Hi Derek (midulster), and welcome to the thread, and thanks for going to the trouble of posting and attaching your backtesting results.
I - for one - don’t have the ‘guts’ to look at your backtest results - if I’m wrong - I don’t want to know about!!! Only kidding.
Having said that though - I’m REALLY starting to think that the way that Parabolic SAR is plotted on MT4 leaves much to be desired - AND - I’M REALLY HOPING that this is the reason why your backtesting gives less than desired results!!!
I don’t know how or why the different calculations of the very same indicator with the very same values can differ so much from platform to platform - it ‘befuddles’ me!!!
Of the (my) three platforms it appears that Delta’s is the MOST correct.
One thing though (and again I stress this):
I am following ‘pure’ Parabolic SAR i.e. not taking into account ANYTHING else AND the reason that this works for me (or should I say has worked for me so far) again is PURELY because of the amount of free margin that I have ensured that I have, the sheer number of pairs that are available to me, and the fact that I (now) trade without stops i.e. I ONLY stop and reverse when I get a new Parabolic SAR entry signal in the opposite direction the next day.
The above is VERY important (I believe) if you are ONLY using Parabolic SAR i.e. ‘pure’ Parabolic SAR.
From what I have seen of late (and because the markets are so quiet at the moment I have had LOADS of time to have a look at things as they stand) if you are only going to stick to one or two or three pairs (like trading ONLY the majors) then I would be the first person to say that you CANNOT use ‘pure’ Parabolic SAR and that you HAVE to take into account information given to you by other ‘confirmation’ indicators such as MACD, or Akrams ‘Alligator idea’, or trend lines, or any of the other fantastic ideas posted here by other ‘Parabolic SAR groupies’.
I mean - as it stands with me right now - about 80% of my positions are sitting at a loss right now (like I said mainly because of the ‘commdolls’) but even although most of them are at a loss right now - that loss would be FAR greater if it was not being offset by the other positions that are showing a profit AND had I not closed all my positions at month end last month I would very definitely be sitting on a nice nett profit about now. This is the MAJOR problem that is caused by ‘bailing out’ when Parabolic SAR has not told you to close your positions (at least in my case) because now it’s almost like I opening positions as indicated by ‘pure’ Parabolic SAR but I’m catching the ‘dying’ or the ‘dead’ or the ‘tail end’ of the trends i.e. the ones that would have cost me money ANYWAY had I not closed out my positions early last month BUT had I had ALL the postions open and not closed ANY of them out at month end last month - like I said - my nett profit would be (probably) JUST as ‘humungous’ as it was about two or three weeks ago (and then of course - being the ‘disciplined’ trader that I am - I would probably have closed them all early AGAIN yesterday or today - and I’d be typing the EXACT SAME MESSAGE as this in about a weeks time and ‘crying on your shoulders’ again)!!!
Now - if you think about this (the above) nicely - I could QUITE EASILY have chosen to stick with GBP/CAD for example - and - had I done this - my ‘luck’ would have been out i.e. this pair has been CONSTANTLY costing me money over the past couple of days / weeks because of the way it’s moving. I would go so far as to say that if you are ONLY going to stick with ONE pair and use ‘pure’ Parabolic SAR then I would recommend that you go with EUR/USD (and - to be quite honest - that statement is made based on what I see happening with James’ 30 minute EUR/USD ‘system’ detailed in the same forum as this thread with one ‘proviso’ and that is that I STILL believe that trading on any timeframe shorter than a day is looking for trouble but that’s just my ‘humble’ opinion and is based solely on the fact that I have NEVER made any kind of ‘decent’ profit on the shorter timeframes EVER and have, in the long run, always lost more than I’ve made on the shorter timeframes).
I would also ‘stick my neck out’ here and say that in backtesting for results of using ‘pure’ Parabolic SAR you must take the nett results over a period of a year i.e. you cannot just pick a random number of days / weeks and then say that it does not work (and I’m not saying that this is what you have done) for the simple reason that we already know that there will be some HUGE drawdowns from time to time but it’s the nett results over a long period of time that matter.
Again - as far as ‘backtesting’ is concerned (sorry - I know that the ‘order’ of this message is a bit ‘jumbled up’) - I would not run any of those backtesting ‘things’ available to you on MT4 - open a chart - and manually go and calculate the profit or loss made using ‘pure’ Parabolic SAR over a period. The reason I say this is that if I look at any chart - any pair or any commodity - and I do the calculation manually - it (almost) without exception indicates that a profit would have been made and there should be no reason why this should differ whether you are ‘backtesting’ manually or trading based on the indicator itself in the future.
Regards,
Dale.
PS - Akram - I sent you a (private) email - please let me know if you got it because the page said ‘done’ after I sent it but nothing more after that.
