GridDancer / GridMaster

Post by **tomele** » Sun Oct 23, 2016 8:42 pm

Hey Rene,

brilliant work as ever.

Will run a terminal with this fantastic stuff along my other tests.

Cheers, Thomas

Post by **acostafulano** » Sun Oct 23, 2016 8:53 pm

Hello Rene,

Could you please explain the difference between BasketProfit and MinProfit in the GridMaster?
I last traded GM 1.4 and it didnt have the BASKETPROFIT setting,

If I understand correctly, minprofit is the small closures that we'll have along the way (positive+negative trades closing out at this value to keep DD at bay), and then EQUITYTP was the equity target the GM needed to hit before the Rinse&Repeat

So... what would the BASKET PROFIT be for?

thanks!

Diego

Post by **renexxxx** » Mon Oct 24, 2016 2:33 am

acostafulano » Mon Oct 24, 2016 6:53 am wrote: Could you please explain the difference between BasketProfit and MinProfit in the GridMaster?

I must admit that this is confusing:

BasketProfit (Default: $10) : If the current floating profit, from all the trades of all the Dancers together exceeds this amount, GridMaster will send the instruction to all the Dancers to close all their trades and start with a new sequence.

EquityTP (Default: $50) : At the start of a new sequence, (i.e. when there are no trades open), GridMaster makes a note of the starting Equity (which is in most cases also the starting Balance, unless there are other EAs at work on the account). If, due to several BasketProfit's and/or Subset Closures (below), the current equity is now at least EquityTP above the starting Equity, amount, GridMaster will send the instruction to all the Dancers to close all their trades and start with a new sequence.

MinProfit (Default: $5): GridMaster is continuously on the lookout to close a Subset of trades, such that this will make it's life easier for the future. One particular 'problem' is the occurrence of a group of BUYs above the current market price and a group of SELLs below the current market price (of the same symbol). GridMaster 'tackles' this problem by closing the Highest BUY together with the Lowest SELL (these two trades together are in a floating loss), together with a set of positive trades (from other pairs) such that the total profit of this subset exceeds the MinProfit figure. This action does not close all trades and GridMaster won't instruct the Dancers to start a new sequence.

Hope that this explains it all.

Post by **szfxtrader** » Mon Oct 24, 2016 7:06 am

I would like to know if the account has more than one EA running, will GM only take care of the trades of GD(same magic number) or it will consider all open trades? Thank you in advance for your reply and help.

Post by **renexxxx** » Mon Oct 24, 2016 7:29 am

szfxtrader » Mon Oct 24, 2016 5:06 pm wrote:I would like to know if the account has more than one EA running, will GM only take care of the trades of GD(same magic number) or it will consider all open trades? Thank you in advance for your reply and help.

You can configure GM with a list of magic numbers for it to manage. However, GM is no magician: if the entries are no good, it will be impossible for GM to close them in profit.

Post by **acostafulano** » Mon Oct 24, 2016 11:05 am

Thanks for the explanation Rene,

Another quick question, as I'm trying to ballpark what type of closures to look for.

As per your default settings and initial results reported in the HGIDance reboot post, you're going for US$5 MinProfit (Closures of subset of trades for US$5) using 0.05 lot sizes.

This would mean, speaking in rough terms, subsets of more or less 10 pips.

Could you please confirm if this logic is correct?

I'm looking to test using 0.01 lots so I want to gauge what type of profits to aim for! (will probably try 3 different settings)

Thanks!

Diego

Post by **renexxxx** » Mon Oct 24, 2016 11:28 am

acostafulano » Mon Oct 24, 2016 9:05 pm wrote: This would mean, speaking in rough terms, subsets of more or less 10 pips.

Yes, that is correct.

acostafulano » Mon Oct 24, 2016 9:05 pm wrote: I'm looking to test using 0.01 lots so I want to gauge what type of profits to aim for! (will probably try 3 different settings)

As a start, you could divide every $-amount by 5, if you are using 0.01 lots.

Post by **acostafulano** » Mon Oct 24, 2016 11:39 am

THank you René.

One small suggestion regarding HGI. (I've been trading HGBnGrid for several months now).

Perhaps include the option to only count TREND or RAD arrows, as I've noticed RANGE arrows are really spammy and dont really work well for trending scenarios.!

René I also wanted to ask a question: will increasing MASTERTIMERSECONDS reduce CPU usage? would you reccomend doing so?

Diego

Post by **szfxtrader** » Mon Oct 24, 2016 5:03 pm

renexxxx » Mon Oct 24, 2016 7:29 am wrote: You can configure GM with a list of magic numbers for it to manage. However, GM is no magician: if the entries are no good, it will be impossible for GM to close them in profit.

Thank you for the information, I am using 0.02 lots/4 pairs, and today until now it made around 0.13%(OOTB).

Post by **renexxxx** » Mon Oct 24, 2016 7:24 pm

acostafulano » Mon Oct 24, 2016 9:39 pm wrote: Perhaps include the option to only count TREND or RAD arrows, as I've noticed RANGE arrows are really spammy and dont really work well for trending scenarios.!

I will add this to the next version.

acostafulano » Mon Oct 24, 2016 9:39 pm wrote: René I also wanted to ask a question: will increasing MASTERTIMERSECONDS reduce CPU usage? would you reccomend doing so?

Yes, it would (reduce CPU usage). I am running 1 GridMaster and 7 GridDancers on one terminal (together with some trade analysis tools), on a crappy 5 year old PC with Windows 7, and my CPU usage is less than 2% (this is with the default 10 seconds for MasterTimerSeconds). By reducing the frequency at which GridMaster is checking for trade closures, you would expect that GridMaster finds fewer trade closure opportunities, resulting in a slower balance increase. However, 'overclocking' GridMaster is not a good idea either. So, you need to find a good balance.