Zigzag Bollinger Band

[snailbeard]

The tests were done with sourcecode changed to meet MetatCrap build 600 requirements.

However, I get errors when I try to use the v600 CrappyTester.

I had to build an (509)ex4 of the v600 compatible sourcecode using MetaCrap version 509
and do the backtest on CrappyTester build 509.

It would interesting to know if a significant number of coders have given up with MetaCrap...

[SpiderX]

The tests were done with sourcecode changed to meet MetatCrap build 600 requirements.

However, I get errors when I try to use the v600 CrappyTester.

I had to build an (509)ex4 of the v600 compatible sourcecode using MetaCrap version 509
and do the backtest on CrappyTester build 509.

It would interesting to know if a significant number of coders have given up with MetaCrap...

Hi snailbeard,

Check out:
http://www.stevehopwoodforex.com/phpBB3 ... =15&t=3439

Feel free to add on to the list.

Cheers

[snailbeard]

The tests were done with sourcecode changed to meet MetatCrap build 600 requirements.

However, I get errors when I try to use the v600 CrappyTester.

I had to build an (509)ex4 of the v600 compatible sourcecode using MetaCrap version 509
and do the backtest on CrappyTester build 509.

It would interesting to know if a significant number of coders have given up with MetaCrap...

Hi snailbeard,

Check out:
http://www.stevehopwoodforex.com/phpBB3 ... =15&t=3439

Feel free to add on to the list.

Cheers

SpiderX, it looks like I have posted in the general v600 topic
http://www.stevehopwoodforex.com/phpBB3 ... &start=100
instead of the coders specific topic
oops

[snailbeard]

I found a few flaws in the pullback logic but after addressing these and rerunning the tests for longer (1 month), I was in a mood to celebrate despite all our MetaCrap issues.

There seem be about a factor of 10 more trades per month than with ZZBB alone.

[snailbeard]

Last week turned out to be a very productive and interesting week. I didn't have to work around any new MetaCrap issues (for the time being) although there are possibly thousands of implied conversions of integer to string which will need to be changed at some point

I have been experimentiing with several variations on the theme of reentry into the current swing in the context of 1 to 3 day swings.

All these entry methods have flaws, so I am thinking about transfering the methods into a single pair daytrader with a small number of easy to change parameters. Part way though the morning I could see if one of these methods should be selected and then an EA left to automaticaly trade one particular method, so that I don't have to hang around for confirmation.

Most entry methods continue to trade the wrong way after an intraday reversal. It is often difficult to spot the difference between a pullback and a reversal. In the attached chart I challenge the reader to define a set of rules which would define this as a pullback for rentry or not, but in general terms not specific to this one example:

[snailbeard]

Following on from the previous post lets move along the timeline and see how the market often lulls us into an obvious pattern and then does the complete opposite:

AUDUSD-H1-one-day-reversal-25July12.png

If I had glanced at the first chart and not studied it carefully I could easily have dived in head first only to regret it.
However, if I had taken my time and studied the chart carefully I could say that the 25th presented only higher highs and higher lows. So if we mask out the previous days history, all we would see is higher-highs and higher-lows (no hint of going down). However, it would be unsafe to ignore what happened in the recent past, so at what point would you say the price is nolonger a pullback and we can trade it in the opposite direction?

There are several things we can consider:
1. Where is the price now in relation to the begining of the day?
2. Is the price higher or lower than previous day's extreme?
3. What clues are in the candlestick patterns?

[snailbeard]

In this chart we have a continuation of a general trend.
The blue lines indicate and upward bias.
The red line highlights a series of lower lows, but then never goes below the recent low, so we can label it a higher low, then price moves up again past the open price.
This pattern is very common in steady trends and can be used to confirm a buy trade or reentry into the main trend.

[snailbeard]

There are times when applying this pullback-reentry logic can work really well and produce excellent trades.
Here is an example of great returns on this method when the market is moving with reasonable volatility:

[snailbeard]

Running the pullback-reentry method (PBRM) over 12 months gives a better idea of the long term reliability of the method. It is the complete antithisis of the zigzag-Bollinger approach. ZZBB trades infrequently, gets the vast majority of trades correct but makes a modest growth, whereas, PBRM often trades several times a day,
can suck up profits for several months at a time, but it can also eat into your capital for three months, which would convince most users to switch it off permanently. PRMB is quite bipolar if used raw but can be fantastic if used correctly. Could we make it ZZBB's sidekick?
Perhaps there are other things we could look at to change its parameters which could be adapted to volatility.

[snailbeard]

Looking at some of the failed PBR trades might help us improve the ratio of wins to losses.
In the following chart we have a collection of confusing lines.
With the benefit of hindsight I can tell you that it is possible to buy or sell and come out with a profit, but only if you choose an appropriate target and stoploss.
So you know the entry point. If I tell you that you can gain 30 pips either way then there is only one parameter left to adjust: stoploss - depending on the spread - we need a stoploss of about 40 pips.
However, having stoploss permanently bigger than the take profit is rarely beneficial, so a more analytical solution is preferable.
We can easily identify several channels and a triangular contraction.
There are more upward channels than downward channels.
Also if we classify the channels by time
We have
4 day channel up
2 day channel down
1 day channel up
We also have a 3 day triangular contraction
We also have price closer to the bottom channels.
According the M15 and H1 (pre-repaint) zig-zags: price is moving away from the low to the top of the channel.
If we take all this into consideration then buy is favoured over sell
or the safe option is to do nothing.