Trading Chaos

[n5032245]

Sinner,

I will try to summarise what you have said in this post to better understand you.

You trade BW Chaos model

1. during the London open to NY open
2. combining it with Elliot wave, resistance lines and fib retracements

You also sometimes layer the entries, but have held back on going into those, to keep it simple.

A couple of questions.

1. Do you still trade those same times? i have noticed as the thread went on that you started to trade the asian time period as well. can you give some more info on this?
2. you do not use fib, EW, etc on every trade. what makes you combine the two on certain BW signals, but not others where they would apply?
3. you said at the start of the thread that you set up the order in the evening and walk away. do you set a time expiry on it? if you just set it up, and another similar directional BW signal was released, you would not have the correct entry settings (or if the MA's crossed and the gators mouth was closed). can you help us understand how you do this?
4. the trailing stop of the green/red MA. do you always use one or the other? how do you set this up in metatrader4? i can see the trailing stops option, but can't set it to the MA 5 period.
5. do you ever layer with multiple fractals, AO signals, etc? what makes you use layering in comparion to not using it?

(i hope these questions are a little more constuctive then the last)

thanks.

 

[sinner]

Sinner,

I will try to summarise what you have said in this post to better understand you.

You trade BW Chaos model

1. during the London open to NY open
2. combining it with Elliot wave, resistance lines and fib retracements

You also sometimes layer the entries, but have held back on going into those, to keep it simple.

A couple of questions.

1. Do you still trade those same times? i have noticed as the thread went on that you started to trade the asian time period as well. can you give some more info on this?

Hi Mate, when I started the thread I was "working" trading. For the last few weeks I have been on "holidays" from trading, so I have a bit more time to identify levels for breakouts during Tokyo. Sometimes this means catching a move a bit quicker, but GENERALLY speaking a level established during Tokyo won't get broken out of till Frankfurt or London anyway. So whatever you do in the morning isn't confirmed or even active until 7-8 hours later.

2. you do not use fib, EW, etc on every trade. what makes you combine the two on certain BW signals, but not others where they would apply?

I don't include everything I use on the charts all the time but everything I use on the charts is included here in one screenshot or another. This is simply to keep them as clutter free as possible. Once you know where a level is on the chart you don't really need the lines there as you can see the PA respecting those zones.

3. you said at the start of the thread that you set up the order in the evening and walk away. do you set a time expiry on it? if you just set it up, and another similar directional BW signal was released, you would not have the correct entry settings (or if the MA's crossed and the gators mouth was closed). can you help us understand how you do this?[

Great question bro. Generally my first trade of the day using this style is a breakout of the Tokyo range. If there is no valid fractal to represent this I will use a breakout of yesterdays hi or lo (or possibly the NY sessions EOD range) with the trend. Generally you are right, a better breakout signal will appear, but I still leave my order on at that area because a breakout of yesterdays high or low is generally more reliable for a few pips. I will include an example explaining what I mean in a sec.

4. the trailing stop of the green/red MA. do you always use one or the other? how do you set this up in metatrader4? i can see the trailing stops option, but can't set it to the MA 5 period.

Check the Alpari link on the first post. Which MA you use depends on the volatility of the pair (how you measure and decide is described in the link - stoplosses section). Sometimes if I am catching the first impulse from a correction and the original R:R is good I won't trail the stop at all and just leave the trade on after taking half profit at 1:1 R:R. I trail the stops manually, just by checking the chart every hour and moving it to the new value (sometimes I might use the MA of the bar before the one that just closed to give a bit of room or whatever).

Don't get too hung up on squiggly lines, Chaos style also has another method of trailing stops as described in the same link (in the same section) based on the statistics of how many bars a wave tends to last before retracing.

5. do you ever layer with multiple fractals, AO signals, etc? what makes you use layering in comparion to not using it?

(i hope these questions are a little more constuctive then the last)

thanks.

Generally I am aiming for a daily profit target if I don't think there is going to be a trend developing around my trade. So if I think a breakout can provide 20-30 pips on AUDUSD or EURJPY or whatever is moving, then I won't add on positions (why add on risk for a reliable signal?).
But if the wave is slow i.e. will only have a prospective 10 pip move to make 20-30 pips then can trade breakout of every bar when the signals are reliable. Conversely if the move is really big and I want to capitalise on it i.e. will have a 100 pip prospective move and want to make 200-300 from it.

So for example of range breakout to decide which fractal level I wish to sell from, last two days trading E/J (not including todays action):

 

[n5032245]

Sinner,

on the Alpari explanation of BW chaos model, it mentions that

Close all open positions if a signal in the opposite direction appears. Bullish divergence / bearish convergence between the Awesome Oscillator and the price signals that the trend is about to end:

does that mean that the first fractal buy signal that is shown in the attached file is invalidated by the following sell fractal signal? which would mean that the third fractal buy signal is entered?

 

[n5032245]

Sinner

I am better understanding how you manage your trades but have a couple more questions surrounding your

1. entry

Great question bro. Generally my first trade of the day using this style is a breakout of the Tokyo range. If there is no valid fractal to represent this I will use a breakout of yesterdays hi or lo (or possibly the NY sessions EOD range) with the trend. Generally you are right, a better breakout signal will appear, but I still leave my order on at that area because a breakout of yesterdays high or low is generally more reliable for a few pips. I will include an example explaining what I mean in a sec.

do you always look for confluence of a breakout with a fractal to initiate a trade? would you do this on the majority of times you enter a trade?

2. stop loss

Sometimes if I am catching the first impulse from a correction and the original R:R is good I won't trail the stop at all and just leave the trade on after taking half profit at 1:1 R:R. I trail the stops manually, just by checking the chart every hour and moving it to the new value (sometimes I might use the MA of the bar before the one that just closed to give a bit of room or whatever).

how do you calculate this R:R? money management seems to be what would make or break this system (as always)

from what i can understand, you combine your knowledge of EW, Fib, etc to set stops. more examples of thses would be appreciated.

3. targets/exits

can we get some more info on whether you set specific pip targets or follow the BW chaos model exit paramaters.

and how is the trading looking today and this evening.

i will have to start looking at trades and posting myself now that i am gaining a greater understanding.

 

[sinner]

In the case of your chart, only fractals in the direction of the gator are valid.

Sinner

I am better understanding how you manage your trades but have a couple more questions surrounding your

1. entry



do you always look for confluence of a breakout with a fractal to initiate a trade? would you do this on the majority of times you enter a trade?

This question, doesn't make sense. They are the same thing. Think of it like this:

Trading a "breakout" of the days high or low is another way of saying trading a breakout of yesterdays highest or lowest fractal. The high or low is by definition a fractal.

2. stop loss

how do you calculate this R:R? money management seems to be what would make or break this system (as always)

from what i can understand, you combine your knowledge of EW, Fib, etc to set stops. more examples of thses would be appreciated.

Go back and look at the EURGBP trades as an example for the 1:1 R:R trades. I will try to find examples and explain this q soon. Go and read the section on stoplosses on the alpari page like I said.

3. targets/exits

can we get some more info on whether you set specific pip targets or follow the BW chaos model exit paramaters.

It all depends. I've provided examples in this thread of both the styles you mentioned (for example see the blue box EURUSD chart exiting with 3 AO bars against you). I'm terrible at holding positions so generally if I see anything around 1:4 1:5 R:R I will just take it right off the table hehe - usually it goes further after I get out. Targets/exits all depend on YOU I reckon. Make a plan and stick to it.

and how is the trading looking today and this evening.

Well...um...if you can't spot the giant bear paw on the chart for the last few days...might still need some practice?

 

[sinner]

Posted this on forexfactory too, apologies for the double post.

December is supposed to be the "low volume hard to trade" month, right? Right? So how come I pretty much cleaned up across a lot of pairs trading the daily compared to November?

Total spread cost: 10 pips
Total carry cost: 14 days short - 6 days long
Total profit cashed in: +~350 pips short
Floating P/L: +~160 pips long
Monthly ATR: ~600 pips

 

[sinner]

Still holding December longs in this pair...well into the money now, getting some good cash from the carry too.

 

[sinner]

Sometimes I just can't be bothered holding or when trading chop like E/J has been for the last 15 days (I am also swing short from 135.691) I will just cash it right in.

Cash in E/J swing short here for +1000 pips.

Any questions?

 

[sinner]

Hi guys,

I have written a rudimentary Metatrader 4 EA for some backtesting results. It trades only on each bar close using only stop orders to enter the market and limit orders to exit. It just trades 0.1 lots with no MM right now.

Include a screenshot of the EA in action (using visual backtest mode) as well as backtest results of EURUSD H4 from 2008.05.26 to current date. I coded a few different methods to trail stops and currently it is trailing stops using the "alligator jaw". There are more "profitable" stop techniques available for use within the system but I haven't coded them yet and wanted to see what the results would be like trying to stay in each trade as long as possible. Full EA source code attached, comment or uncomment various bits to disable or enable features for now.

I have diverted some of my profits from the recent GBP freefall to trade the ASX20 using this system on the daily chart and would like to continue documenting my thoughts on this thread. To that end I have imported 2 years of daily ASX20 data into Ninjatrader 6.5 and installed the Chaos indicators. Is it possible to get this thread moved out of the forex section into the systems section?

If so I will begin posting trading ideas later this evening.

 

[professor_frink]

Is it possible to get this thread moved out of the forex section into the systems section?

done

 

[sinner]

Thanks frinky.

Alright, first up I apologise as my AMP and WPL data seems broken somehow (or at least the import into NT6.5 doesn't work good), so I can't look at those charts right now.

Coming into the new week, I am interested in three stocks:

1. BXB
Fridays daily bar closed as an IB and we got a valid Accelerator Oscillator signal to place a buy stop above Fridays high.


2. CBA
This is a bit of a cheat trade, as I am just getting started trading the ASX20 using this technique I missed last weeks signals. However, Fridays low close might give us an opportunity to enter at a slightly cheaper price.


3. WDC
The 3rd of March high has formed a "buy fractal" as of the Friday close. Buy stops above this high could provide a good first entry into a new uptrend.


Plan for Monday 8th March 2010:
Buy stops
BXB
Entry: 7.34
Stoploss: 6.80 (below Monthly Pivot Point at Feb 26th low)
Signal: AC 2 green bars > 0 + inside bar above MPP
Targets: none

CBA
Entry: 55.35
Stoploss: 52.98 (below last swing low and MPP)
Signal: AO 0 cross + AC 2 green bars > 0
Targets: none

WDC
Entry: 12.36
Stoploss: 11.97 (below potential swing low forming)
Signal: buy fractal above gator jaw
Targets: none

Sell stops:
None.

Fills:
None.

Exits:
None.

Order Adjustments:
None.

 

[sinner]

Hi guys,

I apologise for not writing a trading plan for today (9th March) last night, as I was extremely ill last night.

Today I am still not feeling so great but mustered the strength to import yesterdays daily candles to Ninjatrader and look through the ASX20.

The only chart of interest for me today (thankfully, I don't have the strength right now for anymore) is TLS. We have developed a strong sell fractal at the March 2 low.


Plan for Tuesday 9th March 2010:
Buy stops:
None.

Sell stops:
TLS
Entry: 2.8
Stoploss: 3.11 (MPP)
Signal: sell fractal
Targets: none

Fills:
BXB long: 7.34 (no slippage)
CBA long: 55.6 (terrible slippage of almost 30c into the open yesterday)
WDC long: 12.38 (2c slippage)

Exits:
None.

Order Adjustments:
None.

 

[sinner]

Morning guys,

Plenty of interesting developments in yesterdays trading:

ANZ developed a strong "wiseman 1" reversal candle on the 8th followed by an inside bar.


CSL posted a strong "wiseman 1" reversal candle


Plan for Wednesday 10th March 2010:
Buy stops:
None.

Sell stops:
TLS
Entry: 2.8
Stoploss: 3.11 (MPP)
Signal: sell fractal
Targets: none

ANZ
Entry: 23.64
Stoploss: 24
Signal: wiseman 1 "bearish divergent bar" + inside bar
Targets: Evaluate at the MPP (22.44)

CSL
Entry: 35.74
Stoploss: 36.76
Signal: very strong wiseman 1 "bearish divergent bar"
Targets: Evaluate at the MPP (33.54)

Fills:
None

Exits:
None.

Order Adjustments:
Possibly move BXB stoploss up closer to the $7 mark at EOD or tomorrows open after yesterdays strong performance. If unsure will just leave at current price.

EDIT:

Two notes.

1. It is worth pointing out that I have done no backtesting or anything as a decision point to trade the ASX20 using this method. I have simply diverted some of my recent trading profits into testing this market. So it is just as much an experiment for me, but I am taking it seriously.

2. Can anyone please provide a poor soul with an odd years worth of AMP and WPL EOD data? I import mine from Comsec currently, but NT doesn't seem to like those two stocks. I am not so much concerned about AMP because the index is already overweight financials, but I would like some exposure to oil.

 

[sinner]

Another extremely interesting day on the market. CBA attempted a challenge at the 56 level and failed it seems. BXB made a new high for this trend also but posted a bearish/neutral close. ANZ posted a new high as well so I will be removing my sell stop for it tonight. WDC had a rather bearish day and came close to stopping me out, a shame if that happens because this trade looked really good to me on Sunday night.

Of interest to me today is the NCM chart. We have posted a clean "higher high" buy fractal from March 8 highs after strong test of support. If I was secretly harboring hopes of a hyperinflationary collapse of the economy, surely NCM would be the stock to hold :



Plan for Thursday 11th March 2010:
Buy stops:
NCM
Entry: 34.64
Stoploss: 32.12
Signal: buy fractal
Targets: Some gap filling around MR2 and MR3 would be nice. Will evaluate if we get there.

Sell stops:
TLS
Entry: 2.8
Stoploss: 3.11 (MPP)
Signal: sell fractal
Targets: none

Fills:
CSL short: 35.74 (no slippage)

Exits:
None.

Order Adjustments:
Remove ANZ sell stop (higher high posted, wiseman signal invalid).
BXB long stoploss moved up to $7 (might adjust this down a few cents at tomorrows EOD, $6.97 I guess).
CBA long stoploss moved up to $52.99

 

[Lone Wolf]

G'day Sinner,

Just looking at the backtest results you did for this system on the H4 EURUSD - The equity curve looks nice overall, but I notice that there can be long periods of loss. The worst of which was from about March 09 through to December 09 where the system slowly lost money the whole time. That's steady losses for 3/4 of a year. How do you feel about that? Was it because the EA was incomplete? Do you expect your live trading to have the same issue? Were you able to identify the market conditions in which the system performs poorly? More importantly, can you do anything about it?

Looking at the charts during that time, price looks to have a fair bit of chop with an upward bias. But if you stop trading when it looks like sideways action you might miss the next big move.

I'm not saying the system is a bad one, when it does make money it seems to make it quickly. I was just curious as to what your tolerance level is. How long is too long for a losing streak? You said the ASX20 trading is an experiment. Considering the EURUSD backtest shows long loss periods, how long would you allow the ASX20 trading to lose before you decided the system isn't working?

I enjoy reading your posts by the way, thanks for sharing.

 

[sinner]

G'day Sinner,

Just looking at the backtest results you did for this system on the H4 EURUSD - The equity curve looks nice overall, but I notice that there can be long periods of loss. The worst of which was from about March 09 through to December 09 where the system slowly lost money the whole time. That's steady losses for 3/4 of a year. How do you feel about that? Was it because the EA was incomplete? Do you expect your live trading to have the same issue? Were you able to identify the market conditions in which the system performs poorly? More importantly, can you do anything about it?

Looking at the charts during that time, price looks to have a fair bit of chop with an upward bias. But if you stop trading when it looks like sideways action you might miss the next big move.

I'm not saying the system is a bad one, when it does make money it seems to make it quickly. I was just curious as to what your tolerance level is. How long is too long for a losing streak? You said the ASX20 trading is an experiment. Considering the EURUSD backtest shows long loss periods, how long would you allow the ASX20 trading to lose before you decided the system isn't working?

I enjoy reading your posts by the way, thanks for sharing.

Hey mate.

Great questions!

The Chaos system is (as I said in the first post of this thread) essentially just an accumulation of a few well known trading ideas like Dr. Elders Triple Screen Trading techniques, fractal breakouts, Elliott Wave momentum shifts, Elder-Ray style trades using the AC oscillator, etc.

All of these techniques have one thing in common: they are all known to traders as LTTF methods. LTTF stands for "Long Term Trend Following". You may have heard the well worn trading phrase before "let your winners run and cut your losers short".

Well that is the goal of all LTTF systems. To provide a set of highly reliable signals which will almost guarantee you money during a strong trend (because all you are trying to do is hop in on the trend). Now, these same signals are probably utterly useless during a period of extended chop.

The EA provided, is just a rudimentary mechanical one, it has no discretion or money management (MM is known to smooth the equity curve considerably). But you can see it still manages to adhere to the phrase "let your winners run and cut your losers short". So we don't really care about the periods of extend chop, because we know that over a large enough set of trades we will eventually run into a trend which will pay off our losing streak and then some.

This is a really important point to understand: the market is very paradoxical. For any given trade the outcome from our perspective before we place said trade is essentially (although not 100% deterministically) random. But if the market were truly random, trends would almost never occur. Yet, trends occur all the time.

So for all LTTF trades, it is a truly essential part of our trading psychology keep this point in mind. We are trading the market structure. We wish to make money from higher highs and higher lows or lower highs and lower lows. That means standing first in line to hit the bid or offer when everyone else is fading the trend or already entered the trend 100 pips above you using some MA cross or something.

But the fact remains, we are trading the market structure secure in the knowledge that eventually a trend must eventuate from any breakout of consolidation. We will never be late for any trend that develops, we are always first in line. Nobody wants to buy the highest point, it even seems silly when you first think about it. But, think about it some more and you will realise 1 pip higher than the current high is the only confirmed place that a new trend might be developing.

I hope that sort of answers the fundamental question here but I will reply to each of your actual questions now:

That's steady losses for 3/4 of a year. How do you feel about that? Was it because the EA was incomplete? Do you expect your live trading to have the same issue? Were you able to identify the market conditions in which the system performs poorly? More importantly, can you do anything about it?

1. Steady losses after 20% gains, I feel fine with that. I programmed the EA to take what the market structure provides and it did exactly that. For 3/4 of the year, the market was in unfavorable conditions for LTTF trades but the EA only lost ~3000 of the ~9000 it had made. Pretty good work if you ask me!
2. EA is not incomplete, it was just a rudimentary program to run some backtests, there is no way I would trust an MT4 EA to do my job as a trader.
3. RE: live trading. One of Bill Williams core rules is "never trade when the alligator is sleeping", this is a discretionary rule that is hard to plug into the EA but very easy to apply visually when you open your charts. But this is the core rule to keep you out of chop trades so the EA is obviously trading in some cases where I wouldn't. There is no issue about "missing the good trade" as you said because we only ever start trading in the direction of the trend once a swing high or swing low has been established outside a consolidation. i.e. None of the good trades for us are in the consolidation anyway. Please see the Alpari link at the beginning of this thread for more info on using the alligator to time your trades.
4. See above.
5. Yes, you can do something about it: stand aside. I show many live examples in this trade of standing aside until the trend is confirmed.

I was just curious as to what your tolerance level is. How long is too long for a losing streak?

Seriously, great question.

It is worth looking at this PDF for the answer (think I linked it previously in this thread too).
View attachment Trading%20for%20Infinite%20yield--%20Utility%20theory%20and%20you.pdf

You said the ASX20 trading is an experiment. Considering the EURUSD backtest shows long loss periods, how long would you allow the ASX20 trading to lose before you decided the system isn't working?

Well, EURUSD backtest is a single trading instrument. Trading the ASX20 is very different because you are essentially taking on 20 instruments which are not necessarily correlated (although often can be during periods of extreme market strengh or weakness). This actually allows a much better risk profile than what can be gathered from trading a single instrument. As you can see I am holding both longs and shorts against different components of the index, whereas if I was trading just the index itself I would probably have to be long everything. e.g. tomorrow if I wake up to find the SP500 tanked 50 points overnight I might get stopped out of my long trades but the short ones will remain happily intact.

It might be worth looking at this PDF to answer some of your questions in that regard. Trading LTTF it is very often important to spread your trades out across a variety of instruments to achieve a smoother equity curve.
View attachment turtlerules.pdf

I honestly hope this helps and I wasn't just rambling for the last 30 minutes as I typed this post. Please get back to me if you feel unsatisfied, I will try to respond again.

 

[tech/a]

I maybe wrong but I think your going to find it difficult to make a profit without a bias to either the long or short side of the market.

Interesting thread.

 

[lukeaye]

Hey mate.

As you can see I am holding both longs and shorts against different components of the index, whereas if I was trading just the index itself I would probably have to be long everything. e.g. tomorrow if I wake up to find the SP500 tanked 50 points overnight I might get stopped out of my long trades but the short ones will remain happily intact.

Are you sure about that? You have picked two stocks, which traditionally when the market tanks, do not follow suit. I wouldnt short defensive stocks in a hope that when the market tanks they will cover your longs, because those longs will go down alot harder and alot faster then CSL or TLS.

You might want to short stocks which more closesly correlated with the index. TLS and CSL are prob the worst two i can think of.

But good luck anyhow i know how much you value my opinion.

 

[lukeaye]

I maybe wrong but I think your going to find it difficult to make a profit without a bias to either the long or short side of the market.

Interesting thread.

I havent come up with a way either.

Sinner have you had success with a strategy using a long and short bias?

 

[sinner]

I maybe wrong but I think your going to find it difficult to make a profit without a bias to either the long or short side of the market.

Interesting thread.

Lovely post tech, just come in here, spout some statement without any thoughts or reasoning to back it up, gee thanks for the insight. How come you aren't this helpful all the time?

Since you provided absolutely no thoughts at all, I am wondering why you believe I should bias my trading by believing the NCM share price is somehow related to the TLS share price is somehow related to the FGL share price?

As I already stated in the above post, during periods of extreme market strength or weakness share prices across sectors are often strongly correlated. So what? I don't trade a bias, I trade the chart. When all the sectors are tanking or uptrending at once it'll show up on the charts. I won't need a bias, I will be following the same plan then as I am today.

Until then, I will be happy to ignore the index, hold longs on CBA, shorts on CSL and whatever other setups my charts not my bias indicate.