Basket Trading BALUDA'S CSS Crossovers

[neocount]

Sorry about that Raj. I was actually testing 2 different settings. I went with atr based though because a fixed number is too confining for big pip movers like Guppy or EJ. A 20 pip tp grid on Guppy can be hit in a matter of seconds and then you have to decide to redo the grid or not. I tried a MA sl last week and lost 75% so I decided not to go that route.

Hi Neocount!

Getting confused by your last post. In your first post you wrote "Basket TP set at D1 ATR", but in the post above you wrote "TP for the grids is set at 20". Would love to really understand your trading rules and see if we can all improve and refine. Can you perhaps give an example?

Cheers and happy pippin' [/quote]

[syberiapoya]

I've been running the M5H1D1 set file (GTF version) with 5 different variants for the last 3 weeks. Completely hands off...
1- TP at 0.5%
2- TP 1%
3- without GTF
4- include the NZD & CHF crosses
5- max 3 trades per symbol every H4 candle (instead of D1 candle)

For the first 1.5 weeks, each one had a different profit, drawdown & margin use but the results were pretty close.
The first number is profit percent & the second number is maximum drawdown.
1- 6.5% - 5%
2- 7.5% - 5%
3- 5.5% - 6%
4- 6.5% - 9%
5- 9% - 9%
The 4th set used the highest amount of margin & the 5th set was next.

But from the middle of the second week until the end of the third week no baskets were closed on any set. During this time, 4 of the set files went into a 10 percent drawdown. The 3th set file went into a 14% drawdown.

At the end of the third week, all set files except the 4th set file closed their baskets.
Currently the 4th set file has a drawdown of about 30%.
---------------------------------------------------------------------------

I think this is an EA with a lot of potential. But it must be disabled during news with very high impact or when there is very low market volatility. And it's best to close all of the trades manually at the end of each week.

[TraderJoeForex]

I've been running the M5H1D1 set file (GMT version) with 5 different variants for the last 3 weeks. Completely hands off...
1- TP at 0.5%
2- TP 1%
3- without GMT
4- include the NZD & CHF crosses
5- max 3 trades per symbol every H4 candle (instead of D1 candle)

For the first 1.5 weeks, each one had a different profit, drawdown & margin use but the results were pretty close.
The first number is profit percent & the second number is maximum drawdown.
1- 6.5% - 5%
2- 7.5% - 5%
3- 5.5% - 6%
4- 6.5% - 9%
5- 9% - 9%
The 4th set used the highest amount of margin & the 5th set was next.

But from the middle of the second week until the end of the third week no baskets were closed on any set. During this time, 4 of the set files went into a 10 percent drawdown. The 3th set file went into a 14% drawdown.

At the end of the third week, all set files except the 4th set file closed their baskets.
Currently the 4th set file has a drawdown of about 30%.
---------------------------------------------------------------------------

I think this is an EA with a lot of potential. But it must be disabled during news with very high impact or when there is very low market volatility. And it's best to close all of the trades manually at the end of each week.

Just checking in and so nice to see this summary. This is a helpful sanity check for the settings I decided to run with. I think you mean GTF (Global Trend Filter) rather than GMT not to confuse anyone.

Do Keep up the testing and great work... I have been testing a few new ideas on demo too which may offer some further improvements. Will report back in New Year.

[syberiapoya]

Glad I could help... Your original set file proved to be one of the best ones.

Yes I meant GTF. Fixed it...
And thanks for everything you've been doing.

[lawrencew]

@Mathias

Thank you very much for the dashboard EA, it's a work of art!

I was just wondering: would the full EA be still usable when the hgi_lib expires? What about the simplified version? Because HGI is not integral to the strategy and I also don't use it at all, I thought it might be useful to have a version which is not dependent on hgi_lib.

Happy holidays!

[MathiasJ]

@Mathias

Thank you very much for the dashboard EA, it's a work of art!

I was just wondering: would the full EA be still usable when the hgi_lib expires? What about the simplified version? Because HGI is not integral to the strategy and I also don't use it at all, I thought it might be useful to have a version which is not dependent on hgi_lib.

Happy holidays!

Thanks!
You can easily try it by removing the hgi_lib, then restart the terminal and try to load the EA with settings not using HGI, neither as entry trigger nor as exit trigger.
It should work, but I have never tried it.
BR
Mathias

[lawrencew]

@TJF, I was looking at your setting and realised that you put AvoidNews_ClientComputer_GMT_Offset=2. Since you already set your local time to GMT, I believe this variable should have been set to 0 instead of 2, right?

@Mathias,
Thanks for the reply, I'll try it out and post the result.

Meanwhile, I have a few questions on the settings:
1. GlobalMarketTrend (GMT) Delta: how is this different from GMT Level? And does its "Candles" and "Shift" variables work the same way as CSS Delta?
2. After activating AvoidNews, should it show up above the table on the main panel along with other activated settings? Because it's not showing currently and I've set UseThisSetting to TRUE.
3. If I set the possible trade exit conditions (eg "Close if lower CSS Delta") to FALSE, then the "CSS Delta Exit Settings" below it (with timeframe, type, min, candles, etc variables) won't matter right? CSS can have whatever Delta value and individual trades won't be exited?
4. Related to above, my Min and Max for entry are 0.2 and 99, respectively. If I want to make individual trades exit, I need to put the "Min" in the Exit settings to -20 or something like that (substituting for -20 with other negative numbers, NOT positive numbers like 30)?
5. "Exit_ExpirySeconds" there is no toggle to activate/deactivate this. Is this active?
6. "CloseTime (local) to close all open trades hh:mm" --> no toggle as well. Is this active? What day is it active?

Thanks a lot in advance Mathias!

[MathiasJ]

@TJF, I was looking at your setting and realised that you put AvoidNews_ClientComputer_GMT_Offset=2. Since you already set your local time to GMT, I believe this variable should have been set to 0 instead of 2, right?

@Mathias,
Thanks for the reply, I'll try it out and post the result.

Meanwhile, I have a few questions on the settings:
1. GlobalMarketTrend (GMT) Delta: how is this different from GMT Level? And does its "Candles" and "Shift" variables work the same way as CSS Delta?
2. After activating AvoidNews, should it show up above the table on the main panel along with other activated settings? Because it's not showing currently and I've set UseThisSetting to TRUE.
3. If I set the possible trade exit conditions (eg "Close if lower CSS Delta") to FALSE, then the "CSS Delta Exit Settings" below it (with timeframe, type, min, candles, etc variables) won't matter right? CSS can have whatever Delta value and individual trades won't be exited?
4. Related to above, my Min and Max for entry are 0.2 and 99, respectively. If I want to make individual trades exit, I need to put the "Min" in the Exit settings to -20 or something like that (substituting for -20 with other negative numbers, NOT positive numbers like 30)?
5. "Exit_ExpirySeconds" there is no toggle to activate/deactivate this. Is this active?
6. "CloseTime (local) to close all open trades hh:mm" --> no toggle as well. Is this active? What day is it active?

Thanks a lot in advance Mathias!

Hi!
A lot of questions!
I will try to anwer..:
1. Global MarketTrend Delta is the direction of the indicator, just like CSS Delta. Candles and Shift works the same.
2. You need to set "Use News 1 filtering for Entry" to Normal for it to work as well.
3. You are correct.
4. It can actually be a positive value as long as it is lower than the min value of the normal CSS Delta trigger. But it is normally negative or 0 yes.
5. Under "Other Trade Exit Conditions" you can set "Close if expired unprofitable". That is the enabler.
6. Again under "Other Trade Exit Conditions" is "Close if time pass CloseTime".
Good questions, I hope to update the helpfile somewhat more to make it easier for you to read about all the features.
BTW, I consider renaming Global Market Trend (GMT) to something else in future in order not to confuse it with GMT time...
BR
Mathias

[TraderJoeForex]

@TJF, I was looking at your setting and realised that you put AvoidNews_ClientComputer_GMT_Offset=2. Since you already set your local time to GMT, I believe this variable should have been set to 0 instead of 2, right?

...I don't use the news filter myself but this makes sense if it is turned on.

[TraderJoeForex]

Just checking in and so nice to see this summary. This is a helpful sanity check for the settings I decided to run with. I think you mean GTF (Global Trend Filter) rather than GMT not to confuse anyone.

Do Keep up the testing and great work... I have been testing a few new ideas on demo too which may offer some further improvements. Will report back in New Year.

Quick update ahead of next week ... In my continued testing to address some of the weaknesses from my M5H1D1 set the one with most potential also seems to be the one of the most simple.

Here instead of using three timeframes for diff and delta I just use the most important one .. D1.
I still keep the Global Trend Filter on H1 too. This seems to get into trades a bit quicker and close baskets a bit quicker on average so far.

I tried this setup with a non basket approach some time ago (http://www.stevehopwoodforex.com/phpBB3 ... 03#p149403) but lots of small losses wipe out the bigger gains. I am still experimenting with this non basket set using H1 HGI trend arrows for entry and M15 for exit to refine further but have not really made a great deal of progress no thanks to the sleepy market at this time of year. So I also decided to try a less is more approach with a basket too and here is this new set below:

In the past 10 days it has banked a fair bit more than M5H1D1 +8.2% magic 3 vs +5.3% magic 1 (...although drawdown has been proportionately higher too so it's not a clear advantage yet).

I am attaching here for those that may wish to test .. I would like to see what happens with different Global Trend Filter delta settings since this is something I have not really got as far as tweaking yet.

This is definitely NOT for a live account yet. Just because it has shown promise for me in the past 10days is no indication of how it will perform once some volatility comes back into play in January. Hopefully very well but we need to test more and see.