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

[martin01]

Originally Posted by dpaterso

GOOD (Tuesday) morning all!!!

I thought I'd carry on my 'live trading post' here (because I'm going to reach the maximum number of words on that one very soon).

Well: after the last update of that post on 04/11/2008 at 03h43 I simply HAD to get some sleep. So: I placed stop orders (instead of stop and reverse orders because I did not want to get into new trades that I could not manage) as per the SIS at the HSP's that had been formed by that time. I awoke to a nice surprise this morning: none of the stops had been taken out so I'm still in the same three trades!!!

So here we go (I'll detail these trades until stopped and reversed i.e. that should be a good enough test for our change).

04/11/2008 08h39

+980 pips!!!

GBP/JPY: +548 pips
EUR/USD: +210 pips
EUR/JPY: +225 pips

(the above breakdown probably does not tally with the total pips because the prices are moving as I enter the results here).

Profit still locked in by previous stop and reverse orders at the previous HSP's but Trailing Index SAR is now the closest stop.

04/11/2008 09h11

+962 pips!!!

GBP/JPY: +542 pips
EUR/USD: +198 pips
EUR/JPY: +210 pips

(The above breakdown probably does not tally with the total pips because the prices are moving as I enter the results here).

Things have changed somewhat now. EUR/USD has formed a new HSP so the SAR order has now moved to there i.e. more profit locked in. GBP/JPY and EUR/JPY still have profit locked in at the HSP's previously mentioned. If price continues to retrace to the top of this hour then they will have triggered their Trailing Stop SAR's which will, of course, result in greater profits upon the stop and reverses for both pairs.

Hello Friend,

congrats!! $1000 in 3 days gr8. I had been trading this market few years back . And i was trading with the broker AVAFX. I have again decided to enter this market. And now i am searching for brokers for trading. I have found my previous broker changed with so many facilities. Now i am confused which facility should i ask my broker.

Well congrats once again for $1000 in three days.

[dpaterso]

Hello and 'welcome back' to the market!!!

I just need to make things clear though (just for others who may stumble across the last posts):

It's not $1000 it's +/- 1000 pips (I'm not 'at liberty' to disclose my $ per pips values).

Also: that was the current and 'best case' scenario at the time i.e. EUR/USD has just stopped and reversed at less profit than was included in the 1000 pips running total (see the end of my previous post).

As you may or may not know: we've done LOADS of work on Wilder's Swing Index System and I'm currently just 'proving' our changes i.e. that's why there is not point in me simply taking profit when I 'think' or 'feel' it's enough i.e. I have to follow the system through from start to finish to 'prove it' (before I 'bet the farm')!!!

Obviously: I'd recommend Deltastock or GCI if you're going to 'join us'!!!

But again:

Welcome back. Always nice to have some 'new blood' (especially with some experience) on the thread i.e. we're ALL stiill learning!!!

Regards,

Dale.

[dpaterso]

Hokay:

Now:

This is interesting.

Take a look at the attached chart.

One of the other ways of using the ASI indicator is to draw trend lines on it and see how they correlate with the price chart. Sometimes: trend lines drawn on the ASI are may not be so obvious on the price chart (as with RSI i.e. they're clearer on the indicators than they are on the price chart or they may be used to confirm trend lines drawn on the price chart).

Now you'll note that I've drawn a trend line joining the last two ASI highs. You'll then note that, using an ATR(14) multiplier of 0.9, the SAR bar was exactly the same whether you used the trend line or the TRAILING INDEX SAR to stop and reverse!!! And THIS is why J.'s backtesting showed that 0.9 was 'the sweet spot' for this pair.

Now:

THIS makes it easier to determine the correct ATR(14) multiplier for a particular pair wouldn't you say??? Simple drawing trend lines on the ASI for an instrument or pair and cross referencing them to a particular ATR(14) multiplier will give you the 'correct' multiplier for that instrument / pair!!!

J.:

Take a look.

Regards,

Dale.

By the way:

I'm now long GBP/JPY, EUR/USD, and EUR/JPY, and short USD/BGN (an 'old favourite of mine). Let's see how these trades 'play out'. I'm not going to post details every hour on the hour as I have been doing i.e. I think we've now been able to prove that our 'fix' works. If, for some very obscure reason, the nett results of all my SIS trades is a loss, I will post those results just the same way as I would post results of the profits.

[dpaterso]

You're not gonna believe this BUT:

I'm up +/- +182 pips nett on the four pairs since stopping and reversing (and the pip count is climbing as I type)!!!

I am NOW going to take a 1 hour nap!!!

Later.

Dale.

Edit:

Sorry, but I'm so 'chuffed' with our 'fix' that I just need to 'share' (hate that word i.e. reminds me of Opera Winfrey):

04/11/2008 14h09

+/- +502 pips nett on the SAR'd positions + USD/BGN!!!

I just took a nap as well. It seems this system works better when I sleep!!!

04/11/2008 15h06

+/- +757 pips nett on the SAR'd positions + USD/BGN!!!

I'm speechless. I've made big profits before as you know but not on this scale and particularly not PER COUPLE OF HOURS!!!

OK. Well. I think I'll go give Pivots another bash eh???

04/11/2008 16h01

+/- +736 pips nett on the SAR'd positions + USD/BGN!!!

Do you know why this works (well now anyway)??? It's definitely based on the principle that support becomes resistance and resistance becomes support.

04/11/2008 17h10

+/- +950 pips nett on the SAR'd positions + USD/BGN!!!

HSP's / LSP's have formed on all positions so profit locked in once again.

04/11/2008 18h14

+/- +1456 pips nett on the SAR'd positions + USD/BGN!!!

HSP's / LSP's were formed on all positions so profit locked in once again and orders not yet executed. It looks like the TRAILING INDEX SAR will become the SAR at this point.

Hmmm. I'm thinking now. I wonder if I could improve that 8/21 EMA System in Carter's book???

04/11/2008 19h08

+/- +1610 pips nett on the SAR'd positions + USD/BGN!!!

HSP's / LSP's were formed on all positions so profit locked in once again and orders not yet executed. It looks like I'll get a retracement here now and then new HSP's / LSP's will be formed and more profit locked in. With any 'luck' the Dow will make 10 000 today (these pairs all track the indices) so the new HSP's / LSP's and / or TRAILING INDEX SAR's will not get executed until a higher price is reached. If not: I estimate that about 50% of the current profit showing is already locked in.

04/11/2008 20h04

+/- +1488 pips nett on the SAR'd positions + USD/BGN!!!

As above for now.

[kaalilaatikko]

I found a bug in my fresh backtesting code, calculating L slightly wrong (having my indicator correct though). Moreover, to slip a bit in the implementational details, I returned to simple moving average in both my indicator and my bactester/to-be trade engine. Because of finding that bug, I had to do some more backtesting. Results can be seen in the attached picture.

The results are not that splendid as what Dale is showing at the moment, except for GBP/JPY. I'm not sure whether this is caused by momentuously spectacular market right now, the fact that this backtesting does not take SIS entry/exit rules into account correctly (considering only closes), or what. But the work continues.

It seems for me even stronger than before that the candle count 14 for ATR is a good general choice. I haven't found any essential benefit from moving it anywhere from there. The best ATR factor fluctuates a bit more. But then, there are not that many pairs that are good for this system. 0.9 is the first guess, but the optimum could be somewhere near.

And this time I am ruling the 5 min chart completely out, though it showed tempting on the last round. No proper profits for any of the pairs here.

Note also that spread is not considered here, so it would take these figures down a bit.

J.

Edit: while browsing through the detailed data of GBP/JPY backtesting trades, I noticed that the losing trades often come in streaks. I am going to try the following in my trading algorithm: After every losing trade jump out and wait until getting a fresh entry signal, i.e. when the previous significant HSP or LSP is exceeded. I believe that this should then eliminate some bad trades that are caused by temporary ranging.

[dpaterso]

Hello J.,

Just so that you know:

I'm using 0.9 x ATR(14) and my ATR(14) uses an 'unsmoothed' EMA. Using those settings is what's giving me these results on the 1 hour timeframe.

Regards,

Dale.

By the way: I posted a post for you earlier today (the one with the trend line drawn on the ASI which I figured could be used to find the 'sweet spot' for the ATR multiplier for a particular pair).

[chdorry]

One thing to remember, the SIS is designed for trending markets which we definitely seem to have today. For example, we're seeing a pretty hefty move in a single direction before a SAR.

In Dale's trades for examples I see the following moves for the day based on Delta platform (since candle closed last night)
EURJPY +399.5
EURUSD +315.6
GBPJPY +89.3
USDBGN -378

So while Dale is getting great results off the one hour timeframe, we definitely are seeing a trending network. So while the SIS is king right now, it will cause some problems/issues when we come to times of consolidation/ranging markets.

Again, it boils down to trading the right systems at the right times, and now is definitely a trend following system time.

Also, not trying to downplay the success but most indicators work very well in a trending market, but by no means is the SIS a magic bullet for any and all pairs, as the backtesting has confirmed and as the market changes (we won't be ranging forever, I can assure you of that)

Great trading this week to date Dale, hopefully you've recovered your losses from last week (with interest).

Just making sure people realize there's more to this than turning on a 1 hour chart and watching the SIS indicator, you need to understand how the markets are moving (trending vs ranging, etc) and also have some basic understanding of what is driving the markets short term.

EDIT UPDATE: I hope I'm not sounding negative here, as I'm really a fan of technical analysis and Wilder's systems, I just wanted to be realistic as there is a lot of garbage out on the internet where you "buy my software and you will make BAZILLIONS in forex trading without doing anything". So even with a great system, *YOU* still factor into the trading process as to whether or not to deploy the system (even if its automated, you have to turn it off and on) as well as greed, fear and many other emotions when there is real $$ on the table.

[dpaterso]

Good (Wednesday) morning everyone.

STOP RAINING ON MY PARADE CRAIG!!!

(I am definitely only kidding with you)!!!

You are absolutely 100% correct. The trades that I've just done (and closed out manually last night and I'll tell you why later on in this post) worked out so well because prices were trending fantastcally no question. Which, unfortuanately, does bring us back to one of our 'age old SIS problems': how to correctly and TIMEOUSLY detect the presence or absence of a trend!!! While I've been watching these trades move I have indeed 'slapped' onto the same charts all the indicators that are SUPPOSEDLY capable of indicating either the presence of a range or the presence of a trend and alll of them appear to be pretty useless. They all lag FAR too much and of course: on the shorter timeframes, especially with volatility, they lag even more. I will spend some time today seeing what I can come up with because I certainly don't want this 'bang' to 'fizzle out'. I'd also agree that the SIS may possibly be suited to certain instruments / pairs and not others. Could this be because there are certain instruments / pairs that spend more time trending that others??? I believe also that you could not use the SIS on shorter timeframes (shorter than the daily or possibly the four hour) with instruments / pairs that have big spreads. I think we've got 'a ways to go' here yet!!!

Anyway: I closed out all my positions manually last night at 23h00 or thereabouts for a nett gain ALL CLOSED positions of around 1641 pips. That added to the previous days gain of 732 equates to 2373 pips over a period of 48 hours or so. And therein lies the problem!!! It is just impossible to trade like that so, while those gains are indeed fantastic, they're unrealistic simply because no person on earth (well not any one that I personally know of) can trade continuously for 48 hours straight EVEN IF you attempt to sleep 'between hours'. I mean: I feel like a bit of a 'zombie' this morning and I can tell you that already I'm having a problem concentrating (no matter HOW many Red Bull's I've had this morning)!!! The fact of the matter is that it's no good making a gazillion pips in two days and then spending the rest of the week losing them because you're having a hard time concentrating. That's pointless. Anyway: I've 'slowed down' a bit today. I'm in one single trade (GBP/BGN) and I'll calmly stick with this pair for the day. I will attempt to address our 'ranging / trending issue' as well.

Thanks to all for the good words and support and input. It is really appreciated!!!

Regards,

Dale.

[midulster]

Morning all,

Just wanted to say hello - I've still wimped out from trading the daily markets and hourly trading isnt an option for me. But I do still read here religiously every day and its great to see you folks are still working on and enjoying some stellar results - I'm absolutely goggle eyed when I see the pips you're getting Dale.

Hopefully the good work being done on the 'constant' will help with the dailies when/if things calm down again.

All the best

Derek

[dpaterso]

Hey 'Father Ted'!!!

Very nice to hear from you.

Congratulations ON YOUR RESTRAINT!!! That's 99% of 'the battle' won let me tell you!!!

I've been constantly checking all the timeframes and I assure you that this 'final fix' has sorted the SIS out once and for all. It looks just as good on the daily charts as on any other chart. The only (obvious) difference is the fact that you miss the instraday movements both up and down if you're trading daily whereas I'm 'catching' those intraday movements on the 1 hour charts. But this 'fix' has defintely eliminated most of the 'whipsaws' from the system which is the one and only reason why I had doubts about the SIS i.e. every time I used it in the past I'd get stopped and reversed many time until I simply stopped stopping and reversing and just took profits when they were 'on the table'. Now I feel confident that I can follow the SIS trades to their logical conculsions each and every time without having ot 'panic' about losing some profits. Having said that: by it's nature you will lose some profit on a stop and reverse and sometimes quite a lot. At one point yesterday I was up over +2000 pips on those trades but, by the time I closed them out (at the time they were very close to their stop and reverse points anyway), I'd lost a lot of pips. Having said that though: it's the overall long term result that counts I think. As I said to someone yesterday: I'm happier to make a certain minimum amount of pips per day that I can count on than I am to make these wild profits one day and either nothing or lose money the next. I believe that now that we've sorted this thing out I'm at that point i.e. I am quite happy to simply trade the system 'mechanically' knowing full well that even if I'm 'up' by a 'gazillion pips' I am probably going to lose a good portion of them when I have to stop and reverse because that profit that I have at the end of the day is 'real' profit and something that I can count on.

DO keep in touch though. There are many changes in those indicators that I sent to you so we do need to go through them again. You know how to get a hold of me!!!

Regards,

Dale.