Automatic Loss Recovery System (ALR) - read this first

[milanese]

[ img ].... BTW, how to add pictures directly?! Img here does not automatically load external pics.

For all ,who do not know how to add pictures to their post read please this: http://www.stevehopwoodforex.com/phpBB3/kb.php?a=2

Cheers

Tommaso

[Lagrange]

[ img ].... BTW, how to add pictures directly?! Img here does not automatically load external pics.

For all ,who do not know how to add pictures to their post read please this: http://www.stevehopwoodforex.com/phpBB3/kb.php?a=2

Cheers

Tommaso

Thanks Tommaso!

I closed GBPCHF at small profit with original ALR strategy:

[TheInsider]

Ok, here's an updated version of my ALR Calculations, which I hope you will find easy to use and understand.

The result are the same (ALR_with_Spread_v8.5 mq4 vs ALR Calculator v2.1 and ALR_kotsch_v2.3), but they both differ slightly from the original ALR.

The formulas in the two latter Excel-files are IMHO unnecessary complex compare to the formula used in ALR_with_Spread (which uses the same formula for both direction/for both Buy and Sell), just alter Tot Buy Lots and Tot Sell Lots.

2014-11-03

I have re-uploaded a new version - v.1.4

Fixed bugs and added more calculations and some examples.

[TheInsider]

Hi Fibotrend,

this idea also crossed my mind, but if you try to implement it with an EA it is not practical.

Lets say your first hedge trade is 50 pips away from entry. It opens and price goes 10 pips into it then reverses back in original direction, so you close the hedge when (the hedge) is at BE. Now lets say on the way back in your original position favour, it oscillates at the -50 pips level dome time, It goes for example from -48 to -52 pips 10 times, this means you would every time open and close the hedge trade, paying spread and commissions, plus it is very painful to watch

You idea would much better be implemented manually, for example if the hedge goes to BE, you can put a pending order 10-15 pips BELOW the first hedge and increase hedge position size, so in case it reverses again you still catch it with the hedge, but you are at the same time giving it room to move in the ALR zone.

I've been playing around with numbers and thanks to Kotsh advice I've developed a trailing ALR version. What I'm trying to achieve is killing the trade ASAP after the hedge position kicks in. Therefore I thought we should have a recovery zone for the hedge, the same way we have it on the original trade.



The ratio 1:3 works well, as those levels correspond to fibo's likely to be hit. In our example we open the 2nd hedge 15pips above the -50 level, then we put a sell order 30pips below for the short hedge. As this trade has a long bias we keep trailing our orders in this way, long hedges 15pips above, short hedges 30pips below.

Since we are opening each position closer to each other, lot sizes are much smaller when compared to the original ALR version:

1 = 0.1
2 = 0.134
3 = 0.065
4 = 0.08
5 = 0.085
6 = 0.106
7 = 0.113
8 = 0.145
9 = 0.155
10 = 0.206
11=0.218

The smaller the lot size the smaller the margin, risk and exposure. Unfortunately I don't have any formulas or spreadsheet for the model. I've worked out the size for each hedge manually, any help from the math gurus in the forum would be very appreciated.

Another huge plus is that the orders are shifting away from the ALR zone. Basically if the price keeps bouncing in the range it will trigger 5 orders only.

Any feedback is welcome. Cheers

Thanks for your idea. I believe that I understand what you mean, and partly agree that it's good. However, I think it can be even better...

First, to fully understand what ALR Zone means (based on the actual calculation that are made) the ALR Zone is actually the position for a soft StopLoss, where instead of closing with SL it opens a new hedged position; a position in the opposite direction.

So a new hedged positions will be opened at the time the price reaches this soft SL.

The original idea is to open a new hedged position with the same SL and TP. Those values are static.

Added hedged positions will be opened as soon as the price reaches the soft SL, and always with the same static SL and TP.

I think the better solution is instead of using static SL and TP is to actually Trail SL in steps once it's X pips in profit, and to do so for both the original position (which will reduce the SL and risk), as well as for the hedged positions (which initially will be opened with the same SL and TP as the last known SL and TP for the previous position, but once also the hedged positions is X pips in profit it will modify the SL). As a result, it will require less number of opened positions, the lot sizes will not increase as much as before, and overall reducing the total risk and margin use.

[Fibotrend]

I think the better solution is instead of using static SL and TP is to actually Trail SL in steps once it's X pips in profit, and to do so for both the original position (which will reduce the SL and risk), as well as for the hedged positions (which initially will be opened with the same SL and TP as the last known SL and TP for the previous position, but once also the hedged positions is X pips in profit it will modify the SL). As a result, it will require less number of opened positions, the lot sizes will not increase as much as before, and overall reducing the total risk and margin use.

The idea sounds great, however I'm not sure how reliable MT would be in managing such a complex EA

I would be happy to have a semi-automated EA that takes care of the trailing hedges after the first hedge kicks in. I'm quite confident that the number of turns would reduce drastically comparing to the static ALR.

Another problem with the current EA system is that it takes over the entire trade regardless of market conditions. Ideally the EA should give you the option to set the trailing distance for short and long hedges after the first hedge opens, so it can be optimized according to market conditions.

Cheers

[TheInsider]

I think the better solution is instead of using static SL and TP is to actually Trail SL in steps once it's X pips in profit, and to do so for both the original position (which will reduce the SL and risk), as well as for the hedged positions (which initially will be opened with the same SL and TP as the last known SL and TP for the previous position, but once also the hedged positions is X pips in profit it will modify the SL). As a result, it will require less number of opened positions, the lot sizes will not increase as much as before, and overall reducing the total risk and margin use.

The idea sounds great, however I'm not sure how reliable MT would be in managing such a complex EA

I would be happy to have a semi-automated EA that takes care of the trailing hedges after the first hedge kicks in. I'm quite confident that the number of turns would reduce drastically comparing to the static ALR.

Another problem with the current EA system is that it takes over the entire trade regardless of market conditions. Ideally the EA should give you the option to set the trailing distance for short and long hedges after the first hedge opens, so it can be optimized according to market conditions.

Cheers

I'm already working on such an EA (that trails and dynamically adjust SL and TP for all orders). Will let you know how it performs once I start testing it.

[Dewey McG]

For anyone wondering still which is better, please look at this post where I did a back test using the same EA strategy comparing ALR I versus ALR II: http://www.stevehopwoodforex.com/phpBB3 ... 975#p97975

[blazecrown]

For anyone wondering still which is better, please look at this post where I did a back test using the same EA strategy comparing ALR I versus ALR II: Dewey McG @ Trend Trading EA with Automated Loss Recovery

hi Dewey, thanks for all these great work you've been doing along with Spider and the rest... i believe many are following this ALR topic but are either not posting comments or making replies (for whatever reason)... so for clarification, with all these ALRs posted and revised, may i know which particular ALR are you referring to as ALR I and ALR II respectively... thanks...

[SpiderX]

hi Dewey, thanks for all these great work you've been doing along with Spider and the rest... i believe many are following this ALR topic but are either not posting comments or making replies (for whatever reason)... so for clarification, with all these ALRs posted and revised, may i know which particular ALR are you referring to as ALR I and ALR II respectively... thanks...

Hi

Please read post 1 of this discussion and download the calculator.

Cheers

[Dewey McG]

I will make a new document later explaining the differences.

ALR 1 (the calculator says 2.1) is the first version of ALR based on what Joseph Nemeth developed and later copied (and marketed everywhere) by Dustin Pass. With that one the lots are larger and you are keeping the trades open. Since you are hedging it won't work with US brokers.

ALR II is a version developed by Spider and a friend of his. With this one you close the trade and open a new one, so it is compliant with US brokers. it works in a very similar way, but is able to use smaller lot sizes and much less risk. After doing some back tests (see above) you can see the difference in risk is dramatic.

Going forward I believe we should stick with ALR II.