Pivot Point Squeeze, high probability system

Thanks for highlighting this, you'll notice in the above set of updates that there is now a 'decimal points' parameter, so you can easily set it depending on whether the currency pair trades in 2 or 4 decimal places ... and this will also adjust the MACD calculation - so fingers crossed all the indicators should now work for any currency pair .

Cheers, Johnny.

Johnny Forex

Thanks for the indicator programming looks excellent http://www.forexfactory.com/images/icons/icon7.gif and loading instructions great will try trading tomorrow but looks good.

Thanks again


David

GCTrader

Just a thought as I like Bollinger squeeze systems.

I was wondering whether yourself and others interested in your system fancy getting together on "skype chat" whilst live trading to share ideas and gain a better understanding of how you bring pivots into play etc.

As I said just a thought

Divad

That is a great idea, I will post a time when both John and I can be on skype at night(aussie time) and possibly when there is a squeeze on. That way we can talk about the lead up, what are the fundamentals, technicals, up and coming news and so forth.

My hotmail is [email protected] and my skype is gctrader79

GCT

Hi GCTRADER

Sent you a message on Skype.

Divad

i didnt get it mate, try and add me...

GCT

The squeeze is on although again it is a monday and a public holiday at that. This ensures less volume and a natural squeeze. My advice would be to watch it but would not commit the usual allotment if triggered. A nice trigger down would be nice which would fit in with the fundamentals of the concern over German banks.

GCT

I have been giving some thought to how to improve the system during periods of consolidation, ie. trying to filter out false entries.

Although we have the dynamic trigger line, this doesn't exactly cover consolidating periods fully. It just means that if the market has been narrow, then we need to be even more narrow for a breakout to be possible. However, it doesn't alter the actual breakout criterea, ie. a small breakout (increase in BBW) could still trigger an entry and this would be exactly the same breakout criteria for a market which had not been consolidating.

I am proposing we have the acutal breakout crieteria, ie. the required increase in BBW to be also dynamic - somehow a reflection of where the trigger line is at (but inversely so).

At the moment, the BBW is required to increase by a minimum of .0001 for it to be considered to be widening / breaking out. I believe if the trigger line is lower than 'normal', then the required BBW increase should be larger than .0001 and if the trigger line is >= what is considered to be normal, then .0001 suffices perfectly.

So GCT, we need to come up with a formula for this and then I can optimise the actual coefficient to get the best number. I'm thinking along these lines as a benchmark ...

Trigger >= .0044 required BBW increase = .0001
Trigger < .0044 required BBW increase = .0003

Naturally this needs to be a sliding scale.

rqBBWI=Requried BBW width increase
TRGL=Trigger line value

IF TRGL<.0044 THEN
rqBBWI=.0001 + (.0044 - TRGL)/3
ENDIF

Another idea, which I think I actually prefer (which does not include the above criteria), is once an entry is triggered we note the BollingerBand value at +2.4 std dev and -2.4 std dev at this time. We then do not enter the trade until price passes this level (providing that at the time this occurs an entry trigger is still showing as valid ie. needs to occur probably within a couple of hours of the supposed breakout).

This means that the breakout has to be validated as quite significant before we enter. Naturally, this may mean we get later entries sometimes - but it might tend to keep us out of minor channel trading / price varaitions during periods of prolonged consolidation, where the system currently tends to poise itself for even quite a small breakout.

Everybody's thoughts please?

Of course if you have any other ideas for filters in this regard, please don't be afraid to put them forward.

Thanks, Johnny.

There is a continued squeeze (from yesterday) on EJ. Bandwidth is very low and a break out is immanent! GCT any ideas about pending news? I just looked and didn't see anything of significance...
Rick

Correct, there is a squeeze on atm but no breakout as of yet. Keeping a close eye on it now. No truly significant news out this week that concerns us. Hopefully signal a nice trade. Missed an awesome trade on the aussie last night worth 60 pips.

GCT

In long EJ 158.94....missed the signal by 10 minutes

Those Damn HeadFakes!

Something has to be done about them! John and I will not stop till we have come up with a solution that atleast eliminates some of them... Oh the humanity...

Same here...only lost 50 pips (half lot)...learned my lesson from before!!

is anyone still following this system? how are the results?

by the way, i'm interested in setting up the mt4 files from sunflower or Alexish.... however the amount of bollinger band numbers seem to be different than the amount GCT is using.... why is that? i know it's not that big of a deal as long as they show the contraction... but just curious.

I have had the same issue, I think you need to set the MT4 indicator so that the trigger number is calculated using the same number of back-bars that GCTrader and others are using on the other platform.

I think that the MT4 indicator was programmed a while ago and the default settings do not seem to match what most traders are using now...


See the
TriggerCountBar setting in the indicator to see what I am talking about...

Hi Pipbagger,

That value is about right, atm I'm using 45. The lack of quality signals of late has me looking at other pairs however it did pick the top of euryen on friday. The volatility doesnt seem to be contracting much lately at all so there havent been many nice squeezes in a while. Still keeping my eye out in these uncertain times...

GCT

hey GCT

how do you determine the Number of bars to look back? and why do you use that number for ERUJPY.... i dont particular trade ONE pair... i look at multiple pairs and look for oppotunities within all of them.... so was curious if you could explain why you pick certain number for certain pair.

also what pivots do you use? in the chart you just posted... are those weekly or monthly pivots? (and also support and resistance i'm assuming?)

could you upload the code to that perhaps? would really appreciate it.

My first post here. sorry for my bad English.
First I have to thank GCTrader1 for share this great system.

after adjust some parameter and add in money management, in the back test, I am able to turn $400 to $29000 in 3 month.

so amazing.

the parameter: EUR/JPY H1
B=24, l=0.0078, a=6.2 p=24, macd(9,14,1)

stop loss 20 pip, use 100% capital, $200 per lot $1.

The stratege is so unique.
after read the code again and again, I still have some question to ask.

1. the furmula

if (round((BollingerBandWidth[b](close))*10000)-((BollingerBandWidth[b](close[1]))*10000))>=1

it is eg: round(2.51) -1.99=3-1.99=1.01 > =1
round(3.49) -2 =3-2 > =1
some times 0.52 dif, sometimes 1.49 dif
why ?
should it be round( 2.51-1.99) ?

if (round((BollingerBandWidth[b](close))*10000-((BollingerBandWidth[b](close[1]))*10000))) >=1

This has little effect to the stragegy, but this should be more logical.

2. the exit strategy.
if abs((round(MACDline[5,13,1](close[1])))) - abs((round(MACDline[5,13,1](close)))) >= A then

It means only when there is a sudden, big drop in price, we will only sell to close position.

but if the price is slowly drop.... it will never close the position.
I have enounted once , it slowly drop until I lose 90K out of 100K.

But in overall, this still works very well, so amazing.

I have tried other exit stragegy eg: when moment(12,close) < 0 as in some books. and that result in very low performance.

Can you explain the exact rules you are applying on the pivot point in coordination with this strategy...for example let say the price brakes the main pivot by like 5 pips and are also having a cross of the center cable of the bolling bands. Would you wait for a bigger break of the pivot and aim for R1 or S2?

My second question is, lets say you do enter on a small break of the pivot with a cross of the center cable of the bolling bands and on the next candles it crosses the cable the other way again...would you swap positions?

Thanks Guys, please help me out.

That sounds like over-leveraging and curve-fitting to me..

Be careful on curve-fitting when running backtests. I'll agree this system backtests suspiciously well on recent EURJPY data, but try to backtest it any further back (<'07) or on any other currency, and the results aren't nearly the same (if even positive at all).