T101 basket trading system - math analysis - page 12

 

it's been long since i observed a break-even function may help avoid many of such losses. Today i've seen twice 150 pips + floating profit then turned to a stop loss.

i cannot post link, TSD is an awful forum, but check forexfactory's steve's multi-purpose trading manager (google it, maybe). it is easy to set SL/TP/Trail stop/Jump stop for basket trades.

echelon4x:
Hi Charvo, Today has been poor for BasketBull, my net profit so far now reduced to 12.21%.
 

Hi Charvo,

I've seen the multi-purpose trading manager before and it works well, however I read somewhere that the BasketBull developer advises against using external software to close trades for some reason. I assume that EA could potentially get confused as to what it's tracking. Try it out and see if you get better results, it could be just the developer being over-cautious, however I know that the program keeps track of various variables which are stored as persistent environment variables which may have something to do with it.

I believe that the parameters can be tweaked so that you always take profit at a % value of account (see the parameter instructions for "start trail" and "trail"). I think the % value of account is currently set to 10%, so you need to get some hefty winners for the system to close trades when at a particular profit. I think that you can change these parameters and still retain the current trailing stop logic, though it's not something I've tried.

EA's like this will have bad days it's part of the joy of trading I think the only sure-fire way to take as much profit as possible from the EA is to manually intervene from time to time and "close all" based on what you're seeing on the charts. The discussion thread for the EA mentions that the EA works best with a "Head Trader".

At the moment BasketBull appears to be starting to recover a bit and is showing 3.26% running profit for me (around $360) - these are all sell trades. It looks as though the selling may well continue, so I'll leave these running. Last week I noticed that several times the EA got to $500 in profit and then this was cut back to take profits at around $250 with the trailing stops.

Ok, in typing this I see that BasketBull has decided to take profits. It took 251.8 pips for $266.78 which brings my net profit up to 14.47%. A few more of those sort of trades for the rest of the week will be nice

 

Great! I'm happy to see you keeping pipping in.

Yeah, i read in the manual, roughly feeling the idea that don't let external script/EA to interfere with the positions. Good reminder, I shall continue look into the code.

Looks like I should double-check my .set, my EA on Alpari Demo is doing buy basket this U.S. afternoon. now in negative

echelon4x:
Hi Charvo,

I believe that the parameters can be tweaked so that you always take profit at a % value of account (see the parameter instructions for "start trail" and "trail"). I think the % value of account is currently set to 10%, so you need to get some hefty winners for the system to close trades when at a particular profit. I think that you can change these parameters and still retain the current trailing stop logic, though it's not something I've tried.

Ok, in typing this I see that BasketBull has decided to take profits. It took 251.8 pips for $266.78 which brings my net profit up to 14.47%. A few more of those sort of trades for the rest of the week will be nice
 

Hi Charvo,

Which Alpari are you using ? UK, RU or US?

You can have exactly the same *.set and template files as me, but unless you run on FxPro demo, your trades won't necessarily end up the same as mine. If you're based in the US then FxPro isn't a live option for you unfortunately because they closed all US accounts at the end of Jan this year (I assume because the excessive form-filling for US regulations got too much for them, though the official line was that they wanted to expand more in Europe and Asia).

I noticed earlier today when I switched on GoMarkets demo briefly that the indicators on GoMarkets BasketBull weren't the same as those on FxPro.

I assume that broker differences will be mainly down to two things:

1. Spreads

2. GMT offset

Both of these can lead to slightly different indicator readings which could affect the direction of the traded basket.

The good news is that all the indicators are adjustable via the parameters so you may find that you need to tweak these a bit to get better results.

 

yes, it is UK demo. i've already re-applied your .set, restarted running now.

i guess it is about spreads and data feed causing differences.

GMT is not mentioned in either manual or code.

but data feed depend on every pair's data feeding. the manual mentioned, that every customed indicator need to wait for each pair's new tick to arrive to calculate a new indicator value .

echelon4x:
Hi Charvo,

Which Alpari are you using ? UK, RU or US?

I assume that broker differences will be mainly down to two things:

1. Spreads

2. GMT offset

Both of these can lead to slightly different indicator readings which could affect the direction of the traded basket.

The good news is that all the indicators are adjustable via the parameters so you may find that you need to tweak these a bit to get better results.
 

Hi Charvo,

GMT doesn't need to be mentioned anywhere, however if any part of the code or indicators relies on the close of a bar which is more than 1H, then that bar will be very different between brokers on GMT+3 and brokers on GMT-4 for example.

Alpari UK is on GMT+2 and FxPro is GMT+3 which means that the 4H, 1D, 1W and 1M bars will be slightly different.

Data feeds will be different between brokers and I agree that this is probably a contributing factor to differences in signals in our case. I've not seen quotes being too different between brokers (apart from in times of big news), however it's possible that a small price spike from one broker can trigger a trade which isn't triggered on the other broker.

 

Hi Guys,

Something to be aware of if you're following my results...

All my systems disconnected for a while this morning and I was left with a basket and some straggling trades which I had to manually close.

I took this opportunity to change the timeframe (parameter tf) from 60 to 240, so BasketBull is now running on 4H charts (as the developer originally intended for version 7.0).

Using indicators based on the 1H timeframe seems a bit too fickle - great results last week and rubbish results this week so far. I believe that the 4H timeframe will give much more consistent results over the longer term.

 

Running on Instaforex

Just wondering if anyone has gotten this to run on Instaforex yet?

I copied all files from Alpari dir , but still no luck

 

Basza,

Are Instaforex quotes to 4 decimal places ?

If so, try setting the parameter "min_lot_mtp" to 29 and see if that works.

 
echelon4x:
Basza,

Are Instaforex quotes to 4 decimal places ?

If so, try setting the parameter "min_lot_mtp" to 29 and see if that works.

Thanks echelon that did the trick. Now working.

Reason: