has anyone studied Toby Crabel's ideas?

How do you know when a trend has started? Ended? This forum is for discussions about trend indicators and signals.
Post Reply
richard
Roundtable Fellow
Roundtable Fellow
Posts: 94
Joined: Tue May 18, 2004 10:17 pm

has anyone studied Toby Crabel's ideas?

Post by richard »

Toby Crabel has some interesting studies on historical volatility systems.

Have you ever used them in your trading systems?
Nathan
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Sun Jul 13, 2003 3:52 am

pattern.

Post by Nathan »

Yes the consepts, no to the specific patterns.
richard
Roundtable Fellow
Roundtable Fellow
Posts: 94
Joined: Tue May 18, 2004 10:17 pm

Post by richard »

I wonder if anyone is trading systems based upon cyclicality of volatility and the rate of change of volatility?

The rate of change of volatility can be used in a trend following system to change the exit signal, for instance. Perhaps a MA used for exit (say a 5 day) can be moved to 3 day in event of certain rate of increase in volatility, that type of thing just to use a dumb example.

I am looking into this idea which seems key to me.

Any thoughts or experiences?
damian
Roundtable Knight
Roundtable Knight
Posts: 815
Joined: Tue Apr 15, 2003 8:43 pm
Location: dusseldorf

Post by damian »

The rate of change of volatility can be used in a trend following system to change the exit signal, for instance. Perhaps a MA used for exit (say a 5 day) can be moved to 3 day in event of certain rate of increase in volatility,
Richard, I use 'open trade management' methods that are a function of, for want of a better description:

a) directional volatility
b) ummm, directional volatility with respect to time.

A related example might be the Strength Filter (I think that is what it was called) in the OTRs which is used to determine which is the strongest market out of a group. You may want to check it out yourself, but it relates volatility (ATR) to units of momentum (change over time). You could take such a strength measure and tighten stops according to the reading.

If you will excuse a diversion into a related topic:

I have mentioned this before, but the challenge that then fronts you is how to re-calibrate the system so that it takes an entry on the 'base' entry rules. for example, on a tightened trailing stop you may exit at above HH50.... and it just so happens that you have n entry rule that says "buy tom at HH50 stop". Ergo, you would re-enter the position the day (after) you exit the position... not what you want. Individual flair is needed in structuring re-entry rules.

I have tested such a system and it returned excellent results (see an overtly gushing post of mine a while back that has since been brought back to account by vigilant forum members ;)). I have traded such a system and it returned a loss, however backtesting over the same period which I traded also produces a roughly equivalent loss. So at least the system is performing as testing indicates. The problem is, among other issues, that I have a knack for selecting every market that will not trend until my money runs out :wink:

[sorry, i dont think any of this has anything to do with the original Toby Crabel theme]
anita venetian
Contributing Member
Contributing Member
Posts: 6
Joined: Wed Aug 25, 2004 9:59 pm

Post by anita venetian »

Are there any indicator packages for sale that use Crabels techniques? I understand that opening range breakout is one of his tools but assume there are others. His performance does not look good from a return prospective but the low volatility has helped him to raise a lot of money.
blueberrycake
Roundtable Knight
Roundtable Knight
Posts: 125
Joined: Mon Apr 21, 2003 11:04 pm
Location: California

Post by blueberrycake »

One of the main challenges that I see with Crabel's methodology, is that he looks to create a list of price patterns, that he in turn uses to confirm his main setup. However, many of the patters have so few trades, that it's not clear how anyone can verify them statistically with any degree of certainty.

-bbc
Nathan
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Sun Jul 13, 2003 3:52 am

f

Post by Nathan »

"However, many of the patters have so few trades, that it's not clear how anyone can verify them statistically with any degree of certainty."


My comment is not specifically with regard to any particular idea of toby crabel's.

However, it does have to do with defining and testing ohlc patterns and patterns in general.

Instead of thinking of testing ohlc patterns as an exersize in finding specific, one of a kind patterns, think of it as looking for general market tendancies that can take many specific forms.

If u understand and consider a tested pattern conceptually, u see that often many individual "patterns" are all really basically the same idea, or what i call the same "macro" pattern.

So, while one "Pattern" may only happen 5 times a year in a market, if u identify the macro concept, u may find it happens 100's of times a year, or something like this.

Perhaps combine some of the empirical with a willingness create abstract classifications.

THis is my approach.
Post Reply