Help me write an EA, thanks in advance - page 23

 
Good morning, I think it is easier to add in the settings if we want to set a breakeven zone depending on the deposit or even if a group of orders will be closed in + depending on how many even and odd orders will be closed. Suppose a user with a small deposit has set 2 orders in the settings - so the advisor will not stop opening orders and just change profit/losses of all orders by 13pp and change the stops by 33pp, and then the advisor will open at 000 or a little bit in the plus, and can immediately start trading again, it is also good to put in the settings so that the user can select the parameters of the safe zone of his choice
 
and you get a picture like this
 
edikjefimov:

---------------------------------------------------------------------------1.2550 profit from bai and stop from sells here



--------------------------------------------------------------------------1.2500 бай 0.01/0.04

-------------------------------------------------------------------------.1.2480 сел 0.02



------------------------------------------------------------------------1.2430 stops from bays and profits from sells here

so maybe you can see better? i have not open more than 4 orders, and how long may the price stay in one place? not very long )) because i have to walk

Your bicycle was invented a long time ago and has been studied from all sides, see the branch Avalanche, (by the way, there are also EAs). If you don't use the analysis, this EA will eventually sell out. If we use thechanalysis for entry and limit the number of knees, then in principle it may work profitably, though with a rather large drawdown.

Here are the results of testing one of the variants of such a TS: starting lot 0.05, maximal lot 1.95, maximal number of open positions - 5.

SymbolEURUSD (Euro vs US Dollar)
Period1 Minute (M1) 2009.01.02 06:01 - 2012.01.12 23:59 (2009.01.01 - 2012.01.13)
ModelBy open prices (only for Expert Advisors with explicit bar opening control)



Bars in history1114538Modelled ticks2218833Modeling qualityn/a
Chart mismatch errors0




Initial deposit10000.00



Net profit37418.37Total profit101985.91Total loss-64567.54
Profitability1.58Expected payoff86.42

Absolute drawdown522.62Maximum drawdown4697.34 (13.98%)Relative drawdown18.37% (2637.28)

Total trades433Short positions (% win)232 (66.38%)Long positions (% win)201 (64.18%)

Profitable trades (% of all)283 (65.36%)Loss trades (% of all)150 (34.64%)
Largestprofitable trade4595.18losing transaction-4730.45
Averageprofitable deal360.37losing trade-430.45
Maximumcontinuous wins (profit)11 (670.38)Continuous losses (loss)4 (-1408.07)
MaximumContinuous Profit (number of wins)4595.18 (1)Continuous loss (number of losses)-4730.45 (1)
Averagecontinuous winnings2Continuous loss1
 

Schedule:


 
khorosh:

Graph:

An avalanche, yes. So? The code is similar, even more primitive: there is no check of maximum lot, min. lot step, distance of stops and traps from price and other.

You should test an avalanche on all ticks by the way, then its disadvantages will be seen in the full glory ;-)

 
evillive:

An avalanche, yes. So? The only programmer who posted his code here followed in the footsteps of other avalanche writers. The code is similar, even more primitive: there is no check of maximum lot, min. lot step, distance of stops and stops to price and so on.

You should test an avalanche by all ticks, by the way; then its disadvantages will be clearly seen ;-)

Some experts (paukas) recommend to test it at opening prices of one-minute ticks because the ticks in the tester are artificial.
 

If you know that an avalanche can be tested not only on M1, but surely on all ticks, the result will be the same on M5 as on H1 if the algorithm is programmed correctly)))

Moreover, even artificial tester ticks are more accurate than the opening prices or the more so the check points. Moreover, if an EA is working tickwise, it would be a big mistake to test it at open prices, it is only to mislead someone or to make nice reports )))

 
evillive:

If you know that an avalanche can be tested not only on M1, but surely on all ticks, the result will be the same on M5 as on H1 if the algorithm is programmed correctly)))

Moreover, even artificial tester ticks are more accurate than the opening prices or the more so the check points. Moreover, if an EA is working tickwise, it would be a great mistake to test it by opening prices, it can only mislead someone or make nice reports))

I did not feel lazy and ran it on ticks. The difference is not essential. But time consumption for testing is very significant. The difference may be considerable if we use scalping. But in this case profits are from 10 to 65 pips and stops can only trigger in an emergency.

SymbolEURUSD (Euro vs US Dollar)
Period1 Minute (M1) 2009.01.02 06:01 - 2012.01.12 23:59 (2009.01.01 - 2012.01.13)
ModelAll ticks (most accurate method based on all smallest available timeframes)



Bars in history1114538Modelled ticks41399417Modeling quality25.00%
Chart mismatch errors0




Initial deposit10000.00



Net profit35182.18Total profit100203.65Total loss-65021.47
Profitability1.54Expectation of winning79.60

Absolute drawdown534.75Maximum drawdown4778.54 (14.94%)Relative drawdown19.18% (2670.68)

Total trades442Short positions (% win)238 (66.39%)Long positions (% win)204 (64.22%)

Profitable trades (% of all)289 (65.38%)Loss trades (% of all)153 (34.62%)
Largestprofitable trade4560.08losing transaction-4716.41
Averageprofitable deal346.73losing deal-424.98
Maximum numbercontinuous wins (profit)12 (635.66)Continuous losses (loss)4 (-1328.32)
Maximumcontinuous profits (number of wins)4560.08 (1)continuous loss (number of losses)-4716.41 (1)
Averagecontinuous winnings2continuous loss1
 

khorosh:

Не поленился, прогнал по тикам. Разница не существенная. Но затрата времени на тестирование очень существенная. Разница может быть существенная, если используется пипсовка. Но в данном случае профиты от 10 до 65 пунктов, а стопы могут сработать только в аварийном случае.

Did you test the author's self-made device or is there such an advisor in kodobase? I would like a more recent test, 2009 is not really relevant.

 

Hello all, I tried an avalanche, it is not what I said, it does everything to drain the deposit, there widens each successive order corridor instead of narrowing and go into breakeven on a long flat and start over again, so the profit may not see a user because moving away the order and the profit is moved away, in short, a complete sinker