patrick.net

Graybox says

What is happening in the current markets as far as I can tell is counter-trend trading which can be made up of whipsaw action.

Daily SPX prices having been in mean-reversion mode since the late 1990's.

The USD will take a break for a while, but its long term trajectory is up. As global deflationary debt gets destroyed, the dollar will be in ever higher demand because most of the world's debt is denominated in dollars.

http://www.globaldeflationnews.com/u-s-dollar-indexelliott-wave-update-for-week-ending-11072014/

Graybox says

Peter P says

Daily SPX prices having been in mean-reversion mode since the late 1990's.

That very well may be, I'm not familiar with with the expectation or how that is calibrated. Mean-reversion can be calculated on many different time frames (yrly, wkly, daily etc.) and reference points.

You will get different results for different time frames.

The personality of a market is determined by its traders and their strategies.

Graybox says

Basically speaking EW is a system of Fibonacci sequences and can be useful, however is subject to "price" not the other way around.

EWT has nothing to do with price but rather pattern. Fibonacci levels are not part of Elliott Wave Theory and are, at best, guidelines for anticipating support and resistance levels.

If you are truly interested in learning EWT then you can start here...

http://www.globaldeflationnews.com/introduction-to-the-elliott-wave-principle-the-key-to-understanding-market-behaviorpart-1-the-basic-8-wave-cycle/

If you are interested in a sense of history, start here...

http://www.globaldeflationnews.com/the-elliott-wave-principle/

I am very skeptical of any grandiose theory on the market. In the end, it is a highly dynamic game that challenges your epistemology as much as your knowledge.

Graybox says

Nothing to do with it except to be subject to price meaning your chart may or may not reflect price as your chart clealy displays. Your pattern does not confine price and your chart is merely showing the EW concept and what it predicts price to do. darlag says

Fibonacci levels are not part of Elliott Wave Theory

http://en.wikipedia.org/wiki/Elliott_wave_principle#Fibonacci_relationships

From the link: This quote from RN Elliott,

Elliott developed his market model before he realized that it reflects the Fibonacci sequence. "When I discovered The Wave Principle action of market trends, I had never heard of either the Fibonacci Series or the Pythagorean Diagram".

Elliott developed his theory independent of the Fibonacci sequence. When he later discovered the sequence he realized how often the wave ratios reflected them and thus its usefulness in anticipating wave relationships. But Fibonacci is not a component of EWT, nor is it required to practice the Wave Principle.

Graybox says

I would think twice before totally trading against the trend in earnest and would be more apt to buy into the selling in support of the trend.

Trading against the trend is never a good idea. EWT does not do that. In fact, it provides markers within the trend that, used properly, allow for determining minimum risk trade setups as welll as points of entry and exit.

EWT does not purport to know or predict where prices are going. It provides a framework of wave structures upon which to (humanly) assess the most likely direction of movement.

Markets are no more predictable than the humans trading them. Elliott Wave theory is simply good at recognizing (subconscious) human tendencies by looking at the patterns in the unfolding price structure.

darlag says

Trading against the trend is never a good idea.

In many cases, a "trend" is known only after the fact. By the time you recognize a trend, much of it is already history. By the you are out, it has been reversed for a while!

Most actively-traded markets (e.g. stocks) have a rather strong mean-reversion component. Highly-manipulated markets (e.g. forex) may still tend to trend.

Graybox says

Trends are very recognizable, trend is your friend sort of thing.

Not in the beginning. You won't know whether it is a counter-trend move of another trend.

Similarly, it is hard to tell if there is a counter-trend move inside the trend, or the trend is ending.

Trend-trading can be very profitable in markets with long, strong trends and fewer counter-trend moves.

Of course, order flow information like price areas can be extremely helpful.

It sounds very reasonable. Price action around certain area is a very good piece of information. Some price levels are important because a large group of traders may be shifting from hope to fear or vice versa. :-)

Graybox says

Volume in the currency markets

What do you use as a proxy for currency volume? As a non-centralized trading market that information is not readily available.

Graybox says

That's not something I can really post on the board.

Of course. :-)

If you could reconstruct the FX order book in any meaningful manner you would have a sizable edge.

Graybox says

Are you a programer Peter?

For many years I have been. I try to think of myself as something else nowadays.

Nice! What kind of leverage did you use?

Graybox says

33/1 which is a greed safety measure.

Is 33:1 somewhat related to the Kelly Criterion of your strategy?

Graybox says

No no my approach is not that defined. Most either don't trade enough or over trade never defining their systems.

You may want to play with Monte Carlo simulations to see the effects of leverage on the possible equity curves. Usually, this involves randomizing the order of trades and/or simulating trades using the trade statistics.

In the end, you want to see if you like the expected outcome and if you can live with the worst-case result.

Here is an example:

http://www.financialpicks.com/systems_201.html

Graybox says

I'm more liken to a cards counter playing the odds.

The Kelly Criterion was actually used by the MIT Blackjack team. Card-counting provides the edge. Kelly exploits the edge by maximizing return.

Optimal bet-sizing has as much to do with your return as entry and exit points.

BTW, you can experiment with different bet sizes using Excel formulas. No coding is required. You will be amazed how much difference it can make.