Vegas Algorithm auto-trader by Steve and Squalou | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Algorithm for Belkhayate Timing and Barycenter 0 replies

Anyone wants to program this profitable algorithm into an EA? 7 replies

what type of an algorithm is this? 0 replies

Question for VEGAS and others using Vegas system..... 2 replies

Attachments: Vegas Algorithm auto-trader by Steve and Squalou
Exit Attachments

Vegas Algorithm auto-trader by Steve and Squalou

  • #41
  • Edited 3:52pm Nov 10, 2010 3:41pm | Edited 3:52pm
Quoting Zr-fx
Hey Steve and SQ

The EA looks good in operation. However in the interest of testing I ran it in the backtester with $100k pot and 1 lot. Everything else default. Although backtesting is not reliable but it produced an interesting result. A very smooth negative curve

I don't think I've made a mistake but I'm sure someone will let me know!

I think the majority of the losses are due the trade being closed as soon the MAs cross back in the opposite direction. But I'm note certain of this.

Is a way to be able to disable this close...
ZR
can u run a backtest for todays trading and see if EU picked up a winner in the afternoon. Its one i watched live and saw it hit target on the aqua line. Be interesting to see if the backtest took out there or when the MAs crossed back at each other. trade entry time was at 14.32 GMT.
The Kamakazi
  • #42
  • Nov 10, 2010 4:07pm Nov 10, 2010 4:07pm
Quoting Bl888
ZR
can u run a backtest for todays trading and see if EU picked up a winner in the afternoon. Its one i watched live and saw it hit target on the aqua line. Be interesting to see if the backtest took out there or when the MAs crossed back at each other. trade entry time was at 14.32 GMT.

Here it is. It does pick up a cracker of a trade but only after a couple of losses finishing the day break even.
Attached Image (click to enlarge)
Click to Enlarge Name: Vegas 10112010.gif Size: 8 KB
Attached File(s)
File Type: pdf Vegas 10112010.pdf   74 KB | 346 downloads
  • #43
  • Nov 10, 2010 4:17pm Nov 10, 2010 4:17pm
Nice One SQ,
I took a manual EURUSD trade on one computor and the robot took the same trade on the other computor by itself and this time it held on to it.......Where do I send the gift......

whip
  • #44
  • Nov 10, 2010 4:42pm Nov 10, 2010 4:42pm
I ran a simple count of the trades TYPE from results of the year to date BT:

571 Trades
108 (18.91%) s/l
1 (0.18%) t/p
462 (80.91%) close. Of those 81% where a loss.

I may be completely off the mark here but it seems 80% of trades are closed by the EA, such as when the MA's cross back or other conditions (hitting the outer band?). Of those 81% closed at a loss.

I think it is worth testing with this feature disabled just to see what happens.
Another thought is perhaps only close the trades when the MAs cross back if it is in profit or BE. This is just off the top pf my head in the interest of improving things. Not based on any analysis other than the above so please be gentle if you want me to shut up
  • #45
  • Nov 10, 2010 5:08pm Nov 10, 2010 5:08pm
Quoting Zr-fx
I ran a simple count of the trades TYPE from results of the year to date BT:

571 Trades
108 (18.91%) s/l
1 (0.18%) t/p
462 (80.91%) close. Of those 81% where a loss.

I may be completely off the mark here but it seems 80% of trades are closed by the EA, such as when the MA's cross back or other conditions (hitting the outer band?). Of those 81% closed at a loss.

[font=Calibri]I think it is worth testing with this feature disabled just to see...
interesting stats. Looking at the m5 charts manually looks like entrys on this system are perfect with great potential. Now the next task is to test the trade management. As this is a scalping algo as he calls it, then it may be worth having a tigher stop and jumping stop. Ive left mine with 20 pips stop. 88 pips TP (my lucky number ) and jumping stop of 10.

lets see how it gets on.
The Kamakazi
  • #46
  • Nov 10, 2010 5:20pm Nov 10, 2010 5:20pm
Quoting Zr-fx
I ran a simple count of the trades TYPE from results of the year to date BT:

571 Trades
108 (18.91%) s/l
1 (0.18%) t/p
462 (80.91%) close. Of those 81% where a loss.

I may be completely off the mark here but it seems 80% of trades are closed by the EA, such as when the MA's cross back or other conditions (hitting the outer band?). Of those 81% closed at a loss.

[font=Calibri]I think it is worth testing with this feature disabled just to see...
Totally agree with your findings Zr-Fx from my limited testing. As far as I understand Vegas's method, he sets a SL that gives a bit of room and doesn't get out of the trade when the EMA's cross back and forth over each other for a bit.

I just had a demo trade on USDJPY that set a good SL of 25pips but closed out pretty quickly with about a 3pip loss due to the EMA's crossing again.

Steve, is there a way we could add a 'EMA Cross Buffer' to the EA which won't close the trade if it is within that buffer from the price the trade was opened? Just an idea.
  • #47
  • Nov 10, 2010 5:40pm Nov 10, 2010 5:40pm
Quoting Bl888
Viv

The PDF file tells you everything about the system. I suggest you read that and learn what it is suppose to do.

Also, backtests are not reliable. Best bet is run it in demo and conclude yourself if its good.

Hi Bl888

Not getting my handbag out - but i had read the manual. It describes the 4H and 5min charts - but this is for manual trading. I didnt see anywhere any specific instructions for the EA itself, which is why i asked.

I wish i had the time to trade the manual method, but life is such a bitch.

Yes - i agree, back tests are not reliable. They just tell us that the EA is doing what it is supposed to do.. but so far, results are not telling us this either!

Thanks for your response.
  • #48
  • Nov 10, 2010 6:17pm Nov 10, 2010 6:17pm
Quoting Viv108
Hi Bl888

Not getting my handbag out - but i had read the manual. It describes the 4H and 5min charts - but this is for manual trading. I didnt see anywhere any specific instructions for the EA itself, which is why i asked.

I wish i had the time to trade the manual method, but life is such a bitch.

Yes - i agree, back tests are not reliable. They just tell us that the EA is doing what it is supposed to do.. but so far, results are not telling us this either!

Thanks for your response.
Stop now. You are heading for trouble. "Guys.. what time frame are you using?" came close to getting a ban from me.
  • #49
  • Nov 10, 2010 6:22pm Nov 10, 2010 6:22pm
Guys, have a play with this one. It turns off the bot's ability to close trades and relies purely on sl/tp/mptm

Attached File(s)
File Type: mq4 Vegas auto trading robot by Steve and Squalou no robot close.mq4   47 KB | 327 downloads
  • #50
  • Nov 10, 2010 6:22pm Nov 10, 2010 6:22pm
Great!
  • #51
  • Nov 10, 2010 6:26pm Nov 10, 2010 6:26pm
Quoting SteveHopwood
Guys, have a play with this one. It turns off the bot's ability to close trades and relies purely on sl/tp/mptm

Please consider uploading this version in Post 1.

福 子 Victor V
  • #52
  • Nov 10, 2010 6:40pm Nov 10, 2010 6:40pm
Quoting camvcvoo
Please consider uploading this version in Post 1.

Done.

  • #53
  • Nov 10, 2010 9:38pm Nov 10, 2010 9:38pm
That was the hardest pdf to read ever. Read it 4 times and I'm sitll wondering if I set up this bot right haha.
  • #54
  • Nov 11, 2010 12:26am Nov 11, 2010 12:26am
Thanks for the 'no robot close' version Steve.

Be interesting to see how forward testing goes with this. Backtesting hasn't been good no matter what settings I used. The best results I could get was from changing the trend detection code from H4 to H1, which obviously highly differs from Vegas's entries.

Fingers crossed going forward.
  • #55
  • Nov 11, 2010 1:50am Nov 11, 2010 1:50am
Quoting SteveHopwood
Guys, have a play with this one. It turns off the bot's ability to close trades and relies purely on sl/tp/mptm

Makes more sense because we are always going to get some sort of retracement and this will hopefully stop it
I have also widened my SL but kept my TP at 25= RR21
Im still playing with ideas of the beast (to limit DD)
If I find a good idea I will let you know
Thats why im not participating much here in testing talk although Im running the bot with you people at the same time and getting the same sort of results

UPDATE

Wowsa Wowsa -- things look good now and also have a look at the 4H tunnel method , that seems to work although forward testing will be a slow thing
  • #56
  • Nov 11, 2010 2:13am Nov 11, 2010 2:13am
Steve + SQ

I was watching EU trade just now sell EU @ 7.32 GMT which had 20 pip stop. the bot closed out the deal before the stop as the MA crossed for an instant and then uncrossed itself before it blasted onside 20+ pips.

As this is a MA based system, an option maybe to close the trade on the MA recross at the close of that m5 candle.
The Kamakazi
  • #57
  • Nov 11, 2010 2:35am Nov 11, 2010 2:35am
Well, let's see if I can get banned on my first post.
Running 'no close' OOTB (except for lot size to satisfy my crim) and get failed order sends due to 'invalid stops'. Any suggestions?
  • #58
  • Nov 11, 2010 2:56am Nov 11, 2010 2:56am
Quoting Pinhead
Well, let's see if I can get banned on my first post.
Running 'no close' OOTB (except for lot size to satisfy my crim) and get failed order sends due to 'invalid stops'. Any suggestions?
have u set criminal ECN to true?
The Kamakazi
  • #59
  • Nov 11, 2010 3:08am Nov 11, 2010 3:08am
I have now.
Thank you.
  • #60
  • Nov 11, 2010 3:17am Nov 11, 2010 3:17am
Quoting Bl888
have u set criminal ECN to true?
Bl888, is that really you in the avatar logo !!!!!

I want some Y fronts like that for christmas , where did you get them from
0 traders viewing now