Bollinger Band-it trading system

[wicha]

Hi,

I would like to introduce my trading system which concept is based on the waves' movement and their reaction when they meet with obstacles. I know it sounds a bit esoterical, but as far as i observed, the price (and the mass human perception that lies behind price change) flows like a small creek that has swollen after a rainfall. Mostly it flows to the smallest resistance, but can break through larger obstacles when in the tension is high. Just imagine the picture and you will understand it.

Every movement has its time period. The smallest is about half day that stretches to app. 2 days (can be seen in 4H TF), a larger one is about 1-1.5 week, and the longer ones are about one month. The beauty in trading (and the difficulty) is that every trader has an individual concept of time scale and they react according to it by different amount of power they have (hobby traders vs hedge fund managers) and these powers take effect on the price in the same time, so you have to understand all of the participants view in order to trade well.

What I use:
My only TF is 4H. Since the system includes many time scales from 2 days to half year, i don't need to watch anything else just this TF.

Compulsory theories:
- SMA theories, how they react to each other, slingshot movements etc.
- Basic Bollinger band theories
- and of course how to use Currency Slope Strength (thanks for the indi, i really like it) (step 0 to choose the right chart)

Default chart needs
-SMAs, 15 - 2.5 days, 62 - 2 weeks, 186 - 6 week , 840 - half year
-Bollinger Bands (later simply BB) SD 1 (no color)
- the smaller is 62 with +/- 0.66 (you can set it in levels) - 50%
- the larger is 186 with +/- 0.525 and 1.645 - 40% and 90%

% means that the price stays between the lines by the percentage of time. the smaller segment is usually the non trading or breathing zone, no clear trend. if you know soccer it is tipically the midfield. The price has to get out from the zoe in order to start a trending move. The second (90%) is the non strict border of the time frame. Beyond this line the price is driven by a larger TF, since 90% of the time the price simply isn't here. There is an event that is called Fat Tail, that is happening when the price reaches to the "out of space". There are no orders on one side, so the price goes till every order is filled. A good example of Fat Tail is when the CHF has untied from the EUR on price 1.2 and there were no orders below 1.175 only SL-s so the price went down to 1.05 to fulifill every SL. In these cases the end of the movement is totally unpredictable, you shouldn't trade it. Sometimes you can, but it's not worth on the long run.

- RSI 14
- two SMA-s on RSI line: 6 and 24



and a sample picture of it:

GBPCAD.png

As you can see on one side its a bit complex system, on the other hand everything is on a picture and easy to use. You dont need to use S/R, because it is on the graph. You know instantly where the price will go (or likely where should go) and where to put your trades and TP-s. The extra you should use is one or two obvious trendline, nothing else.

The system and its precise entry exit rules are under construction, but there are some rules of thumb:

- 1. continuation pattern. the price is out of the BB line makes a curve on it (small red BB). It is inside the black BB, it acts as a support too. On RSI the SMA-s are over 50 (bullish) and in order. Gathers strength to jump.

- 3. Continuation pattern. In addition there is another common behaviour, when the price test the far side BB line (small red) from inside and turns back. It acts as a rubber band, fastens the movement. On RSI there is a Head and Shoulders shape, after lines being in order, it goes to south.

- 4. Continuation pattern, almost similar to 1., but since it is outside the black BB line, if it is not a fake move, it has to be deeper.

- 2. a reversal. maybe the grey circle is not at the right place, the real moment is exactly at where the black line is. the price tests back to the upper BB line and right after it goes under the Brown middle SMA (center of black BB) bullish trend here is clearly over, the price is at least ranging. in Elliott waves i think it is the 1st wave or an A.

- 5 reversal. the only thing you know it's a reversal that the RSI shows you a clear sign.



What are SMA-s and BB-s means or reflects?

- Purple SMA (15) is an intraday SMA. if today the sentiment is bullish, the price is over on it. if not, same in inverse.

- Gold SMA and red BB (62) - it shows the smaller waves' neutral zone. it covers 2 weeks time, by it you can see forward 1-2 days

- Brown SMA and black BB (186) - the basic frame. It cover 6 weeks, by it you see the main macromovements, gives signals to trade longer trades, entries, profit taking places, etc.

- Green SMA (840) - half year time, i've put it to know if the graph is bullish or bearish, no other special meaning. The difference between price and SMA is too big to get precise datas from it.

Basics.
- more the Sma-s are in order the better.
- price moves between BB lines like a pinball ball. like lower BB line -> upper BB -> SMA -> Upper BB -> up high
- after BB squeeze (where the downside movement started) the price should go strongly.
- price is outside an outer BB line, goes inside, tries to get back but cant, reversal (after 5. point)

please comment it if you have questions or you have some tips to it. I have been working on this concept for years, but it's not finished yet. I would be glad for any comments

Thanks

Wicha

[wicha]

yesterday my forecast on GBPCAD was a mistake, i try to analyse how i made a wrong conclusion.

GBPCAD_2.JPG

mistakes

#1 fundamental news: the british parliament voted and accepted the Brexit, that caused weakening in GBP. on a day before the GBP strengthened, i think because of hope that the UK stays in the EU, but not.

#2 GBP was returned fromthe +0.2 on slopestrenght downside, while CAD remained under on -0.1. i hoped i will catch the beginning of a trend, but it remained in the range zone.

#3 the two smaller SMAs show downward shift. to be upward, the price had to go over the red BB line.

#4 the price stayed below the upper black BB line, so it was still in range mode

#5 on RSI, the long green SMA went down to the 50 line. in order to be positive the price had to go up right after a bounce from that green line.

#6 Trendlines on price chart told that the price couldn't break the resistance.


now i wanted to write the price would have the possibility to go up, but the GBP has fallen since then. maybe the lower black BB will keep on 1.625. as i see now it is a retest to south and then north.
the GBP has values on slopestrength on H4 -0.23 to south , while on D1 +0.35 .
CAD on H4 is +0.04, on D1 -0.1.

the purpose of this analysis was to show the working of the system, not to give you the best trade of the day.

[wicha]

Today I want to talk about the usual behaviour of the Bollinger Bands.

Band forms.JPG

(I made a tiny change at the SD of outer bands that is why the picture is different according to the previous ones, i will explain it, but now it's indifferent)

- The direction and the steepness of the SMA gives you an overall picture of the strength of the trend.The more steeper the line, the stronger the trend. If the SMA is horizontal, the price is ranging.


1. The Squeeze. When squeeze happens, the price ranges in a small interval, bands are narrow. After the price breaks out from its narrow range zone, usally a strong move will happen. Sometimes the first breakout is a false one. You will know it because right after the false breakout the price "jumps" back right to the other other side of the field, but these false breakouts happens mostly on lower TF-s possibly because of sudden flow of large sum of trades. On higher TF-s I didn't find an example to show you.

In this example (No. 1 on picture) you see that however the bands are narrow, they are not exactly horizontal. That is because of the price tested the lower bands, no test to the upper ones except of a small try. It even didn't reach the outer band, and right after the second peak the price fell below the SMA and then the lower inside band which indicates downtrend.

2. Breakout. Price is outside of every bands; it can do it easily since the bands are narrow. During breakout, the bands are widening because of sudden move into "unknown territory"

3. During Trend phase the width of the bands remains the same. No more, no less. There is no strict rule about where the price actually is compared to the outer Bands. These bands can be both support or resistance, even median to the candles, but in most cases the price somehow relates to it.

4. On Pullbacks (Counter Trends) the bands gets narrower. That is because the price makes no new highs/lows, is clearly between the greater band zone. First clue: the new low is relatively closer to the SMA than before, even if it is a lower low compared to the previous one. As a rule of thumb, last one was outside the bands, the new one is inside the bands. there is a pullback if it pops back from the inside band, counter if it crosses the SMA and pops back from the opposite inside band. After long trends the price's main goal is this band (like in this case too) in order to diminish the long-standing status quo. Right after it, the next common goal is other inside band (like here) that is mostly horizontal line and acts as a clear pivot line (mostly this line is also a fibo, pivot, etc line).

Here is a good example of Ranging.

Band-range.JPG

As you see the bands are almost horizontal, but not as narrow as in case of squeeze. This lets the price space to pop between the bands, commonly outer bands. The moves are almost symmetrical. I drawed a head-and-shoulders shape that can be easily noticeable on screen as a result of popping between the BB lines. Also there is a deformed one before the first black vertical line.

By examining the sequence of intacted BBs, you can have a good hunch of what would be next.

[wicha]

Setuping the Bands.

The bands consist of 2 elements: the SMA that defines the extent of the sample size, and the standard deviation that defines which percentage of the candles should be inside in the Bands. However this % is not accurate, you can get a good overall picture of how common where the price actually is. On Bollinger's official site he states, that the SD=2 means 95% of the candle close prices should be inside the band, actually only 90% are inside. I think it has 2 main reason. 1.) SMA= 20 is low. 2.) since it is a widely used setup, the robots or traders are prepaired for this behaviour and it affects the indi's performance.

I have tested many setups over the years and I think the most stable ones are that can be divided by 60, such as 60,120,180, 240. My main problem with 20 is that you don't see the larger picture of the driving forces, since the price pops so many times from one band to another. On the next pic i show you what i mean (SD always 1 and 2)

Band-sma_variations.JPG

Every SMA has its strength and weakness,that is why it is so difficult to find the most commonly useful ones. The same is true for the SD values. The best way to exploit their potential if we use different ones altogether.

Here is a table where you can find the SD values and %s that belong to them

z-score-02.png

My main purpose is to find those setups where I can divide the chart into segments of range (neutral), trend, and overbought/sold zones. If i know these points i will know where the environment changes, where the mass perception (traders, investors, robots, etc) thinks its time to do something. These are the real entry/exit points. By using the standard setup I get none of them, because it defines nothing except short term entry/exit points. Ok it is enough or not? Well to me not. I don't know where the price actually is: is it a counter , pullback or range. By a widened sample size i see a more complex picture.

When I defined the basic SMA I chosed 186 because somehow this was the ideal choice. Maybe larger investors' mass perception changes in 6 weeks or something, I don't know. Anyway, 180 is large enough to see the big picture, but rather flexible to act upon the new environment. I chose 62 (2 weeks) by the same reason.

When I started to modify SD values I've tried to find those lines where most of the turning points/ trend starting points are. These numbers are translateable into well known percentages because of laws of nature.

The best SD level of range (neutral zone) seems to me now is +/- 0.525 (40% coverage). Most of the trends starts from these lines (+/-).

To define an outer SD line however is harder than it seems to first because there is a thing i mentioned before, the Fat Tail. The farther the price is from the SMA line, the larger the randomness it stops. At first I've chosen +/- 1.645 (90%), but I had to admit that this line doesn't give me the precision i need. The price hits this line frequently, but more it stops before or runs through it. Then I've changed this number into +/- 1.383 (83.3% or to understand better 5/6) and it became more precise. The weaker trends turned back from here and the real huge ones ended at this point. I wanted to keep 90% and I added the most common one, the 95% (1.96). Outside of this region there is nothing that you can predict. This region is too far from the known territory (SMA). This is the place of greed and panic.

Here is the new one if you want to use it

BBand-It default.tpl

[wicha]

Back to the chart of GBP/CAD.

First lets see slopestrenght:

GBPCAD-SS.JPG

I drew trendlines on GBP just to see if i can read something from it. GBP has a LL on 01.10, a HH 01.19 and both the upward and downward movements are like waves. It's interesting that on 02.02 the "price" couldn't get above the last support trendline and a few candles later from now it has to try again. The downward (resistance) trendline is also in the way of the strengthening, but if it can do it, it would be easier to ride the support. I'm just experimenting now, but maybe it can work; we will see it a couple of days later. CAD is in dive, so GBP CAD has a strong upward potential.

The price itself:

GBPCAD.JPG

I wrote on Friday: "maybe the lower black BB will keep on 1.625. as i see now it is a retest to south and then north. "

As you see, the 1.624 kept the price. What was interesting , the conflict of forces. Lower BB said it is a downtrend ( the price was under of it), but the daily (so to say) pivot says its an overall positive day. The basic (black) BB said that the Price is ranging, from down line to possibly up. It was right below the SMA and tried to break it. It did and now is over the upper inward BB line

The most important clue what would happen was on RSI (I circled that). The price had stuck between the 2 trendlines, and had a strong driving force to up because it could get above all of the SMAs and the support line. It would be easier to break the resistance and go a long ride up (as it did). I call the circled area as Slingshot or Bungee. It has to gather strength to do the move.

Now the price is in nowhereon bias of north. Firstly it is out of the range zone, but in the triangle. It has to defeat the resistance, but after that it has the potential to go up at least to the black BB, but 95% is more possible (now 1.675).

A more simple approach: it tested the 1.625, now is under 1.65. if he goes up, Tp is 1.675 (realistically 1.6725)

[wicha]

A short glimpse to GBPCAD:

GBPCAD.JPG

nothing special, it gathers strength. It could stay above the upper range border BB line, as well as the 2 month old support. Since the resistance and the value of 1.65 are still above of it, now is in the phase of recuperation. it is a common behaviour in the market.

Good entry point was at 1.6439 up. Here is both the major lines, if the theory is correct (to 1.675), these lines should keep. Every indicator says it is an uptrend, except the smaller SMA-s. tomorrow the purple one will be over to the orange and a new uptrend can form.

[wicha]

I'm still searching for the necessary tools to understand both price and CSS behaviours. Here is what I've found.

At first let's see the price chart of GBPCAD

GBPCAD.JPG

The price unfrortunately turned back from 1.648 to south. As I see it is ranging, however it had every chance to do the breakout. What could happen?

-It was crowded out from the triangle and went back to ranging zone on both BB-s. It signals the momentum is powerless
- When it went under the support of 1.6440 (blue "horizontal" line) on RSI it reached the 50 from above. 50 means on RSI is the bias of +/-. so right after it broke out from the supporting fibo fan on RSI, right after it tested the 3.81 fibo line. It was just a couple of minutes, on 8:30 EDT today something happened. maybe fundamental news or a large sum of exchange I don't know but now it seems that it was not an earth shaking thing, because this spike is almost disappeared. The RSI is now at the 50 fibo, what is a good place to jump north till the increasing fibo 61.8.
- Also the price is still in both of the BBs, despite the false breakout to south. It is even still in the smaller BB and that is a big thing!

-So it is a range. I believe in north, but it needs time as I see...

CSS:

GBPCAD-SS.JPG

I drawed two important lines.

1. On GBP the cause of stop in increasing was the increasing trendline. Simply the currency couldn't get across to it, instead it turned back to south.

2. A fibo fan on CAD. Just before the GBP has met with its resistance line, on CAD the fibo 38.1 has just kept its currency from falling, it gathered strength and could easily go back to the positive territory.

By these two informations we see that the GBP CAD hasn't got too much chance to go to north, but still there is a chance to make it

+1 the CAD has a second fibo, where the price reached the last (61.8) fibo line. Since it was a long move, there is a chance that by here it would turn back. we will see.




What I don't understand, why the trendlines work on CSS because they clearly work. Both fans and lines. Is it because of 1.) the main chart has these lines (USD) or 2.) the larger investors watch a simple currency chart, or 3.) there is a mapping of the mass' thinking. If you know some explanation to it please tell me.

By the way: how do you like it? any observation?

cheers

[SteveHopwood]

By the way: how do you like it? any observation?

You over-complicate stuff. Your charts are a confusing mess of squiggly bits. Charts full of squiggly bits rarely confer useful information. It is unlikely that anybody has read your posts with sufficient concentration to work out what you are saying.

[wicha]

Thank you. I know it seems to be a big mess for anybody else and maybe i'm overcomplicating it. The reason i use so many stuff is because i want to see the intraday and long term picture on a single screen. However after a bit practising the bollinger bands are really useful and easy to understand indicators. I will try to rationalize them, thanks again

[wicha]

Thanks again Steve that you made me rethink everything about my system. You were right. On the one hand I found it very descriptive in backtsesting, but as you said there were too much information on it. It was really hard to get any meaningful forecast for the upcoming direction, since the tremendous datas showed at least 3 scenarios (up/down/stays as it is). So i had to reconsider everything.

As I said on my first post, the system I build is not ready yet. I build it because I want to trade on a live account from autumn, till that time I just want to implement everything into a well usable system and practising as papertrading. Years ago I've learnt that there isn't such a large sum of money that you can't lose in a month. From now on I just want to do it well, don't want to trade because of the thrill.

I made some serious changes.

1. CSS has one of the most important part in the system. TF remained H4, but i put up a second CSS to see the daily as well. I need to clarify rules, but some can be set in stone.
- buy strong/sell weak
-buy strong when on H4 is weak and turns back to north (a possible new trend after profit realisation)
- sell weak when on H4 is stronger and turns back to south
- trendlines can give you hint on CSS what should happen.

2. I reconstructed the Basic chart into a cleaner one. my base BB remained 186, but every other thing was thrown out except RSI. SD-s are 0.525 (40%); 1.15 (75%); 1.645 (83.3% or 5/6) and 1.96 (95%). as far as i see now these parameters can give you the most important sections of price movements (non-trade or noise zone; range zone, a normal endzone of the trend; and the last one is the outer zone or the no trade against it zone.

3. RSI. Rsi has the corresponding lines to the upper mentioned sections, such as 44; 36; 25; 20. It needs to be tested first, but these numbers reacted well on trend changes.
- trendlines on Rsi can show valuable hints of the next move.

Yesterday's trades:

+ JPY (yellow) and CHF (blue)
- GBP (red) and NZD (butter color)

SS.JPG

Lets see the four pairs

NZDJPY

NZDJPY.JPG

GBPJPY

GBPJPY.JPG

GBPCHF

GBPCHF.JPG

NZDCHF (i haven't traded it)

NZDCHF.JPG

Every time i calculate the SL to 0.35% of the price with a 5% risk. These trades are beyond the 3:1 ratio. I hope there will be a couple of others.