0400-0700 GMT Breakout Strategy | Page 124 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

FXDD GMT+2 or GMT+3 10 replies

What does GMT shift and GMT shift Sun do in this indicator? 0 replies

MBO = Monday Break Out 8 utc gmt - 7 utc gmt 22 replies

What is the best server time for H4 trading? GMT+0, GMT+1 or GMT+2? 10 replies

web charting package with GMT +2 or GMT 2 replies

Attachments: 0400-0700 GMT Breakout Strategy
Exit Attachments

0400-0700 GMT Breakout Strategy

  • #2,461
  • Jan 30, 2010 3:16am Jan 30, 2010 3:16am
Quoting Pippopotamus
perhaps,...In Scullmeister method document in post 1 the time monitored for H/L is stated as 5-7:15, with no trade taken until 7:30. In post 2139 the time as stated is 5-7:00. Could be important.
I stick with the original method = 5 to 7:30.

Scull changed it later to 7:15 , but SL became box size ahd changed it back to 40 again.

Some months his original system does better than the one he has now and some months it doesn't.

The only reason I stick with the original on is that I an uncomfortable with the R:R if it gets futher away from 1:1.

But it really does not matter what you use, it is about the systems consistency, your personality and what you feel comfortable with.

The difference between oldies and newbies is the amount of changes they make if their system does not perform that month.

Scull pointed out that he had a losing month before and Indy pointed out that the ATR is at its lowest since Aug 2008.

We just have to stick it out and hope for a better Feb.

  • #2,462
  • Jan 30, 2010 3:36am Jan 30, 2010 3:36am
Quoting Indyfx

The problem is that you do not know whether the current range is here to stay, so for the momment I am happy with the current parameters which showed a very small profit in January. I want to use a mechanical system and the problem with tweeking it is that you can tweek it every day and then you do not have a system you are trading what you think the market will do. I hope that this is helpful which was my intention
The difference between oldies and newbies.

  • #2,463
  • Jan 30, 2010 3:47am Jan 30, 2010 3:47am
It is also very interesting to keep an eye on the poll results.

The moment we have one bad month, the No % goes up.

  • #2,464
  • Jan 30, 2010 4:47am Jan 30, 2010 4:47am
Guys i notice a few of you are asking me for my results for Nov, Dec and Jan.

Jan finished on -90 for me using IG, however some of you have reported better results, so that's good to hear. I was annoyingly stopped out of 3 trades in Jan, all hitting my SL by a mere 0.5 pips, then going onto win, so i guess i have just been a little unlucky.

RE: Nov and Dec - unfortunately i have no data for this (or any other months) as my system crashed and all spreadsheets and research have been lost as i was stupid enough not to back up my data. However i do know that if you would have traded all the way through Dec (every single day), you would have been up about +140ish pips or so. I stopped trading the last Fri of the second week of Dec, which at that point i was down about -60, but the second week was ok strangely.

Anyway lets forget Jan and see what Feb brings.
Trade the price action, not the cash!!!
  • #2,465
  • Jan 30, 2010 6:07am Jan 30, 2010 6:07am
Hi guys,

just for add results if somebody interesting by the mondays method, I confirm an no good month of january:

- 4 th jan : loss 3*40 pips (box size)
- 11 th jan: win TP1 25 pips, tp2 and tp3 at BE
- 18 th jan: win TP1 25 pips, tp2 and tp3 at BE
- 25 th jan: no trade cause ema was above price; only short allowed and price was up.

I follow previous 4 years results, it's exactly same, february should be better.

Grant, thanks for share the RBI EA's, I will try to add the days input in
  • #2,466
  • Jan 30, 2010 1:53pm Jan 30, 2010 1:53pm
Quoting grant18265
Hey Ifee,

welcome along.

the time used is 4-6 am UK time year round. at the moment that is the same as GMT.

the 5-7 quoted was in answer to another members question whose broker is gmt +1

cheers

Grant
Thanks Grant for your prompt reply.
Cheers
  • #2,467
  • Jan 30, 2010 3:49pm Jan 30, 2010 3:49pm
I think I found an EMA similar to what you are using that turns red and blue. so you are using sculls directions of +7 and -7 for buffer? also, what if the EMA changes color like partway through the box? do you just use the later color?


Quoting matlobien
My trade results for Jan using Scull's original 5 to 7:30 breakout.

Trades 20
Wins 11
Pips -30

Using the 85 EMA for direction

Trades 9
Wins 6
Pips 30


  • #2,468
  • Edited 9:33pm Jan 30, 2010 9:07pm | Edited 9:33pm
can someone verify grants sept results for me. I'm going through sept and a number of the trades where he got 75 pips, I got -15 because they would have stopped out, 1 was right at the edge so that could be a difference in feeds, but 2 of them would have surely stopped out as they were at least -20 to -25 where stoploss is 15.

am I doing something wrong? box is at 4-6. 5 pip buffer each way. if I used 7 pip buffer up top, it woudl make it worse.

EDIT: november came out identical and october came out better for me. but sept I get like -1 pip after subtracting spread and everything. so is it just me or are the sept numbers wrong? I want to make sure I'm doing it right, i think i am.

finally, i really don't understand the reasoning of making the buffer higher to adjust for the spread, that doesn't really make sense. if doing it for backtests, that won't work either because you are buying at the buffer point and your get in point is 2 points higher, you can't simulate that by adding 2 points for spread because that will throw off a number of trades.

The way I do my manual backtesting is I do it without spread and just buffer, then when I am done, i count the trades for the month and subtract the pips. so 15 trades at 2 pips a piece is 30 pips removed. seems to be the best way. I suppose you could try to add that to the profit targets or something. otherwise, the stops wouldn't be 15 pips, they'd really be 13 pips to account for the spread. doesn't make sense to me. maybe I'm just stupid? I'm used to using tradestation and you enter the number for slippage and commission or spread or whatever and it removes that for each trade, seems only sensible way to do it.
  • #2,469
  • Jan 31, 2010 9:17am Jan 31, 2010 9:17am
rob this is what i got for grant's sep results hope it helps


Quoting maq
grant

i may be missing something as sep was a bad month for fed's breakout same time as your i backtested sep

i got 3* wins 4* lose 3*be

am using 85 ema although it shows as 85ma i have put in exponential in the box
here are my findings
1/9 n t
2 n t
3 75
4 -15
5 nt
7 nt
8 nt
9 -15
10 nt
11 nt
14 -15
15 -15
16 nt
17 be
18 be hit 60
21 be
22 75 ema skimmed box
23 24 25 28 29 nt
30 75
"The person who knows that enough is enough, will always have enough". .
  • #2,470
  • Jan 31, 2010 11:12am Jan 31, 2010 11:12am
Quoting robl45
can someone verify grants sept results for me. I'm going through sept and a number of the trades where he got 75 pips, I got -15 because they would have stopped out, 1 was right at the edge so that could be a difference in feeds, but 2 of them would have surely stopped out as they were at least -20 to -25 where stoploss is 15.

am I doing something wrong? box is at 4-6. 5 pip buffer each way. if I used 7 pip buffer up top, it woudl make it worse.

EDIT: november came out identical and october came out better for me. but sept I get like -1 pip...
HI Rob,

apologies for the lateness of my reply but i have been away.

i have no idea why the results are that diferent, feeds maybe, i just dont know..

i use 7 and 5 for the buffer because whilst using the EA to optimize the settings i had it test buffers from 1 to 10 (ignoring spread) on both sides and 7 long and 5 short came out the best overall, so i have gone with those...

the same applies for the SL of 15, i tested 10 - 50 and 15 came out tops, probably because the Risk reward is so good...

the same also applies for the TP 75, i tested 50 - 200, 75 was not the best but not the worst and i picked it cos it was about half of the ADR for cable at that time...

of course these are the settings i use and they suit my risk tolerance etc, i'm sure there are many members that have made tweaks to suit their style more

if your backtests come up with something better then please post the figure and the reasoning so that all the members can see and we all can make a choice of whether or not to change...

heres to a pip filled Feb

cheers

Grant
  • #2,471
  • Jan 31, 2010 11:25am Jan 31, 2010 11:25am
grant,

thanks for getting back to me.

i'm busy manually backtesting these systems. you say 7 pips buffer for long, I was under the impression that you were using that to cover the spread? so should I just use 7 pips long and 5 pips short for the buffer while doing the testing. as I mentioned in a previous post. I figure all the trades for the month and then go back and subtract the pips for spread, cleanest and most accurate way to do it.

I'm using the 85 ema for these tests. one question. as far as the yellow box, maybe I missed it somewhere, but what is that comprised of? it isn't just the high or the low of the candles inside of it. and if the line is going through the box, is it no trade if in the yellow area, or only if its passing through the candles? I believe there was a question about that and it was okay if its just barely moving through the box, like one corner or something? I don't mean to be anal, just want to make sure my testing is as mechanical as possible so I don't end up changing the results to my liking by accident.

EDIT: one last thing, did you ever test without breakeven? I find a number of times, I miss 75 pips because the breakeven kicked in, but you've tested it over the long haul so you'd know if thats just an anomoly.
  • #2,472
  • Jan 31, 2010 12:27pm Jan 31, 2010 12:27pm
Quoting robl45
grant,

thanks for getting back to me.

i'm busy manually backtesting these systems. you say 7 pips buffer for long, I was under the impression that you were using that to cover the spread? so should I just use 7 pips long and 5 pips short for the buffer while doing the testing. as I mentioned in a previous post. I figure all the trades for the month and then go back and subtract the pips for spread, cleanest and most accurate way to do it.

I'm using the 85 ema for these tests. one question. as far as the yellow box, maybe I missed it somewhere,...
Hi Rob,

My analysis shows that the setting that Grant has are pretty well the optimum. I have made a few changes of my own. See my posts for the backtesting details # 2066 and 2443
  • #2,473
  • Jan 31, 2010 12:45pm Jan 31, 2010 12:45pm
Quoting robl45
grant,

thanks for getting back to me.

i'm busy manually backtesting these systems. you say 7 pips buffer for long, I was under the impression that you were using that to cover the spread? so should I just use 7 pips long and 5 pips short for the buffer while doing the testing. as I mentioned in a previous post. I figure all the trades for the month and then go back and subtract the pips for spread, cleanest and most accurate way to do it.

I'm using the 85 ema for these tests. one question. as far as the yellow box, maybe I missed it somewhere,...
Rob,

the 7 and 5 include the spread so just add/subtract them from the highest/lowest candles in the time period to get the entries...

where the yellow box finishes is this point so, highest candle + buffer and lowest candle - buffer. these are my entries and the text confirms the levels...

if the ema touches the box anywhere then its no trade that includes just clipping the corners

cheers

Grant
  • #2,474
  • Jan 31, 2010 1:00pm Jan 31, 2010 1:00pm
rob,

with reagrd to a no BE move please find attached the comprehensive backtest i did for the TP's using 7 and 5 as buffers and 15 SL.

you will see that using no move to BE is indeed the most profitable way forward.

As i trade clients money with this system i felt that capital preservation and drawdown were key issues for them. Using the move at +25 it preserves their capital and they can still make decent returns which have been greater than a savings account from any bank in the UK that i know of.

please be aware that the backtest was done before the ema in the box rule was introduced so the results will differ from the SS i have already posted... but hopefully you will see where i am comiing from in regard to the settings i use...

cheers

Grant

ps as you will see a lot of work went into this which is why i was more than a little p*ssed off with Mr Mead...
Attached File(s)
File Type: xls GU buffer backtest - long 7, short 5, SL 15, 1% risk.xls   32 KB | 388 downloads
  • #2,475
  • Jan 31, 2010 2:51pm Jan 31, 2010 2:51pm
okay, seriously, I'm not stupid, let me make sure I get this right, the actual buffer is +5 and -3 then? That would be the actual entry on the chart? or +5 and -5 is the actual entry on the chart? like if high was 1.6455 in the box then we enter when chart says 1.6460 but of course we are actually getting in at 1.6462 because of spread. So now where is the stop? is it at 1.645 or 1.647? and then when going short. how do you account for the spread?

like if low is 1.6330 and we enter when chart says 1.6325. do we put the stop at 1.6340 or 1.6338 to account for the spread?

Quoting grant18265
Rob,

the 7 and 5 include the spread so just add/subtract them from the highest/lowest candles in the time period to get the entries...

where the yellow box finishes is this point so, highest candle + buffer and lowest candle - buffer. these are my entries and the text confirms the levels...

if the ema touches the box anywhere then its no trade that includes just clipping the corners

cheers

Grant
  • #2,476
  • Jan 31, 2010 4:46pm Jan 31, 2010 4:46pm
Quoting robl45
okay, seriously, I'm not stupid, let me make sure I get this right, the actual buffer is +5 and -3 then? That would be the actual entry on the chart? or +5 and -5 is the actual entry on the chart? like if high was 1.6455 in the box then we enter when chart says 1.6460 but of course we are actually getting in at 1.6462 because of spread. So now where is the stop? is it at 1.645 or 1.647? and then when going short. how do you account for the spread?

like if low is 1.6330 and we enter when chart says 1.6325. do we put the stop at 1.6340 or 1.6338...
Rob,

examples:-

candles highest point in box 1.6500
+ buffer (7) 1.6507 (entry)
SL 1.6492
TP 1.6582
BE @ 1.6532

candle lowest point 1.6500
- buffer (5) 1.6495 (entry)
SL 1.6510
TP 1.6420
BE @ 1.6470

if you have the inputs for buffer (amount above/below high/low) set at 7 and 5 in the indicator then the yellow box draws at the candle high/low plus or minus the buffers, and it will put the entry price in text a couple of bars after the box draws...

i cant/dont know how to break it down any more for you...

Grant

ps 5 digit brokers need the inputs as 70 and 50
  • #2,477
  • Jan 31, 2010 5:18pm Jan 31, 2010 5:18pm
okay, so the buffer is the point on the chart where you enter, and where the stop loss is set at, but the actual entry will be different for longs such that in the example below, if you went long at 1.6507 and your broker has a 2 pip spread, you would actually get in at 1.6509. is this correct?

Quoting grant18265
Rob,

examples:-

candles highest point in box 1.6500
+ buffer (7) 1.6507 (entry)
SL 1.6492
TP 1.6582
BE @ 1.6532

candle lowest point 1.6500
- buffer (5) 1.6495 (entry)
SL 1.6510
TP 1.6420
BE @ 1.6470

if you have the inputs for buffer (amount above/below high/low) set at 7 and 5 in the indicator then the yellow box draws at the candle high/low plus or minus the buffers, and it will put the entry price in text a couple of bars after the box draws...

i cant/dont know how to break it down any more for you...

Grant

ps 5 digit brokers need...
  • #2,478
  • Jan 31, 2010 6:57pm Jan 31, 2010 6:57pm
No rob
your Entry is your entry level. you put a limit or stop order for the entry at the low/high point + the buffer, thats it dead simple, the spread on the open is kind of irrelvant as u want to enter at the entry point as dictated by the box + buffer
  • #2,479
  • Jan 31, 2010 7:06pm Jan 31, 2010 7:06pm
Hi Seba,

I use http://www.virtualbox.org/ to run windows and MT4 in my wife's Macbook. It runs an intel processor and OS 10.4. The software is free and it's not hard to install and configure, but you do need a copy of win xp or equivalent. I used a free version made for Eee pcs called EeeXP http://modmyeee.com/eeexp.html

Good luck.


Quoting sebasound
HI IndyFx

Thanks for the recap. It sounds very good, a bit more SL room (which is good considering the buffer) and still very healthy RP Ratio.

Unfortunately I spent all day trying to sort out my issue with Mac being incompatible with MT4 and haven't found a solution (free PC simulator) yet. So for now I'll only be able to test the previous three weeks, and won't be able to run the Slope Indicator, only the 85EMA option... rubbish. Nevertheless I'll get going with January from the 11th on in the next days and keep updating that as we go. Hopefully...
  • #2,480
  • Jan 31, 2010 8:04pm Jan 31, 2010 8:04pm
this is what I would think, but then your stops and such need to be adjusted for your actual entry price I assume which would be 2 pips higher or whatever your spread is. I keep hearing people talk about long buffer being higher because of spread. it would seem to me that both buffers would be the same unless the system tests better with a different long and short buffer. but adding spread to the buffer makes no sense.

Quoting kiwiduvet
No rob
your Entry is your entry level. you put a limit or stop order for the entry at the low/high point + the buffer, thats it dead simple, the spread on the open is kind of irrelvant as u want to enter at the entry point as dictated by the box + buffer
5 traders viewing now