality, also because of the mechanism of synthetic tick generation by the tester.

 
Vitalii Ananev:

How can you even plan the size of profits in forex. OK, I understand that dividends on purchased shares can somehow be predicted, but profits from speculation on forex ....

Everything has to be planned. And everyone decides for himself how to do it.

If we look at my strategy, it's medium term strategy based on strong signals on H4 timeframe, when the price reaches TP with high probability. There are about 5-7 profitable signals during the week. Trading is diversified on 28 currency pairs, of which the robot chooses one with the highest potential for movement in the desired direction. There can be 3 pairs in the market at the same time. The deposit is calculated to be enough to open e.g. three cheap pairs (e.g. Novozel-Franc) or two expensive pairs (e.g. Pound-Newzbel) at the same time. I check the trade in the past 30 days with 100% quality history. What was earlier (99% quality) is of no interest. Last month it was 146%. And dividing the account, for example, 10000 dollars in half: the profit plan for 5000 - 146%. Ie 7300 dollars. In this case, the robot will calculate everything itself: you only need a leverage of 1:200, a deposit of $15 and an account with zero indentation (ECN, NDD, etc.).

 
Konstantin Yartsev:

Everything has to be planned. And how - everyone decides for himself.

If you look at my strategy, it is a medium term on strong signals from H4, where the price has to go to the TP. In a week there are about 5-7 profitable signals. Trade is diversified on 28 currency pairs, of which the robot chooses one with the highest potential for movement in the desired direction. There can be 3 pairs in the market at the same time. The deposit is calculated to be enough to open e.g. three cheap pairs (e.g. Novozel-Franc) or two expensive pairs (e.g. Pound-Newzbel) at the same time. I check the trade in the past 30 days with 100% quality history. What was earlier (99% quality) is of no interest. Last month it was 146%. And dividing the account, for example, 10000 dollars in half: the profit plan for 5000 - 146%. Ie 7300 dollars. In this case, the robot calculates everything himself: you only need the leverage of 1:200, a deposit of $15 and an account with zero indentation.

:)

Well, I won't interfere with your fantasies anymore.

 
Vitalii Ananev:

:)

All right, I won't bother you with any more fantasies.

It's not like this: ... the price is highly likely to reach the TP ...

Any planning is built on probability. e.g. dividend yield: you plan to get the declared 10% of earnings per share, but the majority shareholders decide to pay out 5 instead of 10. That's it! Or even stronger: the ruble, in which you have denominated shares, will collapse by 400% (1998) or 50% (2007) or 100% (2014) and your div. dough will dust along with your money ...

 
Konstantin Yartsev:

That's not how it works: ... the price has a high probability of reaching the TP ...

Any planning is built on probability. e.g. dividend yield: you plan to get the declared 10% profit per share, but the majority shareholders decide to pay 5 instead of 10. That's it! Or even stronger: the ruble, in which you have denominated shares, will crash by 400% (1998) or 50% (2007) or 100% (2014) and your div. dough will dust together with your money ...

Too bad no one can understand you.

You could say you are sharing and telling everyone about the Grail!

But don't worry, I hear you.

We are talking about that grail broker for 49 ycd that's in your profile, right?

 
Konstantin Yartsev:

Everything has to be planned. It's up to everyone to decide how.

If you look at my strategy, it's a medium term strategy based on strong signals on H4, where the price is highly likely to reach the TP. There are about 5-7 profitable signals during the week. Trading is diversified on 28 currency pairs, of which the robot chooses one with the highest potential for movement in the desired direction. There can be 3 pairs in the market at the same time. The deposit is calculated to be enough to open e.g. three cheap pairs (e.g. Novozel-Franc) or two expensive pairs (e.g. Pound-Newzbel) at the same time. I check the trade in the past 30 days with 100% quality history. What was earlier (99% quality) is of no interest. Last month it was 146%. And dividing the account, for example, 10000 dollars in half: the profit plan for 5000 - 146%. Ie 7300 dollars. In this case, the robot will calculate everything itself: you only need a leverage of 1:200, a deposit of $15 and an account with zero indentation (ECN, NDD, etc.).

If you have a TS on H4 in the usual sense. Can you explain why you need exactly 100% quality? If a candlestick is 5 pips higher/lower (H4) the TP , SL will not work correctly or what? On H4 it's a month... Statistics, sampling adequacy, haven't you heard?) About splitting the account I agree.
 
EgorKim:

Too bad no one can understand you.

You could say you're sharing and telling everyone about the Grail!

But don't worry, I got you.

We are talking about that grail advisor for 49 ycd that you have in your profile, right?

The strategy is priceless. And 49 is a month's rent.

 
Aleksey Mavrin:
If you have TS on H4 in the usual sense. Can you explain why you need exactly 100% quality? If a candlestick is 5 pips higher/lower (H4) then TP, SL will not work correctly or what? On H4 it's a month... Statistics, sampling adequacy, haven't you heard?) About splitting the account I agree.

The built-in indicator analyses every tick of 28 pairs and for its readings to be reliable it needs 100% quality history, which is only available for the last 30 days.

The strategy opens a position with a pivot (you can open it with any step towards profit: 1, 2, 3 ... pips or even market) which is set in 1 pip increments, i.e. bythe closing price of the H4 signal, and is closed by TP.

Multicurrency virtual stop-loss is triggered at a specified level of account drawdown after M1 closing - and here 99% of ticks quality is not critical.

On "optimization on a long history". All Expert Advisors in the Market are losing on the future market, although most of them are optimized. First, why are they optimized on the nonexistent synthetic history (99%)? And the second question: what is the optimization if it's not a trivial fitting of results, even at 100% quality ticks?

So my recipe is simple. Take a reliable strategy that gives more than 100% per month on history with 100% quality (for example, mine: 146% per month with 29% drawdown). Divide the money in half and trade. E.g. you have $10,000: trade for $5,000. Your plan is 146%. For example, you got 146% of 5000 - it is $ 7300. Add 10,000 and we have 17300. On the next month again divided by 17300/2 = 8650. At the end of the month received the plan 146% - 12629. Add them up, divided and trade. And so on.

And to test, optimize and do other nonsense on fake quotes older than 30 days, you can spend the rest of your life.

Документация по MQL5: Константы, перечисления и структуры / Константы индикаторов / Ценовые константы
Документация по MQL5: Константы, перечисления и структуры / Константы индикаторов / Ценовые константы
  • www.mql5.com
Технические индикаторы требуют для своих расчетов указания значений цен и/или значений объемов, на которых они будут считаться. Существуют 7 предопределенных идентификаторов перечисления ENUM_APPLIED_PRICE, для указания нужной ценовой базы расчетов. Если технический индикатор для своих расчетов использует ценовые данные, тип которых задается...
 
Konstantin Yartsev:

The built-in indicator analyses every tick of 28 pairs and for its readings to be reliable it needs 100% quality history, which is only available for the last 30 days.

The strategy opens the position by a pending order (any step towards profit is possible: 1, 2, 3 ... pips or even market) which is set in 1 pip increments, i.e. by the closing price of the H4 signal, and is closed by TP.

Multicurrency virtual stop-loss is triggered at a specified level of account drawdown after M1 closing - and here 99% of ticks quality is not critical.

On "optimization on a long history". All Expert Advisors in the Market are losing on the future market, although most of them are optimized. First, why are they optimized on the nonexistent synthetic history (99%)? And the second question: what is the optimization if it's not a trivial fitting of results, even at 100% quality ticks?

So my recipe is simple. Take a reliable strategy that gives more than 100% per month on the history with 100% quality (for example, mine: 146% per month with 29% drawdown). Divide the money in half and trade. E.g. you have $10,000: trade for $5,000. Your plan is 146%. For example, you got 146% of 5000, which is $7300. Add 10,000 and we have 17300. On the next month again divided by 17300/2 = 8650. At the end of the month received the plan 146% - 12629. Add them up, divided and trade. And so on.

And to test, optimize and do other nonsense on fake quotes older than 30 days, you can spend the rest of your life.

Maybe I'm wrong, but I'm convinced that the difference of quotes of 100% from 99% affect no more than the difference in quotes of brokers among themselves, and the possible differences that occurred with the same broker tomorrow - the day after tomorrow.

I mean that the optimal strategy should not depend on the quality of tick history, if there is a reliable OHLC M1 or even M5 it should work. If only the size of significant movements in the strategy is comparable to the size of M1 and M5 candles, roughly not more than 2-3 medium candles, i.e. scalpers are an exception. And for the medium term it does not matter at all, the main thing that the candles were five-one minute candles.

If you certainly have some clever indicator, then for God's sake, but then it will also be lost when you switch to another broker, but that's OK).