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

[dpaterso]

Heeeyyy everyone!!! It's REALLY NICE to see some people around over the weekend (normally this thing is 'dead' over a weekend)!!!

Don't worry Akram:

I was just making an observation (about Parabolic SAR) is all. My 'money' is still on, and my 'enthusiasm' is still for, the DMS!!!

That said: I'm now in THREE PTPS trades from yesterday i.e. EUR/SEK, EUR/NZD, and EUR/NOK (all short) (one in NICE profit, one in a LITTLE profit, and one in an 'ever so tiny' loss).

Cody:

Tell you what: if you mange to 'rustle up' $100K then you can just as easily 'rustle up' ANOTHER $100K for me!!! LOL!!!

And that AVA account is not even CLOSE to all the trades I've done since December last year on my live accounts so can you IMAGINE!!! Based on my percentages that account COULD be almost on $200K!!!

But THAT is my point as well: with THAT type of money and these trading systems one would seriously not have to worry about following trades through or anything like that. Let me put it this way: under NORMAL circumstances I could 'live the high life' over here on about $9K per month i.e. that would include mortgage, car, and all other monthly expenses as well as credit card and loan repayments. Now I'm unsure what type of money you would need in the US to 'live the high life' (well: let's say 'comfortable' i.e. not having to worry about 'the end of the month bills' EVERY month and just being able to 'do your thing' like have a hobby or two, go out to dinner when 'the mood took you', you know, that kind of thing i.e. not be 'limited' by money or lack thereof. I'm obviously not taking about 'lavish parties' and 'overseas trips every other day' just 'normal'. What would you need in the US)??? I digress. The point is that with that type of money in an account and no investors to pay or stuff like that one could simply stop trading once you've done a few profitable trades like that and 'come back another day' (unless of course you want to start saving for retirement as well). The point is that there is FAR LESS pressure to follow trades through 'to the last' and the BEAUTY AT LEAST of ALL of these trading systems is that 99% of the time your ENTRIES are 'good' and the positions turn to profit most times i.e. it's the SAR's that can 'kill you' (which is what we're trying to sort out at the moment).

Anyway: I've just had a nice cold swim so I'm feeling 'well refreshed'!!! On with the spreasdheet!!!

By the way: as I do this work I am also coming to the STARK realisation that a combination of the DMS and TBPS 'theories' is not a bad one either (J. also has some ideas on this). What I'm saying is this: as long as ADX is rising you go long at market (as long as the oscillator is green) and short at market (as long as the oscillator is red) you immediately set your TP to the TBPS TP and also use the TBPS SL and also use the SAR as per the TBPS rules. In other words: you're trading the TBPS but IGNORING the TBPS MF i.e. ADX 'replaces' the MF for confirmation of trade direction and you're just using the TBPS daily values for TP, SL, and TBP. I am SERIOUSLY considering THIS alternative. You must realise, of course, that the REASON that the TBPS and RTS TP levels are almost ALWAYS hit is because they represent Pivot Levels R1 and S1 and almost ALL of the time those Pivot Levels are 'penetrated' during a trading day (session). Again: the main 'drawback' of trading a 'method' such as this would be the spreads / commissions involved so it would AGAIN (probably) be preferable to trade only low spread instruments / pairs.

Oh, sorry, I nearly forgot: AVA offers commodities but I'm still waiting for them to add additional historical data to their charts. Otherwise we'll just have to wait for another month or two until the ADX Oscillator has 'settled down' (a 'period of grace' is required for the 'smoothing' to take effect is all). GCI also offers commodities but nowhere near as many as AVA. That said: you're in the USA so I don't think you'd be able to open an account at either of these brokers. You MAY be able to open and account at a broker like Saxo Bank (not sure here) and they also offer commodities. They have a proprietary platform though so no I would not have the 'indicators' for their platform. That said: there is nothing stopping you from using an AVA or GCI demo account for this purpose. I'm afraid you'll have to look into this yourself though. As I say: because you live in the USA you have a problem with these (my) brokers and the trading of commodities and CFD's which really is a pity. There ARE INDEED brokers in the USA where you would be able to trade such instruments but those guys (TradeStation Securities for example) require HUGE amounts of capital and a minimum 'maintenance margin' in the account in addition to a sizeable, personal, nett worth!!! Probably not a BAD thing though i.e. they're really ensuring that anyone that decides to get into this business is NOT 'betting their last dollar'!!!

Regards,

Dale. (forexbrokersonline.net).

[kaalilaatikko]

Originally Posted by dpaterso

(I'm not saying that this IS IN FACT the case because I've not gone and compared ADXR values for all the pairs on a specific historical date but I suppose in order to state this as FACT then that is what I have to do. If you have an easier method of goind this then DO FEEL FREE TO LET ME KNOW)!!! Let me put it another way: do you have ANY way of generating a list for ALL the pairs giving their ADXR value on, let's say, 2007-08-07 (which is where you and I started testing from to 'exchange notes')??? That would very quickly tell us whether or not we would have been even LOOKING to trade those pairs at that time. If this IS possible: could you generate ANOTHER list of ADXR values on 2008-08-12 (which is where I 'deem' these wonderful downtrends to have started for GBP/USD and GBP/JPY particularly).

As a matter of fact, I have an easier method for this. It is called "press the enter"... Please see the attached file.

I've got another idea for all this. How about entering only when ADX is above ADXR? This seems to filter a number of bad trades, but also some good ones. I'm not yet sure which wins.

J.

[dpaterso]

Very funny!!! I don't HAVE an 'enter' to press that will give me the ADXR values!!!

THAT BEING SAID:

As I suspected: GBP/USD and GBP/JPY are both 'non-events' using ADX(14) and would not have been considered if using ADX(7) either and those trades resulted in some SERIOUS pips!!! So correct me if I'm wrong here but I don't see noting the value of ADXR as a solution??? Then again: did it keep you 'out' in August / September / October 2007 as WELL and 'put you on to better instruments to be traded' both then AND now???

Regards,

Dale. (forexbrokersonline.net).

[kaalilaatikko]

It seems to me as well that ADXR ranking is not a solution here. Unfortunately I cannot answer your question about the trades in fall 2007. Backtesting that according to your question would be that laborous, and I priorise studying some other possible paths higher.

Here are some of my late findings that may carry value in adjusting DMS:

(1) When ADX ascends above 50 (or somewhere there), it could be beneficial to place a stop order at the opposite end of the latest candle. Argument: when ADX is high, the probability for it to turn increases, and sometimes the drop can be pretty strong. On the other hand, the trend may continue still for a while, and this could lock the profits nicely while still staying in this trend.

(2) When ADX changes colour while it is descending, the stop order would not be placed immediately, but only after it has turned upwards. Now the extreme point would not be as defined by Wilder, but as the lowest/highest point between colour changing and turning upwards. Argument: this is a risky place to enter, but we might also lose some pips in the best case. Here I would favor not to try to squeeze all of the pips, but rather minimise the risk.

(3) There should be only one stop order in effect at the time. If I don't recall it wrongly, Dale discussed the possibility to have two stop orders in effect at the same time. I run across the same during my studies, but I'm now leaning towards a solution where you cancel a stop order whenever ADX changes its colour.
Argument: the stops seem to be placed in more optimal places when done this way.

(4) When ADX turns downwards, it may not be the best thing to TP immediately, but instead place a stop at the opposite end of the candle, like in (1). Corresponding argumentation also here.

None of these involves any extra parametrisation, which I think is good (except for defining an ADX limit level in (1)). Moreover, if I'm correct with these, it could be possible to keep the entering stop order relatively close to the extreme point, and the number of trades would not be reduced to only very few.

All comments are most welcome!

J.

[dpaterso]

Hey J.,

Thanks for going to all of that trouble for us.

It all looks and sounds good to me and I SHALL be taking a good hard look in the next few hours at all the options presented.

Just one thing (I only mention this because this is what I was trying to explain the other day but you've just managed to put in in 'better words'): your point (4) is what I was trying to explain the other day i.e. you do not TP IMMEDIATELY when ADX turns down but you wait a the CLOSE. If, after the CLOSE, ADX has turned down then you place a STOP ORDER (that will effectively lock in profit) at that point i.e. you don't simply TP IMMEDIATELY AT MARKET when you see ADX starting to turn down for the simple reason that most times price will turn again in your favour WITHOUT closing the position.

(This is for eveyone else to consider i.e. maybe come up with a solution that suits you):

There are many (popular) methods for placing TRAILING STOP ORDERS (which, if I understand things correctly, is something that J. is proposing 'in a fashion'):

Use the highest high or lowest low of the preceding two bars (Wilder).
Use the highest high or lowest low of the preceding five bars (Williams).
Use the furtherest of the two preceding opposing fractals (Williams).
Use a multiplier of ATR (2 x ATR is popular).
Use a percentage of ATR (10% or 20% is popular).
Use a set number of points / pips (facility provided by the broker).
Use a previous Pivot Point / Level (popular).
Use Fibonacci Lines / Levels (popular).
Use ZigZag Support / Resistance (noted).
Use 'round values' (noted).

I suppose the list could go on but these are the various methods I have encountered thus far. Just some 'food for thought'.

Look: at THIS point I'm still 'none the wiser'. I also keep thinking about this: I agree that if you were trading a single pair or maybe one or two pairs (like GBP/USD and GBP/JPY) 'back then' that you would have 'come a cropper'. That said: I take trades on every and any instrument / pair when given a signal 'money mangement permitting' and very rarely will I have 100% of my open positions individually showing a profit at a given time. That said: I have no problem if the NETT result showing for, let's say, ten or twelve open positions is a NETT profit. 'Back then' (during the selected test period) yes: you would have gotten 'well screwed' by GBP/USD and GBP/JPY BUT would these two pairs have affected your NETT performance for the same period i.e. would the results have been the same with a 'basket' of currencies??? It would be real nice to ensure that every single trade got closed out at a profit on every single pair but we all know by now that this is just not going to happen. Do you see what I'm getting at??? I mean this past month: I've opened some trades and closed others and offset some losing positions with some profitable positions and taken profit on others that I was 'just not happy with' and taken losses on others for the same reason. Stuff like that and I've managed to 'produced the goods is spades'. And if the truth be told: for the most part I've not even caught the 'tailend' of those long downtrends that I've been 'banging on about' i.e. I've not really even 'caught a trend' in the past month (other than USD/CHF and GBP/CHF both of which I 'bailed on' early).

I need to do some more backtesting though. As I said earlier (and I've spent a good deal of time looking at this this afternoon): for me I can see that it's going to be EITHER just following 'pure' DMS as we have been doing and maybe using some of J.'s ideas to lock in some profit OR going long at market after the close if the ADX Oscillator is green and going short at market if the ADX Oscillator is red and using the TBPS values for TP and SL and SAR when necessary (as per the TBP). For added 'insurance' it would be wise to ensure that ADX is rising when taking these trades and not to take any trades while ADX is falling.

So far: that's 'the best' I am able to 'come up with'!!!

Regards,

Dale. (forexbrokersonline.net).

[dpaterso]

Good (very early Sunday) morning!!!

'The old man' says in 'The Delta Phenomenon' that sometimes he used to get an idea at 03h00 in the morning and he'd 'jump up' and 'jot it down' (I'm not sure if those are the EXACT words he used and it's far too early on a Sunday morning for me have a 'quick squiz' through 'The Delta Phenomenon' to find his EXACT words). Suffice to say that this has just happened to me (AGAIN) (so maybe there is hope for me yet)!!! LOL!!!

NOW:

I don't know if this is because of something that J. ACTUALLY said or was 'trying to get at' or if it's BECAUSE of something that he said that I didn't quite understand or what BUT it neverthless 'sparked something off in my brain in the wee hours':

I think I have found a solution to THE DMS problem and I believe that it's ELEGANT IN IT'S SIMPLICITY AND USES ONLY WILDER 'LOGIC'!!!

Take a look at the attached chart of GBP/USD. On the chart there is the ADX Oscillator (ADX(14)) and a 'new' 'indicator' called the 'AverageDirectionalIndexDX'. And YEP: YOU GUESSED IT!!! All it is is a display of the DX as opposed to the ADX!!! In 'the book' 'the old man' says as follows: 'The higher the DX, the more directional the movement; the lower the DX, the less directional the movement'. This 'got me thinking' some time back (but until now the true 'power' of this statement has eluded me): when the DX is rising rapidly then the directional movement must be accelerating but when the DX either LEVELS OUT or STARTS to drop then the directional movement must be decelerating not so??? Now take a look at the chart again. WHAT IF you ONLY took signals from the ADX Oscillator the moment there was an upturn in the DX??? And WHAT IF the moment the DX 'levelled out' OR turned down you placed a STOP ORDER / TP ORDER above or below the corresponding bar??? And IF your STOP ORDER / TP ORDER was 'taken' then WHAT IF you re-entered in the direction as indicated by the ADX Oscillator by placing a 'normal' DMS entry order as soon as (if) DX turned up again??? I will go through the whole thing tomorrow again (with myself and you) but take a look and see what you think anyway. The basic idea is this: you're entering a DMS trade based on ADX (which is 'normal') and you're STAYING IN the trade AS LONG AS there is an acceleration of movement in your favor BUT as soon as that acceleration starts to 'falter' you're placing a STOP ORDER / TP ORDER for 'protection' SHOULD this deceleration be the end of the directional movement in your favor and the start of directional movement in the opposite direction. ALSO: you are re-entering the trade IF the directional movement starts accelerating AGAIN in your favor. Make sense???

Now as I was typing this I was reminded of what J. was referring to as 'Velocity' some time back??? Was THIS what you were talking about J.??? If so then I apologize for 'plaguerising' your work. Either way: I believe that this DOES have merit and if you're 'hell bent' on trading the DMS with protective STOP LOSS / TP ORDERS then 'this is the thing for you'!!!

Regards,

Dale. (forexbrokersonline.net).

[sverigesson]

Dale,

Really early Sunday for you, just the end of a Saturday evening for me (and my daughter's bedtime)! After considering exactly what the DX means, and checking out your graph, it seems like there's some merit in what you're saying. I'd check it out myself if I weren't swamped nowadays (I now have great respect for all my university instructors). I hope you guys keep up your research; there've been more good ideas from the last few pages than I've seen in a while!

Have a great Sunday, and then lets rock and roll when trading opens again!

Happy Pips,
Cody

[dpaterso]

I am 'ON FIRE' ladies and gentlemen and boys and girls!!!

To take the above idea a bit further:

First: I just realised that it was possible to display the DX and the ADX Oscillator TOGETHER IN ONE INDICATOR (see attached chat again).

Second (and this is the 'kicker'): using THIS method the 'standard' ADX 'rules' may no longer be applicable (the 'rules' of ADX falling below both +DI and -DI or ADX falling below 20 - 25 or ADXR falling below 20 - 25).

Now here is why I say this:

During my backtesting my WORST LOSS on GBP/USD was -508 pips. The signal was given on 2007-11-27 to go long and the order was executed the next day at 2.07560 (I'm just using the high, MIDDLE, price of the bar here for demonstration purposes) and a signal was given to SAR on 2007-12-05 and the order was executed the next day at 2.02480 (again just using the low, MIDDLE, price for demonstration purposes) resulting in a loss for this trade of -508 pips.

Now because I'm ignoring the 'standard' ADX 'rules' with this 'new' 'method':

A signal to go long would have been given on 2007-11-21 and it was a valid signal because the DX had increased from the previous day. The order was placed at 2.06990 and was executed on 2007-11-23. On 2007-11-30 the DX turned down so a PROTECTIVE STOP ORDER was placed at the low of the bar at 2.05340 and this stop order was executed on 2007-12-05 (which ironically is the same bar / movement in price that caused the HUGE loss above). The loss NOW on the SAME TRADE is only -165 pips!!! The exact same thing happened on the very next trade i.e. the loss was cut by WAY more than half!!! And the best part: this 'method' appears to have VERY little effect on the profits to be made when there is a clear and strong trend in place!!!

And what's more: you are consistently booking profits as well during a trend so every time you re-enter it may even be possible to increase your lot sizes as per our money management rules!!!

Please DO take a look see!!!

OK: at this early hour I'm not ENTIRELY sure that TOTALLY ELIMINATING all the ADX 'rules' is a good thing (although NOT eliminating them doubles the loss of the trades I speak about above. That said: the losses were STILL less than half of the original losses at the time of backtesting). I need to check this a bit more thoroughly. That said: MAYBE the DMS could NOW be used in BOTH a trending AND a ranging market / environment because some of my preliminary backtesting shows PROFITS being made using 'pure' DMS on trades that otherwise would NOT have been taken AND no TBPS signal was generated!!!.

Regards,

Dale. (forexbrokersonline.net).

[dpaterso]

OK:

Last post on the subject (for now anyway).

I thought I'd take a look at one of those very recent 'stellar' trades e.g. USD/CHF (see attached chart).

Using this 'new' 'method' did not affect the trade IN ANY WAY excepting for one small thing: it would have saved you +104 pips on the exit!!!

Look: all this is REALLY showing you is turns up and turns down in the DX on which ADX is based the difference being that because ADX is an AVERAGE of the DX it's a lot slower to react whereas the reaction of the DX is almost instantaneous. The sheer 'beauty' here is that it's 'stock standard Wilder stuff' and all we're doing REALLY is using the 'basis' of the ADX equation to 'get us out earlier' and 'protect our profits that have already been made using ADX before it takes them away from us' or 'limiting our losses'!!!

Now OF COURSE this has got me to wondering whether or not the direction and value of the DX could be a far BETTER ENTRY 'indicator' for the TBPS than the MF????????? Hmmm. I'll probably 'dream' about this now and post my 'findings' in the morning!!!

Regards,

Dale. (forexbrokersonline.net).

[reapr]

Dale,
Is wilders dmi indicator any good as an indicator in MT4. I forgot where I found that. If you are familiar with it I was wondering if it really has any use?

Do you have a blog or site with useful wilder indicators or EA's for MT4?

It looks like most of my trades wil be scalping but could be long term on occasion.

Anyhow I did receive the book and look forward to devouring it. Yes it appears to be antiquated in the technology used to create the charts but from skimming it looks like it may be a rather fascinating read and ... rather useful. In other words worth the purchase.