Any pin will do?

This is an idea that I've been exploring. It's a product of what I have learnt

and what I believe to be true so far in my journey into forex.

I have not traded live yet and will not do so until I can keep a demo account

going for a while. I have found this very difficult as i don't seem to be able to

stick to a set of rules.

I have discovered this and quite a few other things about myself so far and

have been trying to develop a system based on my personality, beliefs and

lifestyle.

Please remember that these are my beliefs and thoughts.
They change and develop all the time and in no way am i asking you to agree. Please comment and critisize.

My thought process has been as follows.

BELIEFS

I have read a lot of information about pretty much everything here at FF and have tried to distill from it a few truths about forex (if there are any)

This is what i am left with.

The only thing that price seems to respect are significant levels, whether you call this support, resistance, PPZ's, previous highds/lows.

Price reacts to these but its pretty difficult to predict how, will it bounce or
break?

Smaller time frames expose themselves more to the noise of forex and also the costs (spread) become more significant. On the other hand if this can be managed then scalping can have a number of benefits, mainly that of growing your account quicker.

Trends are great but they can and usually do end as soon as you join them.
Peope say let your winners run and cut losers short, how do you know a
winner?, you cant predict the future.

The weakest point of any method is human intervention or emotions, wether it be flexing your money management rules or 'knowing' were the price is going.
I think another word for 'discretion' is 'guessing'. If you can not quantify any part of your sytem it is basically guesswork.
Don't get me wrong, some traders are very good and cosistent guessers - im not.

The only way to win over time is to find an edge, stick to the rules which give you that edge, have belief in that system and then exploit it.

I believe that an edge is born through probability. If you can prove that
something happens more often than not, over an acceptably long period of
time, with accuracy then maybe, just maybe you may have found one.


System Theory.

O.k. so now you know where my mindset is i'm going to tell you how i'm trying to trade.

I say trying because this system needs considerable refinement.

What i want is your comments to see if its worth me pursuing this idea or not

I suppose this system is really an offspring from the James16 thread so much credit is due there.

It basically uses price action in the form of pinbars to trigger trades.

A pinbar shows price being rejected form a support resistance line.

James16 embraces this price action along with outside bars and double bar

high/lows with a lower/higher close, tries to weave in some other reasons to take the trade such as fib levels, moving averages etc. and then uses
discretion to make the trade.

I believe that this method works but am not convinced about certain aspects of it. Mainly the moving averages and the frequency of trades.

I mean you can use any m.a. value you want from 1 to 1000, at some point
its going to look relevent to something. And man, i must be really impatient or something but trades are so infrequent on the daily timeframe that i cant
stand it, and it would take even longer to prove it.

Maybe this is a big weakness for me, maybe this method will gravitate to the longer tf's, but if there's any way possible of navigating the volatile waters of the H1 then id like to try. As i said earlier, scalping has its benefits

I think that the main thing that i don't like is the discretion side of it. i really
feel much more comfortable with quantifiable decisions, would i be any more
successfull trying to hand pick trades than i would trading everything that met my quantifiable parameters? This is what im trying to find out.



Indicator


I am only using one indicator and it is attached to this post. It is the

PriceAction AHA 0_2 by hua ai.

I have it set up to just recognise pin formations. I don't like staring at the
screen because that's were the discretion creeps in. When the indicator
alerts, i go to the screen and see if a pin bar has formed

This indicator is adjustable but it seems to have a mind of it's own.
I have tried to look at the code but cant work it out. Some times it misses out perfect pins that would have been sccussful and there are other ways that i would like to alter its behaviour.

So when a pin is formed i enter on the open of the next candle and place my stoploss at the top or bottom of ther pin. The TP is then placed an equal distance away.

Its as simple as that. There is nothing to think about other than if the pin is
big enough.

What i mean by that is, as the SL and TP get smaller, the spread becomes
significant and reduces the risk reward to unacceptable levels.
Basically every pair with its spread ahs a minimum size bar that i will trade
before it becoimes too risky.

Just using this very basic system in backtests it has proven to have a

win/loss ratio of 2:1 or 70% with a risk reward of no less than 1: 3/4

This is for all timeframes H1 and above and uses around 100 sets of data in each test

I have yet to find a pair or timeframe that does anything less than break even.


I have thought about other TP levels but what i want is a 100% accurate backtest which i can do visually with a set tp.




There may be many ways to advance this method:

Define and quantify what makes a good pin bar, and be able to input these variables into an indicator/ea

back test with all these variables to find the most favourable conditions

Dont know about filters - don't really want to get into MACD RSI etc i think the simplicity of this system is the key.

Heres a chart to show visually what i am doing and then the last 10 days on e/u H1




By the way does anyone have any advice on posting charts, this is new to me and its difficult to know what size to use. Also does everyone use paint to mark them? how do i paste them to my post

Any way thanks for reading this and please comment, i would really love to hear your opinions and thoughts

I'm giving you credit for applying such an easy(-to-follow) system, accepting that you're new to trading and should be focusing on really "simple" setups (like your example of pure observations on price-action) in the start. And of course money management.

Thousands of newcomers spend damn much time looking for the perfect indicator and spend 90% of their chart-looking time 10 cm below the price-chart while all the action is above that indicator. It's like being on a theater and only stare at the guest to get a "feeling" of what they think of the show instead of watching it yourself and make your own thoughts and experience.

I haven't studied price-action (like james16 great work) as much as other participants here on FF have done, but I have noticed myself some high-probability events like i.e pin-bars.

If you have the mentally "asset" to only take high-probability events like pin-bars, which don't occur too often on higher TF, and trade your confidence up on demo first, I think you have the probability to do better than many many newcomers in any given market. Maybe a reward for doing consistent gains on demo should be including 1 extra PA like i.e bearish/bullish outsidebar ? I wish you all the best with your learning-curve.

I also look at pin bars for entry and they have a high accuracy rate. The best pin bars are when they're bouncing off support and resistance points or piercing the outside edges of the bollinger bands. Take a look at that and you'll see they're 75% successful. Paris

Thanks Paris ,what you say is encouraging.

I respect that you are using bollinger bands to give confirmation, i think that there would be many ways to add to the 'weight' of a signal. Like i said earlier, James16 uses fib levels and confluence of other factors to signal a grade 'A' pinbar.

I am just trying to simplify things and acknowledge that a pin bar alone shows a strong rejection of price, whether it's because it has hit a moving average, bollinger bands fib level, or whatever.

To determine if its a good pinbar or not, one could simply use the price action as a measure, i.e the lenght of the nose/tail , body size, exposure of the nose above previous price levels.

What i'm trying to do is remove discretion from this system as i believe that is what makes it weak( at least in my case).

I am acheiving 70% success rates in theory and this is without any filters, i just trade every signal, possibly showing that there is no need for any filter at all.

Hope this makes sense and good luck with your pins!

I'm wondering how you all quantify a pin bar.

I have always thought a "short" pin bar had an upper wick that is at least 4 times the body length, and a lower wick that is less than the body length.

Consider splitting your trade into two and letting half run until a reversal signal occurs. This might get you into some longer trends and trades.

As for posting charts, you can practise here:

http://www.forexfactory.com/showthre...posting+charts where literally thousands of Forex factory visitors have honed their skills without fear of mockery.

Hope I helped. Good trading.

Hi RonaldReagan, thanks for stopping by, I enjoy you're posts and am looking forward to seeing what you think of this system. I get the feeling that you are also a very quantitative type of trader.

What makes a good pin is probably the most important question.

To answer this i think we need to analyse exactly what a pin is showing us.

In my eyes it is a graphical representation of price being rejected from a price region.

A long nose shows that price made a thorough bid for freedom but was severely corrected for whatever reason. I think the theory is that when this kind of resistance is felt the price has a certain amount of momentum to carry on in the other direction.

Obviously getting this figure right is a matter of back testing, is it 4:1, 3:1?

The indicator has two inputs

Min nose ratio and max body ratio

I have the default settings of 60 for min nose and 40 for max body.

Being a programmer you will be able to tell what maths is going on here but i think that min nose ratio set to 60 means that the nose must be at least 60% of the length of the bar

max body ratio i think means that the body of the candle/bar must be no more than 40% of the bar length.

In the code there also seems to be some other calculations regarding the amount that the nose protrudes from the previous candle and something else i can't work out.

These are the parameters that i need to work on to find the optimal settings

hey thanks for that i'll go and have a look.

One thing that the indicator does not allow is the setting of a min bar length.

As timeframes are shortened obviously the average candle lenght is reduced, this makes the spread more significant.

Looking at the 5min timeframe e/u for instance, there are bars being flagged up of only a few pips in length.

This smaller size does not seem to make the bars any less valid as far as win loss ratio goes (the pins are showing exactly the same price action just on a smaller tf)and there are a hell of a lot more potential trades.

The problem is that things happen too fast, as in the price only has to move a couple of pips and youve missed your entry.

Also the r:r ratio drops because of the spread.

One way round this would be to adjust tp but in doing so would lower the win rate.

If you set a risk/reward ratio, you can work out what the minimum length of the candle must be to maintain it with any given spread

I really feel that backtesting is the way to go on this technique.

The good thing about it is that the backtests are 100% accurate. This is because you are only using price for your data, you can see exactly where a candle opened/closed etc.

However there are al lot of variables

body length
nose length
nose protrusion
min candle length
min risk/reward ratio
various take profit strategies
timeframes

im sure there are others. All i have done so far is some very basic testing just to see if there is anything to this method and i believe there is.

The rest will take a lot of time but if its profitable then its worth it

Hi lesaurus.
in fact not any pin bar will do but with a sound money management and positioning along with discipline to cut your losses short,let your profits run and to take only the best setups you ll be way ahead from the mass.
Have a good journey...
One thing to remember:THE EXITS SEPARATE THE WINNERS FROM THE 95% OF THE TRADERS AND NOT THE ENTRIES.(just food for thought)
P.S.:STAY AWAY FROM THE INDICATORS...THEY ONLY TELL WHAT HAPPENED IN THE PAST

Though that may be true, keep in mind that the whole premise of quantitative trading and technical analysis is to find distinct, repetitive features in the market, which makes my bread and butter.

This isn't to say technical analysis is superior, to fundamental analysis, but certainly better than shooting in the dark.

Thanks for commenting panos69

When you say let your profits run what do you mean, i've never really understood this.
You can't tell where price is going, or whether a winner will not reverse in the next second and eat all you're profit.

I am using a fixed take profit so that backtests are accurate. With accurate backtests, i have confidence in following the system to the letter and hopefully realising a profit

Sorry Ronald Raygun, i refered to you as Ronald reagan ooops. hope i didn't offend

A pencil and an eraser. I use squared paper. The F12 key (Metatrader strategy tester). And lots of hours. It's fun at first when you find how much money you could have made/lost.

Look at indicators when you take a succesful virtual trade and where they were when you exited.

I could let you know where there is a large stack of used note pads...

Youre right backtesting takes time, and lots of paper. My girlfriend goes mental over the dozens of bits of paper and envelopes with unintelligable scriblings all over them.

I'm trying to guage a reaction from you guys as to whether i'm along the right lines and then if i feel i'm onto something i'll spend some serious time doing the work.

The way i measure a pin bar is the wick of the candle must be at least twice as long as the body. one and three quarters is o.k. too. The longer the wick the stronger the signal. Paris

If you get a pin bar and indicators are showing divergence, this is a very strong signal. Paris

Nice pin bar on the EUR/CHF 4 hr. time frame. The pin is also bouncing off the 62ema support level. Nice! Paris

Youre right paris, however i got in on the h1 pin, currently 1/3 of my way to t/p (11 pips)