Avalanche 6.2 - page 3

New comment
 
EvgeTrofi:

Or test over long intervals (e.g. 10 years). With such a history of trades, with > 500 trades, we can hope that the system is suitable for real.
I don't think so. 10 or 20 lots in a row may not be on the history, such a series may start at any time. Possible ways to solve this problem are to increase the initial balance and switch to 0.01 lots. The only question is in profitability of such an Expert Advisor. It can be too low (in addition, the risks decrease and do not disappear at all). The conclusion - it is possible to earn by martin, but it is risky.
 

The system is suitable except for self-education. Someone has already written that everyone, or almost everyone, has gone through something similar )).

In order not to wind up orders, try to forget about lots, play with money, not lots. For example, determine the maximum fixed percentage of a deposit, which will be in the market. Most likely, you will find the obvious thing - as soon as you stop covering losses with new positions, losses will greatly increase, up to being dumped.

Another thing that concerns not only Avalanche - if you trade from a channel, you always give the market two channels plus spreads. The use of stop orders for entry indicates that you have no idea when and at what price to enter. If you use fixed limits, stops and trall, this indicates that you do not know when and at what price to exit the market. Thus, it turns out that you do not know anything about the market and are guided only by the fact that you will be lucky and that the next trade will cover the previous losses, i.e. pure Martin. In my humble opinion, Martingale has a right to exist when you are about to leave the game with an "either-or" result.

However, there is one important thing about this system that you should pay attention to - it is a constant presence in the market. If the price changes, and it always does, you can always get a penny out of it. All that remains is to understand how to do it)))

 
Piccioli:

... If you're trading from a channel, you're always giving the market two channels plus spreads. The use of stop orders for entry indicates that you have no idea when and at what price to enter. If you use fixed limits, stops and trall, this indicates that you do not know when and at what price to exit the market. Thus, it turns out that you do not know anything about the market and are guided only by the fact that you will be lucky and that the next trade will cover the previous losses, i.e. pure Martin. In my humble opinion, Martingale has a right to exist when you are about to leave the game with an "either-or" result.


I couldn't agree with you more. However, this martingale is very tempting. You say it yourself, price is always changing, so it can't stay inside the channel for long. I am using avalanche strategy tested for 10 years while I am working on other more stable systems.


Symbol EURUSD (Euro vs US Dollar)
Period 1 Hour (H1) 1999.01.08 19:00 - 2010.08.25 08:00
Parameters FixLot=false; RiskLot=0.4; max=100; Step=1000; Band=0; Slippage=30; CloseOnly=false; OnlyOnePosition=false; TP_multiplier=0.8; reverse=0; MaxAge=72; Procent=200;
Bars in history 72147 Modelled ticks 1760073 Simulation quality n/a
Initial deposit 10000.00
Net profit 150906.40 Total profit 416119.03 Total loss -265212.63
Profitability 1.57 Expectation of winning 118.08
Absolute drawdown 784.85 Maximum drawdown 35062.70 (22.68%) Relative drawdown 47.95% (10845.94)
Total trades 1278 Short positions (% win) 640 (55.94%) Long positions (% win) 638 (57.84%)
Profitable trades (% of all) 727 (56.89%) Loss trades (% of all) 551 (43.11%)
Largest profitable trade 20801.77 losing deal -13812.02
Average profitable deal 572.38 losing trade -481.33
Maximum number continuous wins (profit) 12 (2842.24) Continuous losses (loss) 12 (-4916.23)
Maximum continuous profits (number of wins) 24205.77 (3) Continuous loss (number of losses) -22461.91 (6)
Average continuous winnings 3 continuous loss 2

 
I agree. Martin, while risky, has made a profit. He is honest - we are not trying to guess where price will go. The other systems, trend or bounce, try to play oracles, get it wrong and lose out. Personal experience.
 
I believe that if a 10 year old strategy with 1200 trades starts to fail, there must be some kind of cataclysm in the market, which is not impossible, but any other trading strategy is not immune to that either.
 
wmlab:
I agree. Martin, while risky, has made a profit. He says honestly - we do not try to guess where the price will go. Other systems, trend or rebound, try to play oracles, make mistakes and fail. Personal experience.


Well, not really, it's just that trend ATCs require a lot of careful watching and adjusting to the constantly changing market.

Today I added some martingale and something similar to it to my EA, the principle: if the last trade was a loss, then I increase the lot, if it is profitable, then I reset the lot increase to the initial one. Try to apply something similar to an avalanche, maybe the deposit will be less large

double lots(){
int time = 0;double profit = 0;//обьявляем необходимые нам переменные куда мы положим интересующие нас характеристики ордера
for(int i = OrdersHistoryTotal();i>=0;i--){// Перебираем все закрытые ордера
  if(OrderSelect(i,SELECT_BY_POS,MODE_HISTORY)){//если ордер с таким номером (i) в списке закрытых ордеров есть ( не путать с тикетом)
    if(OrderSymbol() == Symbol()){//если выбранный ордер был открыт по нашей валютной паре
      if(time<OrderCloseTime()){//(сравниваем его с хранящимся в пероеменной time) 
        time=OrderCloseTime();//если время закрытия ордера больше - ложим его в переменную
        profit=OrderProfit();//и заодно запоминаем прибыль ордера
      }
    }
  }
}
//по окончании этой процедуры в наших переменных будут сидет наибольшее время закрытия, и его профит. Или по нулям если история чистая.
//теперь мы можем выставлять условия в зависимости от результата процедуры
if(profit == 0 &&time == 0){//действия если история чистая}
maxlot=Lot;
}
if(profit >= 0){//действия если последний ордер был прибыльным, или нулевым}
maxlot=Lot;
}
if (profit <  0){//действия если последний ордер был убыточным}
maxlot=Lot+0.1;
}

return(maxlot);
}

I've added this construction, results are much better.

 
EvgeTrofi:


I use an avalanche strategy tested over 10 years,

Do you use it in a tester? In the "control points" model?
 
Well the test is of course from the tester (the price formation model is not important). And I use it on the real. You can try it yourself. A 5-digit account is recommended. H1 timeframe. It trades well on EURUSD and GBPUSD. I will not reveal the signal details yet.
Files:
needforpips_v2.11_1.ex4  18 kb
 
EvgeTrofi:


.... However, this martingale is very tempting....

Don't worry, it will heal before the wedding)))

Seriously though, one of the ideas I was trying to convey was that you should always operate with money only.

Do you think that the statement you've shown is beautiful, because the balance function is pointing to the sky? Well, I think your statement is simply awful from the financial point of view, and I won't say a word about the technical part. The point is that you have to have 10 tons of free money to start with and you will have 150 tons in 10 years, which with taxes and inflation will turn into 120-125 tons, which is not much. With a normal distribution the profitability of this system will be about 32% per annum, which in principle is not bad, if the risks (drawdowns) tended to zero. And they still tend to infinity.

Well, if you test with modeling quality at least more than 50%....

Small correction: I wrote the word inflation, but didn't count it - so with inflation 120 thousand of notional money in 10 years will be worth about as much as 70 thousand now.

 
Piccioli:

I think your state is terrible from a financial point of view, and I'm not going to say anything about the technical part.


Unfortunately, I don't have anything better yet. What about you? Do you have a more stable and profitable trading system?
12345678910...19
New comment
Reason: