System metric for identifying range bound/consolidating markets

[julius]

As per the title, I'm looking for a way to identify range bound / consolidating market conditions for use as a system variable.

Easy characteristic to recognise visually but I'm not exactly sure to quantify it...

 

[Nick Radge]

What time frame?

You can also do it without a price variable, i.e. using volatility

 

[julius]

Nick, time frame is intraday : ~15 minutes

I've tried daily consolidation patterns & intraday consolidation patterns but not really sold either way on which is more effective

The main reason is to use it as a system filter / trigger

 

[Bin57again]

Julius
Can you not use BBands or StdDev or even ATR?
Some authors have also looked at patterns e.g. NR7 (Farley).
ADX is useful as a filter for trends (e.g. only take signals >25).
Bin

 

[>Apocalypto<]

Nick, time frame is intraday : ~15 minutes

I've tried daily consolidation patterns & intraday consolidation patterns but not really sold either way on which is more effective

The main reason is to use it as a system filter / trigger

nice Avatar! r u a Livermore Fan?

 

[julius]

Bin, I've tried this but volatility often drops without the market moving into a consolidating pattern - as is often the case in up trending markets.

Apoc - yes I am a fan, but it is also a great picture & story

 

[Bin57again]

Julius
I'm not sure I quite follow you - if volatility drops then price should be contracting or at least slowing. You should then be able to see if you're in a pause in an uptrend or if distribution is taking place. Nick Radge's book is good on what he terms "micro patterns". It's not much different to Darvas conceptually.
Maybe the problem is not your approach but your data i.e. the 15 min charts? I find 15 min incredibly choppy. If you're trying to apply your concepts to 15 min then maybe there's too much noise? Have you tried applying your concepts against a higher timeframe (e.g. 1 hr) and just using 15 min for entry? You could also combine your volatility with pivot levels, so for example, if you knew you had a weekly resistance above you coming up in 50 points and you'd had a 2 day rally from a lower weekly open, you might start thinking that going long is a bit risky...then you could look for a slow down? If it consolidates and then pushes through e.g. the Friday effect into the weekend then maybe go long with your stop below your technical placement (which might coincide with the weekly resistance level)?
Bin

 

[Nick Radge]

A drop in volatility is a measure of a pause, but not in the sense that price is actually pausing which I assume is what julius is actually after. I put forward volatility as another gauge is all.

So if you're looking to code a specific pattern there are really two types. The NR7 has bin57 puts forward is good but again its more volatility that pattern based. We can drop that down to what I call an 'inside triangle'

An inside triangle is a wide ranging bar than then has a minimum of two bars within it high and low. These can lead to good moves.

If you want a more defined Darvas Box of other pattern then you're up for some serious coding. I have a TradeStation program that detects my micro patterns - but its 1200 lines of code. Even then, I found it better to eyeball the charts.

 

[motorway]

Use a P&F chart with a box size 1/2 the size of the contraction you seek
( this is a 1 box reversal so if you have to use 2 box use a smaller size )

This will show both pause and contraction

and being "moved base" is all time frames

You can then trade the move you want on what ever time frame bar chart you want...

The P&F chart will show you value areas, point of control, expansion contraction etc etc.. auctioning up down

Use it for any type of trading.. coordinating with the bar chart

time based + moved based is more powerful than trying to make the one chart do both ( eg bollinger etc )

The P&F chart is probably the only type of chart that can both stop in the sense of going sideways and stop as in stop (dead )


motorway

 

[Timmy]

Use a P&F chart with a box size 1/2 the size of the contraction you seek
( this is a 1 box reversal so if you have to use 2 box use a smaller size )

This will show both pause and contraction

and being "moved base" is all time frames

You can then trade the move you want on what ever time frame bar chart you want...

The P&F chart will show you value areas, point of control, expansion contraction etc etc.. auctioning up down

Use it for any type of trading.. coordinating with the bar chart

time based + moved based is more powerful than trying to make the one chart do both ( eg bollinger etc )

The P&F chart is probably the only type of chart that can both stop in the sense of going sideways and stop as in stop (dead )


motorway

Great idea ... why didn't I think of that?

Also, if it is viewed as important in your trading you could also plot a histogram below the PF showing either volume accumulating in each column or time accumulating in each column, or both.

 

[Wysiwyg]

Just read an article suggesting a range bound (sideways their words) market could unfold in the next few years along the lines of the 60's, 70's and 80's range bound market.

In my opinion there is no such thing as a "sideways" market. The markets are trending long, medium, short and in fact any term. So what if the market does not make new highs?

For the last twenty-five years, investors have experienced two types of stock markets—bull
and bear—that either went up or down over a multi-year period. But, there is a third type
of market with which many investors today are not as familiar. It is called a “sidewinder”
and it produces a sideways market - one that barely changes over time.

https://www.aussiestockforums.com/forums/showthread.php?t=14387

 

[Wysiwyg]

It has been a generation since investors have experienced a prolonged sideways market (although they were not uncommon through most of the 20th century).1 On October 1, 1975, the Dow Jones Industrial Average stood at 784. Nearly seven years later, on August 6, 1982, the Dow closed at the exact same 784.

The DOW was at 11890 in January 2000 and was at 11890 during March 2008. Eight years later. Can't see what the author is trying to say here.

The obvious question is, how can 38% of the stocks in an index go up 100% or more over a time period when the index itself barely moved? The answer is revealed not by a finance professor, but by one of the most influential evolutionary biologists of our generation – Stephen Jay Gould.

Could this be because the index weighting of these stock did not cause the index to move in conjunction?

Chart shows the DOW from 1966 to 1981 had resistance at 1000 points. A nice round powerful number. During this period when the DOW did not make new highs, there were periods of 20% to 35% pullbacks and runs back up to 1000 points. Duration of years for these trends to take place. Simply didn't make new highs. Markets move up and down


If there is to be a range bound market unfold in the following years then a resistance level will need to be established and that is 14000 points on the DOW. If the market retraces from there then this author will have a shoe in. A second retrace and he has a new pair of shoes. We shall see.