Griddy Wotsit Mk 2

[keydcuk]

Man, this is like watching paint drying......

Ha! I know what you mean. In the first part of my experiment with the GBP/JPY, I was closing a basket a day. Sunday, 1 basket. Monday, 1 basket. Tuesday, 1 basket. Then something happened... heh heh

I ruined the experiment as I started playing with the near 50 trades that were open. I started closing the positive and negative positions that would null themselves out with the tiniest of leaning in the positive profit. Bad mistake this was!

This did lead me to wonder though... Do you think maybe the GW works better during the early days of the week? Maybe as price tries to establish itself, more winners might take place in the early week?

I don't know... just a thought.

I deleted my demo account due to the ruination of my experiment and started all over again with a new demo account. I still wish I knew how to code as I got an idea to what I would like the EA to do but I am reluctant to ask Steve as it is his pet project, not mine.

Cheers!


Daniel

[Gertje]

Ha! I know what you mean. In the first part of my experiment with the GBP/JPY, I was closing a basket a day. Sunday, 1 basket. Monday, 1 basket. Tuesday, 1 basket. Then something happened... heh heh

This did lead me to wonder though... Do you think maybe the GW works better during the early days of the week? Maybe as price tries to establish itself, more winners might take place in the early week?

I think otherwise.
This 'strategy' is depending on high volatility to make a profit. At the moment the market is in waitingmode, maybe some news coming up, or whatever.
See attached chart of last night NZDUSD after they announced a lower intrest.
It dropped like bomb, only to go into a range again.
[In a few moments it might close another $10,- basket again]

NZDUSDM15.png

I still wish I knew how to code as I got an idea to what I would like the EA to do but I am reluctant to ask Steve as it is his pet project, not mine.

Steve doesn't think like that.
If you come up with proof or good reasoning why it might be a good idea, he will consider it for sure.
.

[Gertje]

I think otherwise.
This 'strategy' is depending on high volatility to make a profit. At the moment the market is in waitingmode, maybe some news coming up, or whatever.

There you go, ECB lowered the intrest from 0.05 to 0.00

[xirtamx]

The problem with the pendings being placed then deleted over and over again is a logic problem. I have attempted a fix in 1h in post 1, but am not sure that is will work.

Here is the attempted fix to the function if you want to make the update yourselves. It begins at line 1324:

Code: Select all

bool IsGapBetweenGroupsTooBig()
{
   //Return true if the gap between the lowest market buy and the lowest market sell is too large,
   //else return false
   
   int ExistingGap = 0;
   int PendingGap = 0;
   
   //Market trades
   if (PendingTradesTotal < (GridSize * 2) )
   {
      if (MarketBuys > 0)
         if (MarketSells > 0)
         {
            //More buys than sells. Market has risen so gap is at the bottom of the buys
            if (MarketBuys > MarketSells)
            {
               ExistingGap = (HighestMarketBuyPrice - HighestMarketSellPrice) * factor;
            }//if (MarketBuys > MarketSells)
            
            //More sells than buys. Market has fallen so gap is at the top of the sells
            if (MarketBuys < MarketSells)
            {
               ExistingGap = (LowestMarketBuyPrice - LowestMarketSellPrice) * factor;
            }//if (MarketBuys > MarketSells)
            
            if (ExistingGap > MaxDistanceBetweenGroups)
               return(true);
         }//if (MarketSells > 0)
   
      //Pending trades
      if (BuyStops > 0)
         if (SellStops > 0)
         {
            //More buys than sells. Market has risen so gap is at the bottom of the buys
            if (BuyStops > SellStops)
            {
               PendingGap = (LowestBuyStopPrice - HighestSellStopPrice) * factor;
            }//if (BuyStops > SellStops)
            
            //More sells than buys. Market has fallen so gap is at the top of the sells
            if (BuyStops < SellStops)
            {
               PendingGap = (LowestBuyStopPrice - HighestSellStopPrice) * factor;
            }//if (BuyStops > SellStops)
            
            //Equal numbers
            if (BuyStops == SellStops)
            {
               PendingGap = (LowestBuyStopPrice - HighestSellStopPrice) * factor;
            }//if (BuyStops == SellStops)
            
            
            if (PendingGap > MaxDistanceBetweenGroups)
               return(true);
         }//if (PendingSellStop > 0)
   
   }//if (PendingTradesTotal < (GridSize * 2) )
   


   //Got this far, so gap is ok
   return(false);

}//bool IsGapBetweenGroupsTooBig()

As feeback,
I don't think the fix has worked, I just saw the behaviour again with version 1 h

[DNTme]

Lots of errors for Invalid stops....while price was moving in EU GU

I think EA need volatility to make money, a reasonable volatility...too much volatily can kill it too.
ECB vols was not good for EA ...

[Gertje]

I think otherwise.
This 'strategy' is depending on high volatility to make a profit. At the moment the market is in waitingmode, maybe some news coming up, or whatever.

There you go, ECB lowered the intrest from 0.05 to 0.00

The dust settled, and I was able to access the VPS again...
4 pairs cleared all positions, adding 1700pips in less dan 5 minutes.....

The winner is EURUSD;

EURUSDM15.png

INSANE........
.

[SteveHopwood]

Here is my live trading so far today:

Hmmmmmmm.png

The sensible stuff is the EURGBP row - three trades filled yesterday and the fourth today after the pair moved decisively down and closed on the low semafor just before the market screamed up again.

The lunatic bit is the GJ row, which is GW at 0.01 lots.

I arrived home from teaching a few minutes ago to find, in round figures, 360 open trades. 280 sells and 180 buys. I have unfettered the take profit, so that was 720 pips.

The equivalent of 100 trades so not have to move far in the correct direction to rack up a lot of pips. The pair duly obliged and the position closed about 10 minutes later. I did not time exactly how long it too to close the position, but I reckon it was probably about a minute.

Not sure my balls are steely enough to trade like that for long.

Another hedged position just closed at +30. I will let the thing run for a while.

[Gertje]

At one point today I was up over $200 using 0.01 lots.....
Then the terminal went postal on me, and I had to install a new Empty4 to gain control over the trades again.
A huge amount of trades opened, and I killed everything, down $53 now for the day.

In 6 trading day's I'm still 42% up.
This is using 5pip grids, 8 pairs.

I started another demo yesterday, using 1/10th of the daily ATR100 as pip-distance and TP.
GBPJPY alone made 250pips with max DD of 100pips.

Adding 2.3% to the balance, I can live with a figure like that for years to come.

To be continued.....

[DNTme]

At one point today I was up over $200 using 0.01 lots.....
Then the terminal went postal on me, and I had to install a new Empty4 to gain control over the trades again.
A huge amount of trades opened, and I killed everything, down $53 now for the day.

In 6 trading day's I'm still 42% up.
This is using 5pip grids, 8 pairs.

I started another demo yesterday, using 1/10th of the daily ATR100 as pip-distance and TP.
GBPJPY alone made 250pips with max DD of 100pips.

Adding 2.3% to the balance, I can live with a figure like that for years to come.

To be continued.....

Like Steve said....would you have the balls deep enough to trade 100k account like that?

[Gertje]

Like Steve said....would you have the balls deep enough to trade 100k account like that?

I guess not.
I'll let you know in a year.....

Problem at the moment lies with Empty4.
It's getting clogged with 8 pairs trading so I intend to set up several terminals trading the same account during the weekend. Spreading the pairs across the terminals should lower the workload on each.
I have a dedicated 8Gb VPS for trading, and it's running 50-80% CPU at the moment...

What I also noticed is that most of the times during the night, positions accumulate in the several hedges, going nowhere and costing commission, swap and so on. My idea is to start to use the tradinghours and stop trading after a certain time. The EA will continue trading a hedge after this time, but stop starting a new fresh grid after the hedge has closed. Start trading again an hour before London open, and get going again.

GW added 3,5% already today......