Parabolic SAR - that's all!

Trading Systems Free Forex Trading Systems
1833

OK - well - still getting ‘analysis paralysis’ (and it’s costing me a fortune on cell phone bills but anyway)!!!

As you know I’m STILL on my Parabolic SAR ‘quest’ and I think I’ve come up with something interesting yet simple.

It only addresses ONE (maybe TWO) of the THREE MAJOR PROBLEMS with Parabolic SAR:

1 - Late entry which in turn results in:
2 - Huge drawdown if the trade goes against you (initially)

The ‘fix’ for the above (I believe) COULD be the following:

You enter THE MOMENT Parabolic SAR is penetrated i.e. you are NOT waiting for a new dot to appear in the opposite direction for entry / stop and reverse. You INITIAL STOP LOSS becomes the high or low of this candle where you have just entered (depending of whether you are long or short). Because Parabolic SAR has been penetrated it follows that a new Parabolic SAR dot will appear upon the opening of the next candle. You are now ALREADY in a position. Wait for THIS candle to close and your stop loss now becomes the high or the low of THIS candle (again depending on whether you are long or short). Keep this stop loss value (it is a stop loss NOT a stop and reverse point) UNTIL Parabolic SAR overtakes this value and then start tracking / locking in you profits.

Make sense???

This has been one of my MAJOR ‘bugbears’ with Parabolic SAR i.e. sometimes the initial Parabolic SAR dot (and therefore your initial stop loss) is MILES away from your opening price and if the trade goes against you at this point it can cost you a fortune!!! If you have a look back somewhere on this thread you’ll find one of my MAJOR ‘fuc*up’ trades (it was GBP/CAD if I’m not mistaken) that cost me over $8K because of this problem and that was ONE trade (and - as a matter of fact - that was the beginning of a HUGE losing streak if I’m not mistaken i.e. things just got progressively worse from there because I then started looking at all sorts of other things none of which worked out).

This is not rocket science (and I don’t know why it’s taken me SO long to see this either). From what I can see it’s those Parabolic SAR trades that start out with this HUGE distance between the initial Parabolic SAR dot and the opening price of the candle giving you this Parabolic SAR dot that go against you and can quite easily wipe you out (if not financially then phsychologically and THEN financially)!!!

OK - obviously there is the chance that you are going to get whipsawed out of a good trade BUT at least the loss SHOULD be negligable and this assures that you’re still around to take the good trades when they come i.e. the trades with trends that have ‘conviction’. On the other hand - if you’re brave - you can reopen a position if and when you have a close above or below the stop loss mentioned above and see what happens. Maybe make a rule that if you get stopped out twice or three time on a single trade then walk away and wait for the next one.

There is no point in taking huge loss after huge loss and making a huge profit on another trade i.e. the two will over time cancel each other out and you’re wasting your time.

I know this goes against Parabolic SAR’s ‘design’ i.e. it’s supposed to provide you with enough room at the beginning of the trade to stop you from getting whipsawsed out of the trade BUT it would appear (especially on forex pairs) that this ‘idea’ is no longer ‘valid’ i.e. the markets move faster and ‘spike’ far more often nowadays than they did in the '70’s and I think this is where the difference comes in to play. Put it this way: if you want to see what Parabolic SAR was ‘designed for’ then just have a look at Soybeans. ‘Pure Parabolic SAR’ seems to ‘fit’ Soybeans ‘like a glove’ with no additional ‘messing around’ but take a look at ‘Pure Parabolic SAR’ on any of the indices (especially at the moment) or on forex pairs and you’ll IMMEDIATELY see what I’m talking about.

Comments???

By the way - as I’m typing this - I’m thinking!!! What about ‘modifying’ Parabolic SAR’s acceleration factor by a factor of either the current ATR or current volatility instead of a ‘fixed’ acceleration factor??? Not sure how to do it or if it’s even possible BUT just food for thought!!!

eurusd this is what your looking for.jpg700×360 46.5 KB

eurusd this is what you want to avoid.jpg700×360 49.1 KB

1834

OK - well - just a slight ‘variation’ on the above.

It would appear ON THE SHORTER timeframes e.g. 30 minutes that you are probably better off waiting for the candle to close before entering a new position BECAUSE, for example on EUR/USD 30 minutes (at the moment anyway), the penetration of Parabolic SAR is more often than not due to a ‘spike’ in the price than a rally through the the penetration point.

Suggestion (in this case):

Wait for the candle to close before opening a new position BUT make your stop loss a pip or 2 + the spread above or below the high or low of the previous candle and then move your stop loss to a pip or 2 + the spread above or below the high or low of the current candle as soon as the current candle has closed.

This only appears to be a problem on 30 minute timeframes and under i.e. on the longer timeframes the penetration of Parabolic SAR is normally due to a sustained rally through it than ‘spikes’.

Again: thoughts on this??? (Trying it out now on the 30 minute charts and looking good so far i.e. been taken out (stopped out) twice BUT took profit on a third trade which was far more than what had been lost in the two previous trades BUT this would not have been so had I followed Parabolic SAR ‘to the letter’ i.e. the two previous losses would by far have negated the profit on the third trade.

1835

Dale,

Good to see you back onboard again trying to tune a winning system. You were sorley missed around here.:slight_smile:

A few weeks ago, one poster suggested using ATR as a way to set a stop loss if the PSAR dot was too far away from the current price. I have used that myself, using 2x ATR (14) as a maximum stop loss position away from the price. Worth a thought.

Ok all, I am going on holiday tomorrow until the 2nd of January so no more trading for me ( or very limited) until I get back so happy holiday season to everybody on here.

Good luck and happy trading.
See you in the new year.

Best Regards
Boca

1836

why not try opeing up a smaller timeframe and when the lower tf aligns with the daily take the trade? just a thought

1837

Hey Boca,

Happy holidays and thanks for the good words. Thanks to my trading I ain’t going on holiday but - next year - well - let’s see!!!

cryten:

You’ve got the idea. That’s a good way to ENSURE that you’re going with the major trend i.e. looking at the daily but then trading the shorter timeframes in the same direction only. I thinkg just about everyone on this thread is doing something like that already though. The problem is still the drawdowns i.e. if you don’t have loads of capital (which I don’t anymore and to be honest EVEN WHEN I DID) this indicator can wipe you out fast if you’re wrong when your stops are too big at the beginning of a trade. What I trying to accomplish is to be able to be wrong 10 times only only lose $10 (for example) but be right time 11 and 12 and end up with a nett of $20 INSTEAD OF being wrong 10 times and losing $1K and then being right time 11 and 12 but ending up with a nett loss of $800. Hope that makes sense. Also - I’ve been looking again - and from I can see the trades that start out with a HUGE gap between the candles open and the first Parabolic SAR dot ON ANY TIMEFRAME are usually (appear to be) the trades that end up going ‘sour’ anyway. Those trades that start with a relatively small gap between the candles open and the first Parabolic SAR dot are usually (appear to be) the trades that turn a nice profit. That’s why I mentioned Soybeans as an example. If you have a look at Parabolic SAR on ANY Soybeans timeframe you’ll notice that at the start of the ‘action’ the Parabolic SAR dot is always neatly ‘tucked’ in close to the candle at entry and exit because there is not this pesky volatility with Soybeans and that’s one place that Parabolic SAR seems to have a weakness i.e. high volatility i.e. it’s SUPPOSED to be able to handle volatility by it’s design but I doubt that the volatility we experience in the markets today can be compared to the volatility in the markets in 1978 (when Mr Wilder developed this indicator). Also - as I have mentioned many times before - Mr Wilder was a commodities trader (I don’t even know if you COULD trade forex pairs then - don’t think so).

1838

Hey My China,

Welcome back and happy new year and marry xmas for all of u guys.So u r back Dale that is a good news .

Well it has been a long time since i opend any postion but to be honest with you guys i got an email from a nice friend that i don’t know him before but i think he was following the thread and he really encouraged me to trade so i started to open some winning postions for this week ending up with 50% profit so far to my original account.

Again this doesn’t mean that the system we (i) are following is like 100% winning one but it has it is oun moments and i came on the right time hitting the right order that is why i do have some bucks in my account.

I agree with u Dale that we shoud conceder what u came up with i will do my test on what u sent and will share it here with you (guys).

What I really like of what u r saying is that when the market is not trending i will not loss alot while when it is trending u get good profit.coz normaly what i am noticing that when the market is trending it is trending it’s maximum in the first 3 days or the first 3 candles. and what i am doing now is that i had 4 postions opened and i closed them based on what i said after the trend gets it’s maximum strenth for the first 3 days i closed my orders that means u will have tripple the size of your Lot and some times it is more than that.Depends how lucky u r!!! I beleive this is better rather than waiting for 20 days with hassel for watching the price eating your margin hoping to get away from your margin!!

Any thoughts ??

Regards,

Akram

1840

I hope that I’m not the last to stay with the PSAR. I believe that the PSAR has several parameters that we can profitably adjust.

First, in he April, 1995 (not a typo…that’s at least how long I’ve been into and out of this stuff) Dennis Meyers talks about adding two parameters to the PSAR:

  1. A STARTING acceleration. Currently all formulas use either 0.02 or the acceleration factor as the starting.

  2. Crossover increment. This is how far the price must be penetrated by the SAR value. Currently the value is zero. As soon as the SAR value is hit, a stop & reverse occurs. Increasing the Crossover increment, according to Meyers, reduces whipsaws.

I’ve already modified an existing Custom Indicator (“Parabolic” which is the same as PSAR–“Parabolic” came as a custom indicator with MT4…I did not write its basics…I only modified it) to provide 1., above. The custom indicator is attached to this post. I am starting to look into 2., above.

Reducing the Size of the First Stop Point

Now another point. The PSAR came from the early or pre-computer days. I looked over some PSAR behavior (I should have looked at the program) and it seems that the initial stop point is the previous high (or low, depending upon what the PSAR wants you to do).

We can do a lot better than that. For example, to calculate the initial stop SAR value (first dot on the other side of price):

  1. Look at the last 10 periods :
    a. If the most extreme (low for buys or high for sells) value is within the last 3 periods, then use that value as the first SAR stop and reverse value.

b. If the most extreme value is not within the last 3 periods, then perform a 10 period “exponential” average (not moving, because it is of the last 10 periods only) to get the first SAR value (initial stop and reverse).

Looking at some of the first SAR values chosen by the MT4 program for the daily EURUSD, no wonder people fear the large stops. Look them over and see how many first SAR values of “my” calculation would be hit. It seems like few to me.

Is this worthy of more research?

Custom Indicator Parabolic3Parms

I’ll attach the MT4 custom indicator to this post. when you select the indicator (called Parabolic3Parms) you can set the Initial as well as the step and maximum parameters.

If you need to know how to load the indicator into MT4 just mention here and I will post some instructions.

I Keep Thinking…it helps keep me away from trading

My thoughts are that not only should the PSAR be optimized for a currency pair, but also optimized for with- and counter-trend buys or sells.

Barry

Parabolic3Parms.zip (1.46 KB)

1841

Hey Barry,

I hope that I’m not the last to stay with the PSAR.

I would not worry too much about that!!! If you’ve read (most) of this thread then you already know my story. Funny thing is - I ALWAYS end up coming back to this indicator!!!

I like your thoughts and input. Unfortuanately I no longer have an account with an MT4 broker (strangely enough NOT due to my ‘less than stellar’ trading activities BUT because it would appear they’ve gone out of business or at very least are still under investigation by the Swiss authorities ALTHOUGH I never really put any serious money into that account anyway thank goodness). I will look at the (your) code though because I am able to modify Parabolic SAR with at least one of my current brokers.

1843

Hi:

Why not get a demo account at a broker who uses MT4 just to be able to try this stuff?

Barry

1844

Hi Barry,

Yes - I suppose - given enough time - it would have ocurred to me to open a demo account with an MT4 broker (I’m a bit slow sometimes)!!! On the other hand - I have no intention of changing brokers so I have to make things work with these two brokers (I can still log in to my MT4 account it’s just that it appears that trading has been suspended although I do have access to historical data for backtesting). With my current two brokers I can add/edit custom indicators (with great difficulty though I might add).

Not wanting to sound like a total ‘pratt’ I did not want to ask you who Dennis Meyers was so I Googled him and ended up at his website. ‘WOW’ is all I can say!!! Have you signed up with them or bought any of their indicators at all?

What I found VERY interesting were the articles on the noise channel breakout systems (I’ve been doing a heck of a lot of ‘experimenting’ with price channels over the past couple of weeks and this has really captured my interest). If you have a good look at the articles and the charts associated with them you’ll actually find (or at least this is how it appears to me) that his idea of filtering noise on price channels really works the same as Parabolic SAR with some sort of noise filter introduced.

Thanks Barry - a lot of food for though there.

1845

Wow, Dale!

I never thought to look for Dennis Meyers on the net. I just took him to be one of the many random authors that submitted to Tech Analysis of Stocks & Commodities. I will look into his website.

Have a great Holiday Season (everyone)!

Barry

1846

Hello again Barry,

I hope you’ve had a good look at his site - there is some REAL interesting stuff there!!!

I’ll tell you something funny though:

Like I said a bit earlier - I have been doing a lot of work (research?) with ‘price channels’ for the past couple of weeks or so. It started out with me ‘checking out’ an ‘envelopes’ indicator that I have on one of my trading platforms. I ‘played around’ with this indicator and then started off on my own ‘tangent’ creating various ‘versions’ of this indicator only to find out a week or two later (from searching on the Internet for something) that I had - on my own - created ‘Keltner Channels’, ‘Donchian Channels’, and ‘STARC Bands’. I sht you not!!! And when I saw Dennis Meyers talking about ‘noise filters’ it really got to me i.e. I’ve had ‘thoughts’ for the past couple of days or so on trying to come up with some way to ‘filter’ market noise to avoid whipsaws and - what do I see yesterday on his site - ‘noise filtering’!!! Why is this 'pising’ me off you may ask??? Well - these guys all have Ph.D’s and other degrees and diplomas - not to mention experience - and - I can only assume - have become very wealthy over the years trading. This business has all but taken away a very good life that I once had (financially and more) BUT it does not seem to be due to a lack of ‘brainpower’!!! I wish I had the answer!!!

Anyway - thanks to you (and Mr Meyers) - I’ve worked a lot today on creating a ‘noise filter’ which has a long bias and a short bias - and it’s looking good. I’ll be testing it from tomorrow night to see whether or not I’m ‘on the mark’ and if I am - I’ll post details of MY ‘noise filter’ here so that we can try and use it in conjunction with Parabolic SAR!!!

1847

Hi all,
I was wondering what this system has evolved into since the beginning.

I mean it started with only the Psar with daily, weekly, monthly charts, what has it grown into?

Also, has anyone come up with a Psar EA yet?

Thanks for the input,
Vanhook2

1848

Hi Dale:

I did look at the site, especially his 5 parameter PSAR. It seems that Meyers wants to move the initial stop even further away from the price action.

I am starting to look into using some other indicators like a PSAR. That is, to form the PSAR indicators on the chart using moving averages.

Still looking…

Barry

1849

Hi Barry,

It seems that Meyers wants to move the initial stop even further away from the price action.

Yeh - that seems to be about right - almost a case of ‘damned if you do - damned if you don’t’!!! In other words - either you use EXTREMELY close stops or EXTREMELY wide stops and the problem with EXTREMELY wide stops is at what point do you now decide to stop and reverse i.e. will Parabolic SAR EVER get penetrated???

Off the subject BUT I just want to make people aware of the fact that you MUST check the output / results of your indicators MANUALLY i.e. DON’T take the trading platforms ‘word for it’!!! I’ve (once again) just discovered a ‘standard’ indicator giving TOTALLY wrong readings / values (it was Wilder’s ATR at Delta i.e. the results of calculating the TR and ATR values manually as per his books / worksheets are TOTALLY different to what the system is giving you i.e. had to alter the code to get the ‘standard’ indicator to agree with the CORRECT results of the manual calculation. I’m NOT saying that it would have / could have resulted in a loss of money BUT that’s just THIS indicator)!!! Beware!!! Not ALL indicators as presented on different trading platforms are equal my friends!!!

1851

Hi Codybear! Do you stick with one currency pair while trading or multiple?

1852

I agree, thats the biggest problem i find, is that people just risk to much, money managment is the #1 reason people fail, if you were to practice the same rules for every trade and were only able to predict 51% of profitable trades you would still make money, but people just can;t seem to do that.

1854

thats nice. did you got that with this psar system?

my bad luck did continue last week I was as low as $48 but now I have $150, so I got good lesson about money management.

only hard thing to me with this money management is when I’ve $150 to trade with and I take 1% per trade, the profits are not that big

1855

Good day all,

Happy new year to everyone here on Babypips. I just came back from an extended xmas and new years holiday in Thailand. Very refreshing.

Ok, I have been live since the end of November last year so that’s more than a month of live trading now and currently I am seeing an increase in 3% in that time.

I am no longer using my demo to trade reguarly so that will be the last of my placing the list of picks here. I will probably post a shot of my monthly log if people are interested to see the trades that are being taken.

I am basically using the PSAr system described eralier on the daily timeframe only, but I am not just blindly following indicators as I like to take a look at the chart pattern to help judge a good trade from a bad.

Money management is so important, I think I cant state that enough. I am risking about 0.75% to 1% per trade.

Ok, good luck to everyone on here for healthy trading in 2008.

Best Regards

1856

Hi folks,

The forum software is telling me that I have not posted in a while (I did not think it was that long ago)!!!

Anyway - Merry Christmas (to those who celebrate Christmas) and a Happy New (Trading) Year!!!

For what it’s worth - I have FINALLY managed to sort out my ADSL - so - I suppose - you could say - I’M BACK!!!

These past few weeks of NOT being able to trade in realtime have probably been the best thing that has ever happened to me believe it or not. If nothing else it’s taught me EVEN MORE patience and to NOT MESS with trades i.e. take the position, set stops, set targets, and leave the position alone. Accept your profits and take your losses and leave it at that!!!

I have been doing a lot of ‘manual’ calculation and analysis (sort of ‘paper trading’) and I have come up with quite a few ‘gems’ that I will share with you all in good time (as soon as I have had time to put them down in a readable form).

I have also spent a lot of time ‘mastering’ the creation of custom indicators at Delta and GCI and I will be ‘coming up’ with a PROPER Parabolic SAR indicator soon (hopefully - I’m still ‘mastering’). Why do I say ‘a PROPER Parabolic SAR indicator’? Well, for one thing, I have noticed that, contrary to poplular belief, Parabolic SAR DOES NOT just start at a certain point and then keep on accelerating ‘blindly’ by the step factor which is exactly what happens on both my trading platforms as well as on MT4 i.e. ‘PROPER’ Parabolic SAR will ‘back off’ or accelerate depending on price and volatility which avoids many of those ‘pesky’ ‘whipsaws’. The only time that Parabolic SAR will create a ‘perfect parabola’ is if each bar in the trend has exactly the same (true) range and closes exactly the same number of points higher or lower than the previous bar (depending on whether you are in an uptrend or a downtrend). This does not happen with any of the three aforementioned trading platforms and I’d actually like to see one where Parabolic SAR is correctly plotted!!! How’s that for a statement!!!

Anyway - that concludes my post for the software!!!

Hope things are going well - and I can assure you - 2008 IS GOING TO BE ‘THE YEAR’!!!