Patterns eg. Double Top / Double Bottom

[cshieldsx]

Just a quick question for everyone, how effective is recognising patterns as a technique ?

For example, a double top pattern, forms across a few months, relatively textbook formation, with confirmed support turned resistance, how often is that just complete fluke and forms and then it turns out that there is no change of longterm trend ?

If you used no other analysis, no indicators and no fundamentals, just staring at a pattern on a chart. The same goes for triangle patterns etc. How reliable are they ?

 

[damianoakley]

Just a quick question for everyone, how effective is recognising patterns as a technique ?

For example, a double top pattern, forms across a few months, relatively textbook formation, with confirmed support turned resistance, how often is that just complete fluke and forms and then it turns out that there is no change of longterm trend ?

If you used no other analysis, no indicators and no fundamentals, just staring at a pattern on a chart. The same goes for triangle patterns etc. How reliable are they ?

Hi,

The reliability of patterns depends on what market and timeframe you are trading.

What you need to do is pick a market, pick a timeframe, pick a pattern (eg - double top/bottom), and then look back over time on every situation where the pattern occurred and see what happened. You will literally need to look back on hundreds of previous situations when the pattern appeared.

This is a lot of work, but it is the only way to determine the validity of trading a particular pattern on a particular market in the future.


Thanks

Damian

 

[dbphoenix]

Just a quick question for everyone, how effective is recognising patterns as a technique ?

For example, a double top pattern, forms across a few months, relatively textbook formation, with confirmed support turned resistance, how often is that just complete fluke and forms and then it turns out that there is no change of longterm trend ?

If you used no other analysis, no indicators and no fundamentals, just staring at a pattern on a chart. The same goes for triangle patterns etc. How reliable are they ?

If by "change" you mean reversal, this isn't defined by a double top but by a breach of the last swing low. The "double top" may be nothing more than the beginning of a lengthy consolidation. If you understand how volume illustrates the balances and imbalances between demand and supply and apply that understanding to your pattern, they are very reliable.

Db

 

[cshieldsx]

hundreds of previous instances ? but on a daily chart of a pattern that may take a few months to fully form....for an equity that only has data back to the early 90's - I'll be lucky if I find 10 instances of the pattern !

 

[Paul Wilson]

Patterns can be a be reliable to trade. But like most set-ups or strategies they can go wrong.

In my experience the best pattern to get to recognise is head and shoulders top. I have just been flicking through the FTSE 100 and there have been a few over the recent months I could have made a killing on. I have had success with triangles, but I haven't traded that many.

If you have some time to burn, check this pattern resource...

http://thepatternsite.com/chartpatterns.html
(mods I apoligise if I shouldn't have added that link!)

The site is owned by a guy called Thomas Bulkowski. He has done lots of research into chart patterns.

You will probably notice that it isn't just as easy as checking the price close. Other things can be taken into account including volume, support/resistance, timescales etc.

The website above will give you some additional information in regards to a particular pattern. If you get his book - The Encyclopedia of Chart Patterns(This is not an endorsement, but it cost me about £40 and paid for itself on the first trade I made from it) it will go in to more statistical details of what patterns perform better, and under what circumstances.

If you are new to this, keep it simple, trade equities have high trading volume(ie are liquid and active). Don't trade anything that you think "might" look like a pattern, or hasn't been confirmed.

If you use leverage you might want to keep with lower price stocks.

Also if you are thinking about trading a pattern, check there aren't any company dates around the corner, i.e annual results.

Any questions send me a message.

Regards - Paul.

 

[Paul Wilson]

hundreds of previous instances ? but on a daily chart of a pattern that may take a few months to fully form....for an equity that only has data back to the early 90's - I'll be lucky if I find 10 instances of the pattern !

Once you get used to it you will be suprised how easy they are to spot.

There isn't much happening in the FTSE at the mo with last weeks shennanigans. When something decent appears I will post up what I find.

That is one of the problems with only trading patterns, is that they aren't there all the time. Really you need to have a few more tools in your box.

I started with patterns. But now I am developing some different ideas for different circumstances.

 

[Paul Wilson]

Here are a few recent ones...

Easy to spot in hindsight. I rode HMSO to the bottom

 

[cshieldsx]

For this type of pattern trading, are you guys spread betting or actually trading the equities themselves ?