Trading Systems in 'New Concepts In Technical Trading Systems' by J. Welles Wilder

Trading Systems Free Forex Trading Systems
1067

Post #1055:

The fact of the matter is (to quote): ‘It is not unusual for the HSP and LSP to precede the HIP and LOP (made by price) by one day’. That’s easy. However: in the second example on the chart (to the right) it would appear that the LOP preceded the LSP by one day. This is not detailed anywhere. So the question is: CAN the LOP preceed the LSP by one day OR are you supposed to use the LSP made on the same day as the LOP even although it is quite clear that there is no LOP formed by the last bar???

By definition on p. 99: “A HIP is a daily high price with a lower daily high price the day before it and the day after it.” And the same for LOP correspondingly.

So I would say that LOP can preceed the LSP by one day as in your example. This is something that I have not noted so far and need to incorporate in my program.

Post #1056:

If you got into a long trade after the close of the last bar in the chart (it’s a close above a SIGNIFICANT HIGH POINT) then you should be using the previous LSP as your INDEX SAR. The problem in this chart is the fact that point B is in fact the last LSP and it has a lower ASI value than the LSP at point A BUT there is NO corresponding LOP anywhere around for point B. There IS in fact a corresponding LOP for the LSP that was formed at point A EVEN ALTHOUGH THIS is NOT the last LSP!!!

If taking the rules as a purist, I would say that because there is no LOP at B, you would then use the LOP at A.
But I would like to take some distance to viewing at this.
Maybe this anomaly is a warning sign. Maybe we should stay out of a trade like this.
Maybe the chart is not well-formed for a proper SIS trade, and there are potential losses waiting ahead.
The chart in the example together with ASI do not show any proper trend, and therefore are maybe not ideal for SIS.
There could, of course, be a breakout waiting, but into which direction? Are we betting here too much?

Post #1057:

This concerns the making of a new ASI high before selecting a INDEX SAR LSP or HSP. Although this is detailed in the book it’s very obscure (to me anyway).
…
However: the ASI has not made a new high for the trade so the INDEX SAR remains at point A until a NEW ASI HIGH is made and only then does the previous LSP become the INDEX SAR!!! ???

On p. 97: "Once in the market, we use the previous swing point as the INDEX SAR."
I understand here that you entered after the candle that made the HSP between A and B.
So I would have placed the initial INDEX SAR at A as well.
But as the trade has now reversed, i would call the HSP of entry as a significant one. This would mean that B is the first LSP after a new significant HSP (which has formed just before the trade, which is undeniably a bit subtle here).
So I would have moved the INDEX SAR now to B.

But if you had entered this trade as an initial one, you would have waited until the significant HSP between A and B had been crossed.
This sounds another case with subtlety involved --> a possible warning to stay out?

I have now been staring at these SIS issues all evenings this week, doing various backtests and formalising the trade management rules for my program. I am becoming more and more convinced that SIS alone is not enough. I have seen days in the historical data where SIS has behaved very badly, and then days when it has been unbeatable. The first days of this week fell clearly in the latter category. What I am after is some filtering criteria that allows you to decide when to jump in and when to stay out.

Some ideas, which all need some refinement (throwing these here for free commenting before having better evidence about how these would work):

  • It seems that losing trades come in streaks. So if you have one or two losing trades, jump out and enter only by the initial entry rule.

  • SIS is at its best when using it in heavily trending market. So we need to find when the market is trending. The problem with the trend-showing indicators may be that they are lagging too much. It occurred to me that
    we could look at these indicators (ADX/ADXR/VHF) at one timeframe below the one we are trading. If it starts showing ranging there but our timeframe is still trending, it might be a good time to jump out. The best pairs could be also selected by an ADX rank compiled from that lower timeframe.

  • When in trend, counter-trend trades are usually not very profitable --> conclusion: don’t enter counter-trend trades. You might consider entering every trend-following trade as an initial one. To define the direction of the prevailing trend, I’m thinking at looking at the timeframe above us.
    If there is no proper trend there, stay out.

  • If there is a huge candle, especially in the opposite direction of the trend, the market is probably reacting to something. Stay out. (There should be a good definition for what is a “huge candle”.)

Dale had also some ideas about defining the ideal ATR factor for calculating L. While I see some merit in doing it in that way, I have not had time to study that further myself. The above pondering has also made me to think that if you have been able to select the time of trading correctly, the value of the factor is not that critical. It might save something in the losing trades, but you will save much more if you are able to stay out from those trades altogether.

J.

1068

The ADX/ADXR is a sound calculation and I believe there is merit to using it. Wilder used it to filter his SIS trades, we should be using it! Yes of course, this thread has come to the conclusion that ADX/ADXR is preventing us from getting into good SIS trades, but Wilder’s SIS uses a filter!!! We need a filter.

We modified the SIS to potentially be profitable in today’s markets, I am sure we can pool our knowledge to modify the ADX/ADXR to compliment the SIS. If we get this filter sorted properly we should be SET!

K.

Your shorter timeframe theory is intriguing. I have read about other systems analyzing the shorter time frame indicators and applying to the longer time frame trade. I think this has merit, and if possible we should backtest a filter on the 30 minute timeframe. Have we agreed that the 1 hour timeframe is best to use the new SIS?

1069

Yeh, OK, so I couldn’t stay away until Monday!!! Who would have thought!!!

Anyway:

Thanks for those two posts. I’m going to go through yours J. in detail once I completed this post.

I have also sent you an email message because I need you to check something in the SIS calculation, something that I think may have something to do with all these SIS issues. Either I or J. will post the results here soon.

I’ve been looking at ADX/ADXR FUNNILY ENOUGH again today!!! Hell: I’ve even been looking at Parabolic SAR again!!! AS FILTERS FOR THE SIS!!! Can you believe that. Great minds think alike as they say!!! I’ve found something that MAY have merit albeit extremely cumbersome and very few trades but, having said that, I’m talking 99.999% profitable trades nonetheless. Simply put: it involves putting Parabolic SAR on the chart, calling up ADX(7), and the ‘new fixed’ SIS, as well as the ‘old without the LIMIT’ SIS. Messy but looks good. The idea is this: you only take short trades when ADX is above 20, ADX is above both +DI and -DI, and +DI is below -DI. You wait for your first short Parabolic SAR entry signal. Once you have that signal you look at the SIS and wait for a new LSP to be formed. You then place your entry stop order below this new LSP. What this accomplishes is to keep you away from those dreaded Parabolic SAR whipsaws and false signals. You know the ones: when Parabolic SAR has told you to SAR, price hangs around for ages doing nothing, and then comes back and stops you out at a loss. By not entering Parabolic SAR trades at market but rather using the SIS for your entries you hardly ever seem to get into those Parabolic SAR trades in the wrong direction. Once in the trade: you now look at both Parabolic SAR and our ‘old without the LIMIT’ SIS for an exit. The reason you’re doing this is because, ironically, our ‘old without the LIMIT’ SIS will stop you out far earlier in most cases than Parabolic SAR will i.e. this emliminated the OTHER age old Parabolic SAR problem of giving back a good portion of your profits before stopping and reversing at the Parabolic SAR signal. You never stop and reverse using this ‘system’ you only ever ‘just stop’ and TP!!!

Take a look and see.

You’ll note that I mentioned ADX(7). ADX(14) for this purpose IS usesless I assure you i.e. you’ll NEVER take a trade!!!

Now to this end I’ve createed an ADX(7) ‘Oscillator’ which I will email to everyone once I’m happy with how it looks (it does work)!!! All this really does is give you green lines above a zero line when all the ADX criteria have been met for a long trade and red line below a zero line when all the ADX criteria have been meet for a short trade. If NO lines are shown i.e. when there are ‘gaps’ in the Oscillator then you are trading in a range (well: as best as ADX can report such event anyway). This makes the above ‘system’ easy to set up and use i.e. you first wait for a Parabolic SAR signal. If this signal concurs with this ADX(7) ‘Oscillator’ you then start looking for a SIS entry point in the same direction and exit and TP as above.

That should satisfy all that want to use SIS filters. If you filtered the SIS any more than that you’d do one trade a lifetime IF YOU’RE LUCKY!!!

Regards,

Dale.

Edit:

At this point the SIS spells disaster on the shorter time frames because of economic data releases BUT I MAY have found a ‘workaround’ for this as well i.e. I’ve also ‘bounced’ that off of J. and just awaiting his input to see whether I’m ‘on the right track or not’. The short of it is this: instead of worrying about HSP’s and LSP’s and SAR orders and the like: I noticed that with our ‘old without the LIMIT’ SIS the TRAILING INDEX SAR seems to ‘get you in sooner and get you out sooner’ and because you’re always waiting for a CLOSE when setting orders for the TRAILING INDEX SAR there are no stops or SAR’s to be ‘taken out’. I can tell you that had I used THIS ‘method’ (if it is indeed a ‘method’) I would not have been falsely stopped and reversed or taken out by economic data releases not even once and all of those trades would have gone in my favor and I’d not have given up a single cent of my gains from earlier in the week. This needs to be test, however, before anyone goes throwing money at it because BELIEVE IT OR NOT I CAN BE WRONG!!!

1070

!!!!_1.gif643×674 11.1 KB

ADX works without doing anything fancy. Trade the signal and manage the money. This is one way I gobble pips.

1071

Hey,

Hello.

Welcome to the thread (I don’t think you’ve posted here before).

Well let me ask you this:

If you have the time would you mind showing us how YOU use ADX???

If you’ve read any portion of this thread you’d see that I’m a ‘Wilder Junkie’ and I hate to contradict ‘the old man’ but I’m just not seeing it and believe me when I say that I’ve tried HARD!!!

From where I sit: by the time ADX tells you that price is rangebound it’s already broken out into a trend and by the time it’s told you that the trend is over you’ve been ‘chopped up’ in the new range. The one ‘magical’ thing about is is then it starts turning down during a trend. There I will agree that it has merit. But as far as using it ALONE as a filter for Parabolic SAR (as an example) is concerned: no go. Not in MY book anyway.

As an example: ‘the old man’ says to only take long Parabolic SAR trades when +DI is above -DI and ADX is above 20 - 25. I can tell you from BITTER experience with Parabolic SAR: that keeps you out of NO bad Parabolic SAR trades.

ADX is the ‘base’ for all of Wilder’s systems i.e. theoretically it’s signals should be taken into account no matter which of his systems you’re trading but I cannot honestly tell you that it’s made a single iota of difference (well not of late anyway).

One other thing I’ve noticed is that ADX(14) on the WEEKLY charts has merit. No question about that. But on the daily and shorter time frames I’m not convinced.

As far as the SIS goes: I can show you places where ADXR was WAY above anything else and ADX was where is was supposed to be and those trades turned sour. I can also show you at least one or two places on any given chart and any given timeframe where the most profitable of SIS trades would have been ignored had you based your entry on ADX.

Anyway: please feel free to give us a new ‘take’ on this. As you no doubt may have gathered I’m getting more and more frustrated with all these systems as the days go by!!! One minute: ADX is ‘the thing’. The next minute: it’s the worst thing to come our way!!! One minute the SIS is making ‘gazillions’ and the next minute it’s feeding the broker and the counterparty to the trades!!! Somewhere there has to be consistency in these things. We obviously, for some or the other reason, have not found it yet!!!

Regards,

Dale.

Edit:

So sorry. I did not see your chart (the ‘icon’ was not being displayed when I first opened this page). Thanks for that.

I’m assuming that ‘Advanced ADX’ is similar to what I coded this weekend i.e. a sort of ‘ADX Oscillator’. Let me ask you this: would you mind TERRIBLY much posting the MT4 code for it??? I could then recode it for our platform.

Edit:

Don’t worry to upload the script. I found it on the Internet and downloaded and installed it. Looks good although it’s giving you nothing more than green lines when +DI is above -DI and red lines when +DI is below -DI i.e. exactly the same as Wilder’s DMS system just presented in a different way. Having said that: maybe presentation is everything because even of the chart that I tested it with it looks good.

1072

OK:

Not quite as ‘fancy looking’ but gives the same signals.

The frightening part about it is if I look at that indicator I find myself thinking: why are we bothering with all this complicated stuff??? It sure looks to me (at the moment anyway) that buying on a green bar and stopping and reversing of a red bar works well!!!

(The attached chart uses our ‘unsmoothed’ ADX(13) by the way).

Who knows!!!

What amazes me is the it is not concerned with ADX or ADXR values and where ADX is positioned in relation to +DI and -DI or anything like that. The Metatrader example posted by pipgobbler simply uses the value of ADX to ‘scale’ the oscillator i.e. the curves formed by the colorbars represent the value of ADX. Not even the ‘Extreme Point Rule’ is factored in!!! Simply put: buy on green (+DI has crossed above -DI) and sell on red (-DI has crossed below +DI). Amazing. I’m tempted to say: ‘if only it were that easy’!!! Again: who knows!!! Maybe it is!!!

Regards,

Dale.

Edit:

Hmmm. All is not what it seems!!! Take the same ADX Oscillator (or the Metatrader equivalent) and ‘slap it’ onto some more charts and different pairs!!! Again: no problem when there are nice long trends as is being shown on the EUR/USD chart. Scroll that same chart back a few weeks to between around April and July this year!!! You would have spent four months getting ‘chopped to bits’. If you were lucky: you broke even after four months of work!!! Sorry folks. It would appear that it’s NOT ‘that easy’!!! As with anything: it will work EXCELLENTY for a while!!! The ‘trick’ is to have it work WELL ALL of the time!!!

Now not to dissapoint you pipgobbler. That’s not my intention and I also don’t know how long you’ve been trading (in total or with this indicator). If it’s worked CONSISTENTLY for you over a reasonable period of time then I’m glad for you. If this is something that you’ve just ‘found’ and it’s working NOW: be careful is all. This indicator relies on the fact that the money made during trends will cover the losses incurred while price is trading in a range. Now if you knew for a certainty that this was the case then I’d say OK. The problem is that NOBODY knows how long a trend or a range is going to carry on for. Hopefully over time: the losses will be covered but that’s one hell of chance to take. And this is one of the reasons why we get complicated again i.e. if you’re going to use ADX then you cannot do so without factoring in rest of the indicator.

eurusdadxoscillator.zip (36 KB)

1073

Good (Sunday) morning all.

Well: not to be ‘outdone’ by Metatrader I’ve created a ‘supa doopa’ version of the ADX Oscillator (as well as a crass attempt to pass the time away until tonight)!!!

Take a look at the attached chart.

On the chart: ADX(7), ADX(13), and ADX(14) are being displayed.

INCLUDED is the ADXR line (the Teal line) as well as the ADX ‘cutoff’ point i.e. the Red line at 20.

Also: when ADX has fallen below both +DI and -DI (in which case, according to ‘the old man’, you should stop trading a trend following system) the ADX Oscillator turns Blue and hereby one of my many problems with this ‘logic’ is clearly displayed i.e. I see prices trending in spite of ADX being below +DI and -DI on all except the ADX(7) Oscillator.

What’s more: when the ADX Oscillator is has turned Blue: one should be looking to trade either the RTS or TBP. I don’t know for sure but from were I sit: both of those systems would have ‘landed you in hot water’ if traded when the ADX Oscillator was Blue.

Anyway: take a look see. Maybe you see the merit in this. If so: use it. If not: don’t use it!!!

Later.

Regards,

Dale (forexbrokesonline.net).

deltaadxoscillator.zip (73.1 KB)

1074

Another ‘offering’.

I’ve modified the ADX Oscillator to draw the Blue lines i.e. ADX is below both +DI and -DI in BOLD for readability.

I’ve also added Black lines i.e. ADX is below both +DI and -DI AND below 20 (also in BOLD for readability and just in case you cannot see that ADX is below 20 i.e. the Pink line).

Looks interesting.

Anyway: I’m about to email the latest version of the Delta Indicators to everyone. Please note that there are (again) some major changes to the SIS and, of course, the above additions.

Regards,

Dale. (forexbrokersonline.net).

Edit:

By the way: I’ve tried using this as a filter for the SIS (it’s just ADX/ADXR presented differently) and again I don’t see the merit. Put it this way: the SIS is to be used when ADX is above 20 (or 25). This would not have kept me either from using the SIS last week or getting whipsawed by the news. The only possible SIS filter that I can see in ADX is to only take long SIS trades when the ADX Oscillator is Red and only trake short SIS trades when the ADX Oscillator is Red. This of course means that you never stop and reverse with the SIS i.e. only stop (which is contrary to the entire design on the SIS). Also: there are many times when the SIS has given you a SAR signal by the Trailing Index SAR and the result of the SAR has been long trending trades which you would have missed out on.

Anyway (for those of you who trade at Delta and receive the indicators): take a look and see. As I’ve said before many times: sometimes I can no longer ‘see the wood for the trees’ because I’m so ‘caught up’ in our ‘Wilder stuff’!!!

eurusdadxoscillatorsmodified.zip (74.3 KB)

1075

Good (very early i.e. 04h27) Monday morning all.

Always a joy to ‘pop over’ to the thread and read all the new posts!!!

Anyway: I just though that you may (or may not) like to know that in my frustration on Friday I opened some positions on the 1 hour charts but then decided to switch to the 4 hour charts to give myself some ‘time away’ (I left the positions opened on the 1 hour charts open because they were still valid on the 4 hour charts). Those positions (as of right now anyway) have all but covered Thursday and Fridays ‘fiasco’, putting me ALMOST back where I was after my ‘pip feast’ on Sunday / Monday /Tuesday last week.

There’s hope yet!!!

Regards,

Dale (forexbrokersonline.net).

1076

Well,

In an effort to pass the time in the ‘wee hours’ I’ve been looking at OUR ADX Oscillator and I’ll tell you one thing that I DO in fact notice and it appears to be ‘without fail’ i.e. when a trend is about to reverse you get an immediate drop in the oscillator i.e. after a peak has formed, as soon as you get a line shorter than the one preceding it, you can ‘bet your bottom dollar’ that the trend is either going to end or reverse. Of course: this is ‘pure ADX’ as detailed in ‘the book’ i.e. it’s just a whole lot easier to see with the oscillator.

(It’s particularly clear on the 1 hour time frame using ‘ADX7Oscillator’).

Regards,

Dale. (forexbrokersonline.net).

1077

Good (Wednesday) afternoon all.

I’m actually pleased that nobody is posting because I’m assuming that this means that nobody is trading and it’s then also reasonable to assume that nobody else is getting ‘fu*ked’ like I am at the moment!!!

I’ve been getting very nicely ‘chopped up’ by the SIS this week again and this through no fault of mine i.e. I’ve not been messing around with other systems; I’ve been concentrating on the trades and religiously placing orders where they should be; I’ve not been ‘second guessing’ the entries and exits; I’ve been doing everything 'by the book. And still: I’m ‘bleeding cash’ like it’s ‘the fashion’!!! I just thought that you should know.

One thing that has come out of this though:

As with every book written on this subject: only the ‘perfect’ examples appear to be given and in these markets: the chances of you getting price movements that even VAGUELY resemble these ‘perfect’ examples are NIL!!! Now I’ve managed to work out that most of the losses THIS week (this is becoming a way of life again by the looks of things) hav been incurred on stops and reverses at losses i.e. entering the system correctly; once ‘in’ placing the INDEX SAR order; price retracts; triggers the INDEX SAR (you are of course now at a loss); place the INDEX SAR order again; price retracts; triggers the INDEX SAR (you’re at a loss again of course); and so on and so forth. BUT: I’ve now figured why this is happening (I’m trading the four hour charts by the way). Every time this has happened it’s been when the INDEX SAR order has been placed at a HSP or LSP where the is no CLEARLY DEFINED HIP or LOP. In other words: in the example in ‘the book’: Wilder ALWAYS has a nicely formed HSP / HIP or LSP / LOP. Now I know that he says that if you cannot see the HSP and corresponding HIP (or you cannot see the LSP and corresponding LOP) that you ‘assume’ that the HSP preceded the HIP by one day. THAT’S not the problem. The problem appears to occur when the bar PRECEDING the HSP / HIP or LSP / LOP does not allow for the formation of a clearly defined HSP / HIP or LSP / LOP. In other words what I’m saying: EVEN ALTHOUGH you may be seeing a clearly defined HSP or LSP BUT because of the PRECEDING bar EVEN IF the next bar closes higher or lower a HIP or LOP will NOT have been formed.

I’ve attached a chart demonstrating what I mean.

The fact of the matter is this: if you cannot see a CLEARLY DEFINED HIP / LOP ‘near’ to your current HSP or LSP then that HSP or LSP must be ignored and you would then go back in the chart and use the previous VALID HSP / HIP or LSP / LOP for your INDEX SAR UNTIL a new VALUD HSP or LPS has formed. If you dont’ do this: you’ll end up ‘bleeding in the street’ as I am now!!! Needless to say this moves your INDEX SAR’s MUCH further away in certain instances but, in trading and concentrating on the SIS, it’s clear that it needs ‘room to move’ and sometimes requires ‘nerves of steel’ to follow.

I don’t know if you agree with me or if anyone is still interested in these systems. I know that’s a ‘shock statement’ to be coming from me but I’d not blame anyone for having their doubts after seeing my last few posts I assure you. Unfortuanately: I have no choice. I have to carry on until I ‘can no longer’ so, in the interim, I may as well keep trying to sort out the problems. The only thing that keeps me going with these systems right now is the fact that I know (for a fact) that Wilder made his money with these same systems. Admittedly: that was a long time ago but, especially where the SIS is concerned, other than market volatility the way the market ‘works’ cannot have fundamentally changed i.e. that ‘line’ that is the ‘true market’ is still there although it would appear that it’s a little harder for me (and the SIS) to ‘spot’!!!

Regards,

Dale.

Edit:

I had a thought i.e. maybe my above explanation is not clear. In other words: the HSP’s and LSP’s should be used as INDICATIONS that you should look for a HIP or a LOP and should not be used ‘literally’ as it were. Another way: when you have a HSP or LSP formed then it’s time to start looking for a HIP or a LOP. If there is no CLEARLY FORMED HIP or LOP direcly above or below the HSP or LSP then look for the ‘nearest’ one’. What I’m saying is that the HIP’s and LOP’s are more ‘important’ than the HSP’s and LSP’s. Because (believe it or not) the SIS is ‘price based’ and not an ‘indicator’ as some would have you believe: these HIP’s and LOP’s are where you look for swing points and, if a HIP or LOP is not clearly identifiable, then there is no swing point. That’s how I interpret things and I can tell you that this would have saves me load so $ this week had I seen the pattern earlier.

gbpjpychopping.zip (69.4 KB)

1078

oh my god!!!

guess what Dale this was my first week to read the book.the moment i decided to do so your posts turned me off!! any ways i am trying to look for the Volatility system like writing them on an excel sheets and see what this man wants me to do when i have a signal with the SAR so my old friend (Dale) shall i give up this system or what???

but to be clear and honest nothing of the systems are working on this market right now i heard someone was saying it is a an excellent time for traders to have some rest till the market goes in it is normal conditions right now it is not as there are 17 banks in the states so far has announced there bankruptcy things has to be crazy at the moment of time but till when???

have a nice trade

Akram

1079

Good (Thursday) morning.

Akram:

I cannot advise you to use or not to use these trading systems. I believe in them as they have worked very well for me this year up until a few weeks ago. Having said that: a few weeks ago my losses were my fault not the fault of the trading systems. After that: I made ‘loads’ about two weeks ago trading the SIS on the 1 hour charts. This worked for two or three days but I sooned realised (again) that the 1 hour charts have lots of ‘market noise’ and later on that week I started to get ‘chopped up’ and gave back most of profits due to whipsaws. I then went to the 4 hour charts and things were good until Tuesday and yesterday when I started get ‘chopped up’ on the 4 hours charts as well. I think I have found the reason for this though (as detailed in my above post as well as further on in this post).

At the moment I’m only concentrating on the SIS i.e. we’ve all used all of the other systems in the book and they appear to have worked with varying degress of success (although from what I gather: many ‘failures’ were not due to shortcomings in the trading systems but rather shortcomings of human beings).

Let me put it another way: none of these systems has COST me money WHEN followed correctly up until now and I personally believe that the only reason the SIS is messing me around right now is because, even after all the time and effort I’ve put into it, there are still things that I’m not clear on (or have not been clear on up until now anyway).

As far as ‘sitting out’ is concerned: that’s probably very good advice. Unfortuanately I have deadlines to meet so I don’t have that luxury. In addition to that: if I can get the SIS to work REASONABLY profitably NOW then believe me: it’ll make LOADS when this lot calm down.

Now further to my post above regarding the HSP’s / LSP’s / HIP’s / LOP’s I’ve attached a chart of a ‘primo’ example of what I’m talking about.

Regards,

Dale. (forexbrokersonline.net).

eurusdchopping.zip (65.6 KB)

1080

I’m ‘gobsmacked’!!!

Take a look at the attached chart and tell me is you can see any differences between the three indicators on the chart??? (I see two extremely minor differences HOWEVER one of those minor differences makes the difference between waiting for another signal or taking the signal given by the bottom two and thus potentially saving LOADS of profit on short positions that I’m in now).

(Take a look at the chart before reading further)!!!

Now let me tell you why I’m ‘gobsmacked’:

The top indicator is the ASI (SIS) ‘in all it’s glory’.
The second indicator simply represents the ROUNDED closing prices.
The third indicator simply represents the actual closing prices.

My question:

OTHER than the TRAILING INDEX SAR and the instructions on HOW to trade the SIS: what on EARTH is the point of the HUMUNGOUS ASI equation???

Thoughts???

Am I again missing something here???

Regards,

Dale.

gbpjpycompareasiclosing.zip (72.6 KB)

1081

Well, well, well!!! Here is where we come ‘full circle’ with the SIS!!!

After my ‘discovery’ (above) I now ‘see’!!!

The MAJOR problem with the SIS is, and has always been, the fact that you can get ‘chopped up’ badly when stopping and reversing. Of course: once that action stops it’s profitable and, I would say, more so than any of the other systems.

I’ve now found this:

The ASI calculation, as with all Wilder’s calculations, goes to great lengths to take into account opening gaps which, when trading commodities and stocks, could be of a ‘material’ nature. However, as you know, with forex pairs, we do not HAVE opening gaps (with the possible exception of the difference between Fridays close and Sunday’s open).

So: given the fact that we have no gaps: the ASI is giving us nothing more and nothing less than a line that is ‘drawn in the sand’ of the closing prices of each period / bar.

Now this may not sound like a big deal BUT:The ASI calculation ALSO incorporates the TRAILING INDEX SAR as you know. HOWEVER: the TRAILING INDEX SAR is a fixed number of ASI points and the distance between the TRAILING INDEX SAR and the LIMIT (which, again, we do not HAVE on forex pairs) are in direct proporation to one another. Now this IS MAJOR for the simple reason that TRAILING INDEX SAR is fixed and based on the LIMIT so therefore what’s happening when you do NOT have a LIMIT is that the TRAILING INDEX SAR is stopping you out far too quickly and is not taking into account the market volatility which, thinking backwards, is not an issue with stocks and commodities for the simple reason that the market volatility is ‘capped’ because of the LIMIT’s imposed!!!

In other words: with stocks and commodities, no matter HOW volatile the markets are, the volatility is ‘capped’ by the LIMIT imposed. Price can ONLY move SO much in one direction or another before triggering the LIMIT. And HEREIN lies the problem i.e. with forex pairs and other instruments where there ARE no LIMIT’s: there is also no ‘cap’ or LIMIT as to how far price can move during a period BUT, by using Wilder’s ASI with the TRAILING INDEX SAR which is fixed, it means that your TRAILING INDEX SAR is ignoring this fact and is always at a preset amount above or below price!!! This has a MAJOR consequence when placing stop and reverse orders.

In addition to the above: the ASI (even without a ‘smoothed’ ATR) is, somehow, and I believe it’s due to rounding errors, actually ‘smoothing’ the ASI ‘plot’. So far today: this made the difference between stopping and reversing with about +274 pips ‘in the bank’ on a position or waiting for the only signal being shown by the ASI to SAR which was far enough away to not only lose profit but actually realise a loss on the same position.

It pains me that after all this time: I’ve actually been looking at something that I thought was so extremely complicated, but in reality, is so simple it’s ridiculous!!!

Now this does NOT mean that this section in the book is useless i.e. it does indeed tell you how to trade this type of system (which I for one would not have had a clue as to how to do this) but, because of the different types of markets, some serious changes are, and have been, necessary.

Now I’ve ‘coded’ a new SIS for my ‘clients’ (and myself of course).

Why it’s better:

For one thing it’s using the ACTUAL prices of the instrument or pair not some convoluted equation to get ‘one size fits all’.

Also: the TRAILING INDEX SAR now ‘adjusts itself on the fly’ according to market volatility.

And in addition: may more HSP’s and LSP’s are drawn on the chart and are more obvious to the naked eye so the chances of them being missed is nil (unless you’re going blind of course)!!!

This system is called the ‘Fintrans Swing System’ (original huh)!!!

I’m going to continue testing today, and, if satisfied that there are no errors, will email the same to my ‘self trade’ ‘clients’.

The irony of this discovery:I was actually trying to find some code to do manual rounding of the ASI this morning as there is no ROUND() function in Delta’s script language. I found the ‘code snippet’ that I was looking for and created a new test indicator to make sure that the code worked before I applied it to Wilder’s ASI. THAT is when this ‘similarity’ ‘jumped out screaming’ at me!!! And one thing led to another and HERE WE FU*KING ARE!!! FULL CIRCLE!!!

Regards,

Dale.

1082

Well, here it is!!!

If you think I’m ‘joking around’ with above posts then take a look at the attached chart of the Dow.

There are two ‘indicators’ on the chart:

The bottom one is the SIS / ASI with all the ‘bells and whistles’ i.e. the correct equation with the correct LIMIT for the Dow. You name it!!!

The top one is the FSS (‘Fintrans Swing System’).

Now there may appear, at a quick glance, that there are no differences but look closely i.e. a (very) few times there is a slight difference in the ‘plots’ and each of these is due ONLY to an opening gap on the Dow.

What’s more:

The TRAILING INDEX SAR (represented by the dots) of the SIS / ASI will ALWAYS be 60 ASI points on either side of the ASI and, when translated to ‘price action points’ means that it will always be exactly the same distance on either side of price no matter how much price has moved during the period. Why? Because it’s ‘governed’ by the LIMIT of the Dow (for example).

The TRAILING INDEX SAR of the FSS, on the other hand, will ‘adjust’ itself after the close of each new bar to the current volatility of the market. THIS in ITSELF should keep away from almost all whipsaws. And what kills me here is the shear simplicity!!! I’ll give you three guesses as to what the distance is between the FSS TRAILING INDEX SAR and the FSS ‘Closing Price Index’ (which replaces the ASI in the FSS)??? TWO TIMES ATR(14) which millions of people around the world use to place stop orders!!! This, of course, ‘lends itself’ to some further experimentation i.e. increasing the multpile of ATR could / would make your SAR ‘individual’ and ‘unique’ to you and / or the instrument being traded (and if you’re one of those people who are ‘paranoid’ about ‘stop hunting’ then this has an added benefit for you). As you can clearly see though: the ‘default’ is 2 X ATR(14).

The other ‘rules’ that I ‘discovered’ still ‘stand’ however i.e. only when there is a CLEARLY DEFINED HIP or LOP above or below a CLEARLY DEFINED HSP or LSP is that point used to place a SAR order. This again should keep you away from almost all the (other) remaining whipsaws!!!

Edit:

Something I forgot to mention: the FSS ALSO gives you the exact value in PRICE of the TRAILING INDEX SAR so one could, if one wanted to, and one was in profit, actually ‘trail’ the price according to the TRAILING INDEX SAR instead of waiting for a close above or below the TRAILING INDEX SAR. Of course: if you use this method you’d NOT SAR but only TP.

Have fun. I think I can ‘put this one to bed’ now!!!

Regards,

Dale.

dowfssasi.zip (56.8 KB)

1083

Dale,

Another cracking piece of work! It is fascinating to watch the output from huge brains of people like yourself and J. I just wish I had more to contribute.

Now that you’ve said it about the limit, limiting volatility it seems obvious. At a first reading it looks like you’ve generated a hybrid of SIS and VS.

If you’ve managed to dump the bath water and hang on to the baby here this could be very exciting.

Derek

1084

Hey ‘Father Ted’,

Thanks for the nice words. At this point they’re even MORE appreciated!!!

It’s a ‘bitter sweet victory’ I’ll tell you. TWENTY MONTHS off and on I’ve spent ‘tinkering’ with the SIS. TWENTY MONTHS I tell you!!! And it comes down to THIS!!!

I’ve been using it today and, although it’s ‘early days’, I can already see the difference especially where the TRAILING INDEX SAR is concerned and the false SAR’s or whipsaws from before. And it’s just as well i.e. even although we’ve ‘saved the baby’ it’s in dire need of resuscitation!!!

Anyway: if nothing else I’ve got to be given 10/10 for trying!!! Hell: I’d give MYSELF that!!!

Again thanks.

Regards,

Dale.

1085

ok Dale,

i will keep on watching the vs system till i know i am ok with it and i know every thing on this system then i will go with the SIS system thanks for your words i was so upset as i thought it is not good systems . will keep u posted and if i need any help i will ask u here

Regards,

Akram

1086

Good (Friday) morning!!!

Well: I know it’s ‘early days’ but the FSS appears to be working far better than the SIS. I’ve been in two trades since yesterday and both would have been stopped and reversed at losses had I been using the ‘standard’ SIS. The FSS, however, has kept me in the trades and both are showing profits. It’s obvious, of course, that far more testing has to be done but so far so good.

One other ‘discovery’ today (it never stops does it???):

Believe it or not: Delta’s charting module cannot ADD!!! LOL!!!

OK: it’s not quite as bad as that but beware (for those of you who are modifying code in the platform).

One would EXPECT that this function:

SUM(HIGH,7)

would ADD the HIGH’s of the last 7 bars???

Well not so!!!

The above adds the HIGH’s of the last 8 bars!!!

In order for this to work correctly you have to use the following:

Let’s say that you wanted to add the HIGH’s of the last 7 bars:

SUM(HIGH,7)-REF(HIGH,7);

or if you’re going to paramterise it:

SUM(HIGH,PERIOD)-REF(HIGH,PERIOD);

Nice huh!!!

This affects every indicator or system where the SUM() function is used e.g. the VHF.

FORTUANATELY this does NOT affect the SIS / ASI but only for one reason i.e. although the SUM() function is used the parameter (PERIOD) being used is a 0 (ZERO). This quite correctly adds each and every single bar (in the case of the ASI it’s adding all the SI’s). Because the FSS does not use the SUM() function it’s, of course, not affected.

What this ‘discovery’ NOW allows me to do is parameterise our ADX and ADX Oscillator.

Just something to bear in mind.

Regards,

Dale.