Random thoughts on price and volume

[tss42]

My thoughts on trading with price and volume, in no specific order:

Volume
I personally don't see the value in volume. Volume confuses me because in each exchange volumes are different. Volumes in index futures are different from volumes in ordinary stocks. I am reasonably sure volume in US exchanges are different from in continental exchanges. This happens simply because trade reporting requirements are different in each exchange. (I am not even going into Forex). My view is a trading system is good if it's market-invariant and timeframe-invariant. The only market parameter that is context-invariant is price. This is why I am a great fan of Dunnigan.

Definition
My view is that trading systems need to be so well defined that a computer can trade it. I hear this debate on mechanical vs discretionary trading and that puzzles me. A trading system needs to be totally mechanical, or are you going to decide in real time whether you have a reversal or a pullback? If you don't know exactly what you are looking for, then how can you ever assess the reliability parameters of your system (e.g., expectation)? You can do all the research on the market you like but that will have to be before you trade. When you trade, you have to trade 100% mechanically according to your rule book. And your rule book should be 100% defined. You should be able to program in a computer what you mean by trend, reversal and continuation, and any other market attributes you are going to use.
Interestingly, if you think about it, all successful businesses work like machines, according to their very well-defined rulebook. They have done their homework beforehand. When you ask for a Big Mac, do you want the staff to be discretionary or mechanical? This is the other thing I like about Dunnigan. I was surprised to see that on Amazon Dunnigan's book gets 0 review. Not a single one.

Support/resistance
Which leads me to my next topic. I never found much value in analysing S-R simply because 1) I can't define them, however hard I try, 2) S-R are not lines but zones, 3) I can't tell if the market is likely to pierce through the S-R zone, or bounce off it, or pierce a little and then bounce, or bounce and then pierce.
In short, they tell me zilch, but that's just me. (Of course I can tell after the event but that doesn't help me when I'm trading).

 

[dbphoenix]

As for trading systems, do you have a consistently profitable mechanical trading system?

As for Dunnigan, professionals know him, beginners rarely do. In any case, Amazon has 8 reviews of his One-Way Formula book.

 

[tss42]

You are right. I have the edition with ISBN 1897597576. It seems there are multiple editions.

Regarding system, I am in the process of defining my system now, based only on price action. I found Dunnigan's One-way Formula very thought-provoking.

 

[dbphoenix]

That's the paperback. You'll find the reviews associated with the hardback.

Good luck with the system. Keep in mind that the OWF was a work in progress. Unfortunately, Dunnigan died that same year.

 

[Rhody Trader]

Volume confuses me because in each exchange volumes are different. Volumes in index futures are different from volumes in ordinary stocks. I am reasonably sure volume in US exchanges are different from in continental exchanges. This happens simply because trade reporting requirements are different in each exchange. (I am not even going into Forex).

Your arguments would be fine if you did comparissons between markets, but you don't. Volume analysis is usually only done as a relative comparisson. As long as the method for volume reporting for a given instrument remains the same, then the data can be useful.

 

[tss42]

That's the paperback. You'll find the reviews associated with the hardback.

Good luck with the system. Keep in mind that the OWF was a work in progress. Unfortunately, Dunnigan died that same year.

Thanks.

Yes, Dunnigan sadly died immediately after publishing OWF, and you can see the improvements over the Thrust Method. I am not copying Dunnigan by the way. It's his philosophy that I really like, and the way he went about defining and improving his systems.

 

[tss42]

Your arguments would be fine if you did comparissons between markets, but you don't. Volume analysis is usually only done as a relative comparisson. As long as the method for volume reporting for a given instrument remains the same, then the data can be useful.

Yes. I however feel it is much better to have a method that is proven to work in all markets rather than having one that is proven only in one type of market (e.g. European index futures). I am a very risk-averse person.

 

[Rhody Trader]

Yes. I however feel it is much better to have a method that is proven to work in all markets rather than having one that is proven only in one type of market (e.g. European index futures). I am a very risk-averse person.

I think you miss my point. When you employ volume as part of a trading system the system looks at volume for each markets independently of any others, just as it does for price (unless you have some cross-market system of course). If you're not dealing with forex (and even then you have futures market volume) then there's nothing at all stopping you from including volume as part of a system which operates across multiple markets.

P.S.: How is having one system for use on multiple markets being risk averse?

 

[tss42]

I understand what you are saying (I think), but my point is either you have to stick to the same (or similar) market forever or re-evaluate and re-test your strategy every time you think of trading a different exchange/market.

Just to give a very hypothetical example, if your reversal entry involves evaluating buying/selling climax based on volume, then you may not see the same (or even similar) level of volume in a different market.

 

[DaxDestroyer]

I use volume very rarely as it can be too easily manipulated. And when I use it, then for short term divergences. But friends of mine use volume and are successful as well.

Seems to be , that there is no holy grail, only hard work and if you follow your rules (the ones, you found which have an edge), you will earn money

 

[Rhody Trader]

Just to give a very hypothetical example, if your reversal entry involves evaluating buying/selling climax based on volume, then you may not see the same (or even similar) level of volume in a different market.

That's true, but I go again back to the idea of relativity. A climax like that is a relative thing, meaning volume is some multiple of the average or recent amount. It's not some hard and fast figure - or at least it shouldn't be since volume changes over time in all markets. Just like you probably wouldn't use hard and fast price movements in most systems because prices also change over time and a 5 point move now might not have the same meaning as it does this time next year.

 

[firewalker99]

Support/resistance
Which leads me to my next topic. I never found much value in analysing S-R simply because 1) I can't define them, however hard I try, 2) S-R are not lines but zones, 3) I can't tell if the market is likely to pierce through the S-R zone, or bounce off it, or pierce a little and then bounce, or bounce and then pierce.
In short, they tell me zilch, but that's just me. (Of course I can tell after the event but that doesn't help me when I'm trading).

You talk about the requirements for a system to be 100% programmable. I'm very curious how you would define (by using strict, unambigious rules and definitions) what a S/R area is. Perhaps these levels are sometimes too flexible to fit into rigid rules I don't know... or you would need to create an algorithm that reassesses S/R throughout the next hours or days or whatever your prefered timeframe is. On the other hand, it's true you don't want to be asking yourself what to do next if a situation materializes during the trading day. You need to have rigid rules, but are these rules always convertible into a completely mechanized strategy?

 

[tss42]

As I said in point 1, I am unable to define support and resistance.

 

[dbphoenix]

Perhaps you could clarify just what it is you're looking for. Your thoughts on volume and S/R are not random but rather uniformly negative, at least in terms of their being of any value to you.

Do you want somebody to persuade you that they are in fact of value? Or do you have a case that you want to win?

Or do you want advice on creating a mechanical system based on Dunnigan's continuations and reversals?

In a nutshell, what's the point of the thread?

 

[Trader56]

One of the (many) problems I have seen with so-called 100% mechanical systems is that yuo end up trying to quantify something created by human activity, and perhaps more importantly, human emotion.

LTCM leaps to mind, as do Wyckoff's comments from 100 years back when people were trying to do the same things then.

 

[dcraig1]

One of the (many) problems I have seen with so-called 100% mechanical systems is that yuo end up trying to quantify something created by human activity, and perhaps more importantly, human emotion.

LTCM leaps to mind, as do Wyckoff's comments from 100 years back when people were trying to do the same things then.

I find this argument rather less than compelling. Many human activities are quantifiable and indeed predictable. Why should the financial markets be in principle any different ?

This is not to say that creating good trading systems is easy. It plainly isn't.

I really can't see what LTCM has to do with it. No doubt their models were inadequate and most of all, too much leverage did them in.

Whether they like to admit it or not, PV advocates are also quantifying "human activity" with S/R levels, volume patterns and whatever. It's just not rigorously defined in an algorithmic sense.

 

[tss42]

Do you want somebody to persuade you that they are in fact of value? Or do you have a case that you want to win?

Or do you want advice on creating a mechanical system based on Dunnigan's continuations and reversals?

In a nutshell, what's the point of the thread?

The purpose of the thread was just to express my own experiences. Sorry if this has come across the wrong way. I wasn't trying to win a fight here.

No, I am not looking for advice/guidance. I have to figure it out myself.

 

[TheBramble]

As I said in point 1, I am unable to define support and resistance.

If you care to let me know any one of your most studied instruments, I'd be more than willing to annotate a chart with SR lines for you to watch subsequent price structure & development around/through those levels and for you to get your own learnings that way.

 

[OpenMind]

Is it definable in a programmable language?

 

[dbphoenix]

The purpose of the thread was just to express my own experiences. Sorry if this has come across the wrong way. I wasn't trying to win a fight here.

No, I am not looking for advice/guidance. I have to figure it out myself.

Understood. However, no offense intended, you likely don't have nearly enough experience to state what needs to be or must be or has to be. I suggest you work on keeping an open mind with regard to volume and S/R since experienced trades have been depending on these elements for more than a century. To dismiss all of that simply because you don't understand them does not bode well for your growth as a trader, mechanical or otherwise.