I fortunately am retired and do not have to leave my computer to monitor my trades.
- #6,822
- Jul 14, 2010 9:42am Jul 14, 2010 9:42am
No trade with Alpari live or demo, no trade with Fxcbs live either!
The price moved so smoothly there was no chance of an M15 upper fractal.
- Both MAs pointing up so we get 2 upward pointing blue arrows and the first thumbs up
- At point A a 15min fractal high forms and this would have generated thumbs up number two
- At point B price has pulled back, but the two bars have exactly the same low so no 5min fractal is detected (must be a price low which is lower than the 2 price lows immediately before and after it)
- At point C there is a small pullback which is a 5min fractal low but this will not locked until point D (3 bars later)
- At point D the high is higher than the 15min fractal at point A and this unlocks the 15min fractal turning it from green to HotPink. This causes the 2nd thumbs up to turn to a thumbs down
- Even though there is a valid 5min fractal at point C, the 15min fractal has been unlocked so no trade ...
So what does this all mean ? I think the MQL4 iFractals function used for detecting fractals only detects basic fractals i.e. must be a price high/low which is higher/lower than the 2 price highs/lows immediately before and after it. I think at the very least having two equal price lows should be treated as a single price low such that a fractal will be recorded if those lows are lower than the 2 price lows immediately before and after those lows.
It will be relatively straightforward to replace the use of iFractals function with something a little smarter. However, it remains to be seen how often this kind of thing has happened this year - maybe this is a one off. If either of those 2 bars had been 0.00001 lower then we'd all be happy ...
Regards,
Piptician
Tested again with this settings :
GMT offset 1
Start_time 23
Broker GMT is +1 so the Tokyo starts at 1.00
What am I doing wrong?Or maybe it is normal results? ;/
Test is about 1.5 month long. Here is the results:
http://www.ipix.lt/images/13699309.gif
GMT offset 1
Start_time 23
Broker GMT is +1 so the Tokyo starts at 1.00
What am I doing wrong?Or maybe it is normal results? ;/
Test is about 1.5 month long. Here is the results:
http://www.ipix.lt/images/13699309.gif
Actually there was an M15 upper fractal, it was the "awol" M5 fractal which caused a no trade. I have been analyzing why Alpari UK (Live) did not trigger today and here is what I discovered (refer to screenshot below):
[list][*]Both MAs pointing up so we get 2 upward pointing blue arrows and the first thumbs up
Still a little difficult for me to understand the fractals, but I am trying.
You claified a lot in your post here,
thankyou again for all you do to help everyone.
- #6,831
- Edited 1:44pm Jul 14, 2010 12:59pm | Edited 1:44pm
- | Commercial User | Joined Jul 2009 | 5,375 Posts
place this indicator on a chart, it will confirm what your gmt offset is.
Open H1 Chart (ANY PAIR) Place vertical line on the LAST BAR and see what time showing on your
chart... What Hour.? (forget about minutes)
Go here http://wwp.greenwichmeantime.com/
Check current GMT (hour only), and compare to your chart.. That's your offset..
If you current BAR chart Hour is 16 (+ some minutes) and GMT showing 20 (plus some minutes) then your GMT_Offset is "-4",
means your broker time 4 hours behind GMT.
Is this really that hard? What you need special indy for?
I don't get it, how you guys even flying between Countries in different timezones?
Cannot figure out +1 -1 hour.
6000+ posts in this thread and still the same questions...
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
- #6,832
- Jul 14, 2010 1:21pm Jul 14, 2010 1:21pm
- | Commercial User | Joined Jul 2009 | 5,375 Posts
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
- #6,834
- Jul 14, 2010 2:43pm Jul 14, 2010 2:43pm
This morning I get some data back which I lost yesterday. In the meantime again at 190 winning trades. So the curve is now like an S with lower magnitutes.....first 192 then 190, 186, now 190.....This seems to be very strange especially as I do think that asking Alpari for any official statement would bring close to zero in terms of clarification......
Does this mean backtesting results are even more limited in terms of their validity as we already thought?
Very sorry for my late respond but I was out off screen for a while.
Definitely agree with you, every my back tests always I try to confirm with other brokers and even that never trust in 100%.(database false).
Regards
Mac.
- #6,836
- Jul 14, 2010 3:25pm Jul 14, 2010 3:25pm
Hi Haishang
Very sorry for my late respond but I was out off screen for a while.
Definitely agree with you, every my back tests always I try to confirm with other brokers and even that never trust in 100%.(database false).
Regards
Mac.
no worries about a late reply plse.
it is just that I seem to spend too much time trying to find the best solution with backtesting and forgetting sometimes that we are living in the real world and not in statistics........
but difficult to always stay on the fine line!
- #6,838
- Jul 14, 2010 4:00pm Jul 14, 2010 4:00pm
- | Commercial User | Joined Jul 2009 | 5,375 Posts
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
