Try My EA! Need Group Feedback... | Page 34 | Crypto Craft

Options

Go to
Permalinks
Print Thread

Related Instruments

Gold/USD

Similar Threads

USD/TRY and EUR/TRY 12 replies

Fibo Group Feedback 23 replies

Need feedback on my trading breakouts system 8 replies

One World Capital... Need feedback 2 replies

Need feedback on FXCM Educational workshops 3 replies

Attachments: Try My EA! Need Group Feedback...
Exit Attachments

Try My EA! Need Group Feedback...

  • #661
  • Dec 30, 2014 10:16am Dec 30, 2014 10:16am
Quoting dgodfrey
{quote} Hey Kosmo, no it's not that it's not there - it's just that you set it to a nominal value. Also, it doesn't factor into the step size, that's controlled completely by GridStep. It does factor into the BE and Targets a little, as well as the middle distance between the 2 halves of the grid.
Ok, got it, thanks.

Following the scary experience from Friday, I have another thought... would it be possible to devise and code in a formula, which will calculate how many pendings are needed on the other side to balance the opposite trades already open, so that when price changes direction from, say, sell to buy (after already having triggered a good portion of sells), and reaches the other side of the box (buy), it'll still hit the target without price 'getting out' of the box? In effect, it'll add pendings to the other side of the box on the go, based on how many opposite trades are already open. I'm sure it'd be a great weapon against getting boxed out, as long as there's margin left, of course.
mind over matter
  • #662
  • Dec 30, 2014 10:20am Dec 30, 2014 10:20am
Quoting dgodfrey
Hey Kosmo, are you having any trouble with executions on your grid when there are sudden moves? I just looked and the problem is that all my trades are bunching at the same levels - I have 5 of the grid buystops executed within a few cents of each other at 1206.60. That's why I'm not reaping the benefits.
Haven't really looked into it, but I guess I'd better do, before going live, lol!
mind over matter
  • #663
  • Dec 30, 2014 10:25am Dec 30, 2014 10:25am
Quoting kosmo
{quote} Ok, got it, thanks. Following the scary experience from Friday, I have another thought... would it be possible to devise and code in a formula, which will calculate how many pendings are needed on the other side to balance the opposite trades already open, so that when price changes direction from, say, sell to buy (after already having triggered a good portion of sells), and reaches the other side of the box (buy), it'll still hit the target without price 'getting out' of the box? In effect, it'll add pendings to the other side of the box...
That's not a bad idea - the mechanism is implicit in the calculation anyway. The algorithm assumes that those orders are there even if they're not. All you would have to do is, if the target is outside the box, have the program fill in the trades at least to the target. Sounds pretty doable. In fact, the program could just use that to build the box on the fly, starting with a small box (and less trades to cancel) and increasing it as required to cover the hedge.
A good man knows his limitations ~ Clint Eastwood
  • #664
  • Dec 30, 2014 10:33am Dec 30, 2014 10:33am
Quoting dgodfrey
{quote} That's not a bad idea
I do beg your pardon, Sir. This is an understatement!

Fantastic! I'll leave the algebra to you then.
mind over matter
  • #665
  • Edited 11:06am Dec 30, 2014 10:35am | Edited 11:06am
Quoting Thepipster
{image} {quote} Excellent!!! ~Dgodfrey...thanks for your updates! I'll wait until we can have a version that places TP and SL at the calculated profit line before downloading and starting a new demo with it. People can test your version to be sure of the math. I don't have time to view my platform on VPS, so I wouldn't be able to observe the comments to see it for myself (sorry). I think it will help us be able to see any improvements we can make to gain an edge.
I'll think about how to do what you want there, Pipster. I need to make an improvement in the program anyway, to poll the actual orders and get that value built into the calculation on what BE and Target is. Could pull that all together with building out the box, and have all the open orders set to take profit or stop out at the target.
A good man knows his limitations ~ Clint Eastwood
  • #666
  • Dec 30, 2014 10:36am Dec 30, 2014 10:36am
Quoting kosmo
{quote} I do beg your pardon, Sir. This is an understatement! Fantastic! I'll leave the algebra to you then.
And here I thought the British were the masters of understatement
A good man knows his limitations ~ Clint Eastwood
  • #667
  • Dec 30, 2014 10:41am Dec 30, 2014 10:41am
Quoting dgodfrey
{quote} And here I thought the British were the masters of understatement
Don't know about that. I only live here, lol! But my roots are Slavic, hence the creativity... (another lol!).
mind over matter
  • #668
  • Dec 30, 2014 10:47am Dec 30, 2014 10:47am
Just to be clear, the BET version does calculate based on the current P/L, it's just that it uses the theoretical grid to find what the target and BE value needs to be. I will have to figure out a way to get the open trades worked in there to do an accurate calculation of the target rather the what they should have been doing had they been executed perfectly. Obviously, with all my trades bunched at 1206 instead of marching steadily upwards from 1201 to 1206, not as much will be contributed by those bad fills.
A good man knows his limitations ~ Clint Eastwood
  • #669
  • Dec 30, 2014 10:53am Dec 30, 2014 10:53am
http://www.kitco.com/images/live/nyg...07787682455684

~kosmo! Now you see what happens when you're patient!

~dgodfrey...looks like your demo trades more like a live account. Gold can have some slippage when live. I like the idea of starting with a smaller box, (maybe 10 orders?), and growing it. Hopefully doing that doesn't make the math harder. So, with your "BET" version, I could set the grid spaces smaller, since I could override the spread?
Too close for pips, switching to points.
  • #670
  • Dec 30, 2014 11:01am Dec 30, 2014 11:01am
Quoting Thepipster
So, with your "BET" version, I could set the grid spaces smaller, since I could override the spread?
Yes, the spread is no longer part of the grid step calculation. It should give you exactly what you set GridStep to be. Dambuster found that for 3 digit brokers the number needs to be 10x larger than what the 2 digit brokers need. I'll fix that in the next go round.

Example: GridStep = 50 gives $0.50 between steps for 2 digit brokers. Gridstep = 500 gives $0.50 between steps for 3 digit brokers. The same logic applies to Spread and Buffer as well.
A good man knows his limitations ~ Clint Eastwood
  • #671
  • Dec 30, 2014 11:19am Dec 30, 2014 11:19am
... as well as the equity target so mine is set to 10.0 which equates to 1% with my 3 digit broker
  • #672
  • Dec 30, 2014 11:39am Dec 30, 2014 11:39am
Quoting Thepipster
~kosmo! Now you see what happens when you're patient!
Absolutely, but that's just one side of the coin; the other is believing in your settings.
mind over matter
  • #673
  • Dec 30, 2014 12:11pm Dec 30, 2014 12:11pm
End of year musings... is gold heading for $1250 (?)
Attached Image (click to enlarge)
Click to Enlarge Name: 1250.png Size: 41 KB
mind over matter
  • #674
  • Dec 30, 2014 12:38pm Dec 30, 2014 12:38pm
@ Kosmo - we can hope!

How's this for slippage? Instead of another gain, the %gain has gone from over 20% to just over 8%
Attached Image (click to enlarge)
Click to Enlarge Name: XAUUSDM15.png Size: 64 KB
A good man knows his limitations ~ Clint Eastwood
  • #675
  • Dec 30, 2014 1:07pm Dec 30, 2014 1:07pm
Just watched the BET version sweep past the sell BE prediction, and noticed that it was when the ASK price hit the BE value it was pretty much even money. So that number is off by the broker spread if your chart is only displaying the bid line.
A good man knows his limitations ~ Clint Eastwood
  • #676
  • Dec 30, 2014 1:13pm Dec 30, 2014 1:13pm
Quoting dgodfrey
How's this for slippage? Instead of another gain, the %gain has gone from over 20% to just over 8% {image}
Ouch! What can be said other than F.XCM is a joke (with all due respect, lol)!

Maybe it's time to do some research. They're trying their best there and the site is a good starting point, but some of the info needs to be taken with a pinch of salt and confirmed at the source. Suffice to look at Armada's spread info...
Attached Image (click to enlarge)
Click to Enlarge Name: 100fb.png Size: 9 KB
mind over matter
  • #677
  • Dec 30, 2014 1:21pm Dec 30, 2014 1:21pm
Quoting dgodfrey
Just watched the BET version sweep past the sell BE prediction, and noticed that it was when the ASK price hit the BE value it was pretty much even money. So that number is off by the broker spread if your chart is only displaying the bid line.
Good to know. I'll tick the 'show Ask line' in chart properties.
mind over matter
  • #678
  • Dec 30, 2014 1:42pm Dec 30, 2014 1:42pm
Quoting kosmo
End of year musings... is gold heading for $1250 (?) {image}
Well it needs to breakout somewhere - hopefully the buy side as I've three trades open (on demo). Seems to be hovering around 1200 at the moment.

Coin toss......
  • #679
  • Dec 30, 2014 2:21pm Dec 30, 2014 2:21pm
Quoting kosmo
{quote} Ouch! What can be said other than F.XCM is a joke (with all due respect, lol)! Maybe it's time to do some research. They're trying their best there and the site is a good starting point, but some of the info needs to be taken with a pinch of salt and confirmed at the source. Suffice to look at Armada's spread info... {image}
You are right about that. Just went through the list of brokers what I can get a live account with - not exactly encouraging lol. I am going to give IKON a shot, at least their spread is a little tighter (or so they say).

The good news is my BET version (probably) cycled, according to where the market is and what the target was at lunchtime. I got a 10% cycle on that account this morning because I had the wrong target - and it got hit anyway I'm back to the more sane target now, and this is the second hit on it at the 1% level. I would link it to an explorer but I seem to have lost the investor password. Guess I'll start it over so I can.
A good man knows his limitations ~ Clint Eastwood
  • #680
  • Dec 30, 2014 2:34pm Dec 30, 2014 2:34pm
3rd cycle today just closed... wow
1 trader viewing now