Avalanche - page 465

 

Checked the arithmetic sequence of lots in my Expert Advisor. I had to increase the number of open orders to 9 in order not to drain the Expert Advisor. The maximum lot is 0.45. The initial lot is 0.05. The results are worse.

SymbolEURUSD (Euro vs US Dollar)
Period1 Hour (H1) 2009.06.01 01 01:00 - 2011.06.16 10:59 (2009.06.01 - 2011.07.01)
ModelBy open prices (only for Expert Advisors with explicit bar opening control)



Bars in history13642Modelled ticks26283Modeling qualityn/a
Chart mismatch errors0




Initial deposit7500.00



Net profit4501.45Total profit54320.16Total loss-49818.71
Profitability1.09Expected payoff11.37

Absolute drawdown3375.48Maximum drawdown3881.89 (48.48%)Relative drawdown48.48% (3881.89)

Total trades396Short positions (% win)196 (59.69%)Long positions (% win)200 (68.50%)

Profitable trades (% of all)254 (64.14%)Loss trades (% of all)142 (35.86%)
Largestprofitable trade4080.96losing deal-4341.20
Averageprofitable deal213.86losing deal-350.84
Maximum numbercontinuous wins (profit)11 (313.47)Continuous losses (loss)5 (-987.09)
Maximumcontinuous profits (number of wins)11368.78 (5)continuous loss (number of losses)-11294.80 (4)
Averagecontinuous winnings3Continuous loss
 
khorosh:

Checked the arithmetic sequence of lots in my Expert Advisor. I had to increase the number of open orders to 9 in order not to drain the Expert Advisor. The maximum lot is 0.45. The initial lot is 0.05. The results are worse.

Have you tried to use anti-martini?
 
Cmu4:
Have you tried applying the anti-martini?

No, I haven't. I think you can improve the results only by improving the quality of entries first of all. Then you will need fewer orders and the drawdown will be less to make a profit.
 
khorosh:
No, I haven't.


Maybe it makes sense.

It will look like this: if you have a profitable trade, the next one will increase by a step, and so on up to a certain maximum value; if you have a losing trade, the volume of the lot will gradually decrease to the starting volume.

You have 11 winning and 4 losing trades in a row. So the anti-martini makes sense, at first glance... although you can't see the average length of continuous losses.

 
Cmu4:


Maybe it makes sense.

It will look like this: if you have a profitable trade, the next one will increase by a step, and so on up to a certain maximum value; if you have a losing trade, the volume of the lot will gradually decrease to the starting volume.

You have 11 winning and 4 losing trades in a row. So the anti-martini makes sense, at first glance... although you can't see the average length of continuous losses.

This is probably good for an averaging martin, but I have a reversal martin.
 
khorosh:

Checked the arithmetic sequence of lots in my Expert Advisor. I had to increase the number of open orders to 9 in order not to drain the Expert Advisor. The maximum lot is 0.45. The initial lot is 0.05. The results are worse.

SymbolEURUSD (Euro vs US Dollar)
Period1 Hour (H1) 2009.06.01 01 01:00 - 2011.06.16 10:59 (2009.06.01 - 2011.07.01)
ModelBy open prices (only for Expert Advisors with explicit bar opening control)



Bars in history13642Modelled ticks26283Modeling qualityn/a
Chart mismatch errors0




Initial deposit7500.00



Net profit4501.45Total profit54320.16Total loss-49818.71
Profitability1.09Expected payoff11.37

Absolute drawdown3375.48Maximum drawdown3881.89 (48.48%)Relative drawdown48.48% (3881.89)

Total trades396Short positions (% win)196 (59.69%)Long positions (% win)200 (68.50%)

Profitable trades (% of all)254 (64.14%)Loss trades (% of all)142 (35.86%)
Largestprofitable trade4080.96losing deal-4341.20
Averageprofitable deal213.86losing deal-350.84
Maximumcontinuous wins (profit)11 (313.47)Continuous losses (loss)5 (-987.09)
Maximumcontinuous profits (number of wins)11368.78 (5)continuous loss (number of losses)-11294.80 (4)
Averagecontinuous winnings3continuous loss


:-))) Yuri, try the sequence of lots according to the Fibo numbers. For my part, I "hasten" to inform you that I "caught up" with you in terms of maximum drawdown...

Netting version of Avalanche - entry by trend in a "higher" TF, the working TF - 15 min, trawl by fractals from the first iteration, increasing volume of lots by Fibo numbers (details in fm377 p. 25), all deals are evenly distributed in time - optimization of parameters (in many respects conditional - "working" round values were taken as a result of its implementation, parameter step change is also great by round values) from Jan 2009 to Jan 2011. The parameter optimization was carried out evenly from Jan. 2009 to Jan. 2011, from Jan. 2008 to Jan. 2009 - backtest, the step of parameter changes - also a large round value. 2009 - backtest, from Jan. 2011 - up to now - forward, Changing the channel width - dynamically, according to the APR value.

Volume of positions - constant 0.01 lots. And you correctly wrote about possibility of its proportional increase with increasing the amount of capital for every 500 units.

Tentatively - up to 400 trades - backtest, then optimisation period and from about 1000 - forward - all trades are evenly distributed in time.

Two nods of the balance graph downwards - around 200th and 900th trade - is an entrance to the market - volume 0,34 lot (eighth flip) with subsequent exit to profit or (almost :-))))) to breakeven.

I am especially glad, that this year drawdown is minimal, I am testing it on demo.

Thank you, that at the time you posted a picture of your avalanche variant with a maximum drawdown around 500... :-))) It was very interesting for me to approach such milestones myself... :-)))

 
Roman.:


Hasn't it been draining since 2006?
 
khorosh:
Hasn't it been draining since 2006?


Since 2002 - minutes. The maximum two downwards runs - the first - August 2004 - coming out with 3.77 lots at 0.01 start - 13th flip with a 48 points channel width, the second - July 2007 - coming out with 5 (maximum micro) lots - 14th flip with a 30 points channel width - there and then turning into a profit. The starting lot is constant.

 
khorosh:
Hasn't it been leaking since 2006?

Seems that to be in a good profit (all other things being equal - control over communication, handling requotes, etc.) you have to start not with 500 cu, but with 5000 cu on micro... And every 5000 c.u. increase the lot size by 0.01... :-)))
 


Here's the avalanche from June 14 to today on the demo

Reason: