The gg53 "Killing Zone" | Page 100 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Broker Challenge "NDD/STP" vs. True "ECN" and "Mkt Mkr" broker talk 46 replies

Dealing with "Invalid Price", "Requote", "Server Busy" in MT4 3 replies

The Ultimate truth by gg53 4 replies

Metatrader "killing" 3rd party developers? 6 replies

It’s our job to trade "Futures" not "Histories", by Joe Ross 7 replies

Attachments: The gg53 "Killing Zone"
Exit Attachments

The gg53 "Killing Zone"

  • #1,981
  • Oct 17, 2013 2:31pm Oct 17, 2013 2:31pm
Quoting FusionF
Kilian I would also like to thank you for your contribution and willingness to create an EA for gg53's strategies. And of course a big thank you to gg53 for sharing it with us and keeps sharing his ideas. So to summarize the strategy only for the Asian Breakout (not the killing zone in general). Timeframe: 15 min Indicators: Gadi_OBV_v2.2 and Gadi_TickVolume_v2.2 for confirmation. Long entry: Breakout of the OBV channel (plus the PercentIncrease) and Gadi_TickVolume_v2.2 is above 0. Short entry: the opposite. SL: The last swing of the VOLUME in...
the white Difference line in Gadi_Currencies v 3.0 Diff needs to be pointing in the same direction as the direction of the trade upon entry.
  • #1,982
  • Oct 18, 2013 12:36am Oct 18, 2013 12:36am
Quoting panamamike
{quote} the white Difference line in Gadi_Currencies v 3.0 Diff needs to be pointing in the same direction as the direction of the trade upon entry.
panamamike,

Using the Gadi_Currencies Diff. is optional (per my previous instructions) - for extra confirmation.
It's not mandatory for this strategy.

G.
  • #1,983
  • Oct 19, 2013 11:46am Oct 19, 2013 11:46am
Here are the rules on one chart. (I cheated by trading on Friday which per system rules is not allowed.)

Gadi_TickVolume being angled upwards is optional and not required. Confirmation from Gadi_Currncies is also optional.

BTW--spread on m broker was much lower when I took the trade. It goes up right at close of business Friday.
Attached Image (click to enlarge)
Click to Enlarge Name: audusd_fxm15 oct 18 entry rules.jpg Size: 317 KB
  • #1,984
  • Edited 2:16pm Oct 19, 2013 12:21pm | Edited 2:16pm
The EA will only take the first trade after the channel creation now.
Start trading time was removed.
The rules to close a trade after the 4th candle is optional now.


Quote
SL: The last swing of the VOLUME in Gadi_OBV.
this will be part of future releases as a swing is really objective. I have to think about it a little more.


The pip value (pipette value to be correct) will be drawn at the indicator.
(optional) As requested via pn there is an alert when an arrow appears. (The alert will only be activated once per candle).
The live arrow creation algo got fixed.

There might still be errors so please test and tell me what doesn't work. The new EA only works with the new version of the indicator!!


See Post 1989 for the EA.
Attached Image (click to enlarge)
Click to Enlarge Name: Chart.png Size: 56 KB
Attached File(s)
File Type: ex4 Gadi_OBV_mod.ex4   8 KB | 879 downloads
  • #1,985
  • Oct 19, 2013 12:24pm Oct 19, 2013 12:24pm
Quoting Kilian19
The EA will only take the first trade after the channel creation now. Start trading time was removed. The rules to close a trade after the 4th candle is optional now. I won't tackle the SL before there are no clear rules defined. The pip value (pipette value to be correct) will be drawn at the indicator. (optional) As requested via pn there is an alert when an arrow appears. (The alert will only be activated once per candle). The live arrow creation algo got fixed. There might still be errors so please test and tell me what doesn't work. The new...
Awesome! Thank you so much for your efforts.
  • #1,986
  • Oct 19, 2013 12:37pm Oct 19, 2013 12:37pm
Quoting Dewey McG
{quote} Awesome! Thank you so much for your efforts.
upps.
!Caution I think I have uploaded a wrong version in the previous post. Just download it again and it should be fine.
  • #1,987
  • Oct 19, 2013 12:48pm Oct 19, 2013 12:48pm
Quoting Kilian19
{quote} upps. !Caution I think I have uploaded a wrong version in the previous post. Just download it again and it should be fine.
Thanks for the heads up.

Haven't started on the EA yet but the indicator alone is amazing. It makes visual back testing easy and the alerts will help to trade more pairs. Big thanks to both you and GG53. I can't wait to trade next week!
  • #1,988
  • Oct 19, 2013 1:16pm Oct 19, 2013 1:16pm
I started trying to test the EA but there seems to be a bug. The EA took no trades last week so I tried visual mode and as you can see from the PIC it doesn't take the trade when a clear signal is given.
Attached Image (click to enlarge)
Click to Enlarge Name: eurusd_fxm15 ea no trade.jpg Size: 216 KB
  • #1,989
  • Edited 2:17pm Oct 19, 2013 1:21pm | Edited 2:17pm
Quoting Dewey McG
I started trying to test the EA but there seems to be a bug. The EA took no trades last week so I tried visual mode and as you can see from the PIC it doesn't take the trade when a clear signal is given. {image}
I will take a lock into it in about an hour or so

Edit: Fixed... I am sorry I don't know where my thoughts were when I coded that stuff .... .....

The only untested thing on my parts at the moment is the trailing stop. have an eye on that please-

See post 1984 for the indicator and the instructions ...
Attached File(s)
File Type: ex4 Asian Breakout (gg) a.ex4   7 KB | 524 downloads
  • #1,990
  • Oct 19, 2013 2:44pm Oct 19, 2013 2:44pm
Quoting gg53
{quote} whydak, Yes, Good catch. I was looking for an indicator to draw it automatically (divergence indie) before posting this feature... You are ahead of me... ...and it doesn't have to be on higher TF's - see Attached (M15). Just another proof that Gadi_currencies is a predictive tool. Thanks, whydak. G. {image}
This is older but have you perused this? Letting indicators draw divergence is actually pretty simple. There are some good examples over at codebase and it's just a matter of changing 3 or 4 lines of code.
  • #1,991
  • Oct 19, 2013 3:50pm Oct 19, 2013 3:50pm
Quoting Kilian19
{quote} I will take a lock into it in about an hour or so Edit: Fixed... I am sorry I don't know where my thoughts were when I coded that stuff .... ..... The only untested thing on my parts at the moment is the trailing stop. have an eye on that please- See post 1984 for the indicator and the instructions ... {file}

I think I found a few bugs. I ran a back test just for the last week and compared with the charts
Attached Image(s) (click to enlarge)
Click to Enlarge Name: results.jpg Size: 126 KB
Click to Enlarge Name: trade 1.jpg Size: 245 KB
Click to Enlarge Name: trade 2.jpg Size: 214 KB
Click to Enlarge Name: trade 3.jpg Size: 207 KB
Click to Enlarge Name: trade 4.jpg Size: 193 KB
  • #1,992
  • Oct 19, 2013 4:24pm Oct 19, 2013 4:24pm
Quoting Dewey McG
{quote} I think I found a few bugs. I ran a back test just for the last week and compared with the charts
I don't see any bugs. I think the EA did exactly what it was told to do. If we want to adjust the rules we need to state them before that.


#1: Well it's an EA after all. Either there is a rule to take a trade or there is not. If you want any changes please make clear suggestions

#2 There is already an build in option to see if the candle is unusually big.

if( iATR(Symbol(),0,1,0) >= AtrMult * iATR(Symbol(),NULL,3,1)){trading = "Range of current candle is to high"; return; }

Are you sure the candle was already that huge when the EA took the trade? I can just look at the current values and not at what would have, when the candle alredy closed.

#3 The breaking question. Maybe the OBV has breached the channel for a split second and returned back to it. You should look at that during visual mode. The EA just takes trades when the obv crosses the channel. The indicator just shows the condition at candle close!.

#4 Why did it close so early. Did it touch the SL? It's hard to judge when I don't see the entry and exit points on the chart. Currently the SL is defined as the Close of the previous candle. If that is to close take the Low. If that is still to close take the value entered in the settings.

#5 same as number 3 I guess.
  • #1,993
  • Oct 19, 2013 5:26pm Oct 19, 2013 5:26pm
"Bugs" was the wrong term. I would wait for GG53 to comment before doing anything but here are my suggestions:

1. The initial SL is too tight. I was using the previous swing low and if it was too close I would use the one before. I then switched to the low of the candle where the last swing low or high for the Gad-I_OBV was. The trade needs a little more room to breathe.

2. I know the rules per the system is that the Gadi_TickVolume must be above 0 for a long and below for a buy, but I take the trade manually if the angle is steep and in the right direction. I would wait to see what GG thinks though.

3. I did see a couple spikes on visual back testing that happened so fast I missed the first time and also when I was trading. I would suggest a delay of some sort--maybe just a minute even-- to see if it is still above or below the channel.
  • #1,994
  • Oct 19, 2013 5:40pm Oct 19, 2013 5:40pm
Quoting Dewey McG
....
I didn't want to sound "agressive" in my last post. Sorry if that gave you wrong impression.

as I said I will look into the SL again tomorrow. I think the increase parameter was used to create a slight buffer. Maybe increasing the value will do the job. Regards the Tick volume. We could calculate a regression of the past 3 or 4 values and predict if the next bar would be below the 0 line .... Just an example. Let gg speak.
  • #1,995
  • Oct 19, 2013 5:59pm Oct 19, 2013 5:59pm
Quoting Kilian19
{quote} I didn't want to sound "agressive" in my last post. Sorry if that gave you wrong impression. as I said I will look into the SL again tomorrow. I think the increase parameter was used to create a slight buffer. Maybe increasing the value will do the job. Regards the Tick volume. We could calculate a regression of the past 3 or 4 values and predict if the next bar would be below the 0 line .... Just an example. Let gg speak.
You sounded fine--I just wanted to make sure I didn't sound rude.

I like both your ideas--will be good to see what GG says.

If this works I don't want my wife to find out--I need her to think I am working
  • #1,996
  • Oct 19, 2013 10:39pm Oct 19, 2013 10:39pm
Just some food for thought for the EA:

Visual back-testing is limited because it only shows close of bar and not what happened on the tick.

However, if you traded only signals on the close of bar on the EU, 21 of the last 24 trades would have been winners. I can't go back further because of the limitations of the indicators. The losers would have been much smaller than the winners.

All 3 losers had a signal in the opposite direction afterwards (The Gadi_OBV crossed over the opposite side of the channel). Each of these would have been a winner.
  • #1,997
  • Oct 20, 2013 2:00am Oct 20, 2013 2:00am
Quoting Dewey McG
Just some food for thought for the EA: Visual back-testing is limited because it only shows close of bar and not what happened on the tick. However, if you traded only signals on the close of bar on the EU, 21 of the last 24 trades would have been winners. ...
Hello everybody,

First I would like to thank gg53 for enlightening us with his view on volume and for sharing, and to all of you for the teamwork.

Just as a suggestion: for the purpose of Kilian19's EA (appreciate it), I believe this specific issue with the large candle may be resolved by looking to the close on a lower timeframe (e.g. 5min TF if working on 15 min TF, 15 min TF if working on 1H TF etc.).
So if the the specific conditions are met on the lower timeframe, thus having a valid trade, we may benefit from entering earlier than we would on the current timeframe.
gg53 - while is your system and rules, please be kind to let us know your opinion on it.

Best regards,
Dragos
  • #1,998
  • Oct 20, 2013 12:12pm Oct 20, 2013 12:12pm
Quoting dragos_
{quote} Hello everybody, First I would like to thank gg53 for enlightening us with his view on volume and for sharing, and to all of you for the teamwork. Just as a suggestion: for the purpose of Kilian19's EA (appreciate it), I believe this specific issue with the large candle may be resolved by looking to the close on a lower timeframe (e.g. 5min TF if working on 15 min TF, 15 min TF if working on 1H TF etc.). So if the the specific conditions are met on the lower timeframe, thus having a valid trade, we may benefit from entering earlier than...
dragos,

Entering earlier on lower TF's is good practice, especially if you are entering on bar close (not lower than M5).
I'm entering on M15, manually, on Gadi_OBV breakout - without waiting for bar close.
I didn't made statistics which way is better - i.e. enter immediatly or on bar close.

Entering immediatly on breakout causes challenges in EA programming. You need to confirm the "breakout" in either a larger boundry percentage or count the number of ticks of valid signal.

G.
  • #1,999
  • Oct 20, 2013 12:41pm Oct 20, 2013 12:41pm
Quoting Dewey McG
However, if you traded only signals on the close of bar on the EU, 21 of the last 24 trades would have been winners. I can't go back further because of the limitations of the indicators. The losers would have been much smaller than the winners. All 3 losers had a signal in the opposite direction afterwards (The Gadi_OBV crossed over the opposite side of the channel). Each of these would have been a winner.
I will watch a movie and after that I will take a look at the EA again. Just letting the EA take a trade on the close of a candle isn't much of a code change). And back testing will be a million times faster
  • #2,000
  • Edited 3:28pm Oct 20, 2013 3:17pm | Edited 3:28pm
This version will now just take a trade at the open of a new bar looking back at the last bar for the signal. If you open one at the close or the open of the next candle should be the same thing right ?

Quote
First I would like to thank gg53 for enlightening us with his view on volume and for sharing, and to all of you for the teamwork.

Just as a suggestion: for the purpose of Kilian19's EA (appreciate it), I believe this specific issue with the large candle may be resolved by looking to the close on a lower timeframe (e.g. 5min TF if working on 15 min TF, 15 min TF if working on 1H TF etc.).

Also this should be not an issue for this version as we now have the time to look at the closed candle.

Dewey would you mind telling me where you see the profitable trades? I can't seem to find them ...
Attached File(s)
File Type: ex4 Asian Breakout (gg) open.ex4   8 KB | 561 downloads
1 trader viewing now