stevehopwoodforex.com
https://www.stevehopwoodforex.com/phpBB3/
Print view

The Awesome Basket Forward Test
https://www.stevehopwoodforex.com/phpBB3/viewtopic.php?t=5240
Page 16 of 24
Author:  c1borg [ Tue Nov 07, 2017 4:15 pm ]
Post subject:  The Awesome Basket Forward Test

Ok thanks for the info might try and fit your settings into the 1q version if there is any benefits code wise.
Author:  pouet [ Wed Nov 08, 2017 4:52 pm ]
Post subject:  The Awesome Basket Forward Test

Hello All,

Thanks for this wonderful ea and all the work you do for testing it.
I just have a question :
Is there a reason why on all setfiles used here in the list of Pairs to use GBPAUD, GBPNZD, GBPCAD, NZDCHF and NZDCAD are missing ?
Author:  c1borg [ Wed Nov 08, 2017 5:17 pm ]
Post subject:  The Awesome Basket Forward Test

I think you can edit the pairs and include the ones you like. From the tests I have going atm it might be better to exclude some of the wilder fast moving ones.
Author:  WhoKnows [ Thu Nov 09, 2017 12:08 am ]
Post subject:  The Awesome Basket Forward Test

c1borg wrote:Ok thanks for the info might try and fit your settings into the 1q version if there is any benefits code wise.
I am no coder, but from reading the thread I would think there were some very important changes from 1m to 1q that may impact performance.

IMHO one should always try to work with the latest versions because they are least bug prone. But also because it shows appreciation for the work of all excellent coders here working tirelessly in an effort to improve the EAs. :clap: :clap:

BUT ...if you have a profitable old version, one should IMHO always - over a period of two weeks and at least 100 trades -, run a new version parallel with the old version until convinced that you can replicate or improve on the profitable old version (with adapted set file when needed).

But back to Awesome:
One very important change would be that the issue of trades closing on the first opposite FB (instead of close on opposite FB ONLY WHEN SS agrees), was solved in 1o. The details are described in the thread.
I do not say that this fix improves results (I did not yet test conclusively myself). The 1o fix made Awesome close trades as originally designed, but this behavior would be differently from the behavior from versions before the 1o fix (as with 1o the setting about 'only close when SS agrees' in the setfile is no longer ignored).
Hence you need to adapt the set file if you want same behavior as before 1o.
According to the AwesomeOriginal thread there were also changes in 1q with respect to the minimum pip distance between trades.

My testing has not been able to keep up with the changes, so I can't yet report on how the changes impact performance.
Author:  c1borg [ Thu Nov 09, 2017 5:33 am ]
Post subject:  The Awesome Basket Forward Test

WhoKnows » Thu Nov 09, 2017 12:08 am wrote:
c1borg wrote:Ok thanks for the info might try and fit your settings into the 1q version if there is any benefits code wise.
I am no coder, but from reading the thread I would think there were some very important changes from 1m to 1q that may impact performance.

IMHO one should always try to work with the latest versions because they are least bug prone. But also because it shows appreciation for the work of all excellent coders here working tirelessly in an effort to improve the EAs. :clap: :clap:

BUT ...if you have a profitable old version, one should IMHO always - over a period of two weeks and at least 100 trades -, run a new version parallel with the old version until convinced that you can replicate or improve on the profitable old version (with adapted set file when needed).

But back to Awesome:
One very important change would be that the issue of trades closing on the first opposite FB (instead of close on opposite FB ONLY WHEN SS agrees), was solved in 1o. The details are described in the thread.
I do not say that this fix improves results (I did not yet test conclusively myself). The 1o fix made Awesome close trades as originally designed, but this behavior would be differently from the behavior from versions before the 1o fix (as with 1o the setting about 'only close when SS agrees' in the setfile is no longer ignored).
Hence you need to adapt the set file if you want same behavior as before 1o.
According to the AwesomeOriginal thread there were also changes in 1q with respect to the minimum pip distance between trades.

My testing has not been able to keep up with the changes, so I can't yet report on how the changes impact performance.
I absolutely and wholeheartedly agree with everything you pointed out. 1m is still performing beyond my expectations and I have a second trial in place which is also amazing. I don't intend to break this with 1q but as you say newer versions are produced to get rid of bugs and add additional features, so worth looking at.
Author:  m0nk111 [ Thu Nov 09, 2017 11:48 am ]
Post subject:  The Awesome Basket Forward Test

here are my results with Awesome
i removed CHF pairs because this broker has different margin on those pairs
https://www.myfxbook.com/members/m0nk11 ... 40/2291211

now im running basicly the same test with Awesome Bells and Whistles and NewsFilter on a different broker where CHF has normal margin. (difference is no double trades across timeframes, enabled newsfilter and added CHF pairs) ive uploaded that setfile in the B&W thread. i'll post that set here when i also have myfxbook results
Author:  WhoKnows [ Thu Nov 09, 2017 12:55 pm ]
Post subject:  The Awesome Basket Forward Test

Thanks m0nk111 for contributing your 4 week MTF approach results. Very impressive.

Some questions if I may:
Which Awesome version did you use?
Is there a reason why you have ImmediateMarketOrders on False?
I was surprised that only 3 H4 trades were taken, any idea why?
Author:  m0nk111 [ Thu Nov 09, 2017 1:15 pm ]
Post subject:  The Awesome Basket Forward Test

demo started on 16 okt, so i used the version from then, which is v1q, the latest.
ImmediateMarketOrders false so im sure it goes the right direction, but the buffer is only 10 pips
i used a very small deposit. so when the MinimumMarginPrecentage of 1400% gets hit it stops adding new pairs to trade. H4 only has a possible signal every 4 hours, by then M30 already took all margin%, so thats why.

WhoKnows » Thu Nov 09, 2017 12:55 pm wrote:Thanks m0nk111 for contributing your 4 week MTF approach results. Very impressive.

Some questions if I may:
Which Awesome version did you use?
Is there a reason why you have ImmediateMarketOrders on False?
I was surprised that only 3 H4 trades were taken, any idea why?
Author:  marx64 [ Sun Nov 12, 2017 2:07 pm ]
Post subject:  The Awesome Basket Forward Test

So my Awesome 50€ TP survived the huge drawdown and closed in profit last week. Honestly I'm quite surprised, as I thought it will never survive 600€ DD. But it did.
I will change the setfile a bit today and use 200 pips emergency SL for safety purposes. It survived last time, but there are pairs which can move 1000 pips without any meaningful retracement and that would cripple the account, as there are only two closure options: reaching the global TP or margin call. Actually there is a third option as well- close on opposite FB if SuperSlope agrees but I have never seen any closures by that and not sure if it's working or not.
Author:  TraderJoeForex [ Sun Nov 12, 2017 2:57 pm ]
Post subject:  The Awesome Basket Forward Test

marx64 » Sun Nov 12, 2017 2:07 pm wrote:So my Awesome 50€ TP survived the huge drawdown and closed in profit last week. Honestly I'm quite surprised, as I thought it will never survive 600€ DD. But it did.
I will change the setfile a bit today and use 200 pips emergency SL for safety purposes. It survived last time, but there are pairs which can move 1000 pips without any meaningful retracement and that would cripple the account, as there are only two closure options: reaching the global TP or margin call. Actually there is a third option as well- close on opposite FB if SuperSlope agrees but I have never seen any closures by that and not sure if it's working or not.
My H4 sets close on opposite FB. Works perfect and ensures low drawdown however may not be desirable on the lower time frames.
All times are UTC Page 16 of 24