Markets Untradable Using Count-Based Strategies
Posted: 30 May 2007
I realize that most posts in this forum concern software or support issues, but after almost 6 years of studying price behavior and trending I've come to a conclusion that I feel obligated to relate to my fellow traders.
The conclusion is that all markets are untradable using count-based strategies such as EWT - and probably untradable using the more mainstream technical analysis strategies as well. When I say "untradable" I mean 'incapable of being consistently profitable.' Every trader will of course have periods of profitability and loss - my experience has been that 'loss' wins out in the long run. Here's why (at least from the perspective of using a count-based strategy):
Count-based strategies such as EWT are 'resolution dependant'. All of these strategies are based on some sort of counting pattern. With EWT it's all based on the numbers 5 and 3. The pattern in turn is tied to a specific resolution or level of detail (if you show me a 5 point count on your screen I can turn it into a 7 or 13 or 21 point count just by changing the resolution). Any trend can be as complex or simple as you like. It all depends on how close you want to look at it.
Per any given resolution you will find a hodge podge of trends with varying complexity (and that means varying counts). So trends vary their complexity from trend to trend within resolution A and from resolution A to resolution B. The conundrum for the trader and analyst is that there is no way of matching up these patterns with their respective resolution ahead of time. You cannot know that the current trend will appear as pattern X in resolution Y and you therefore will not be able to locate the terminal for that trend.
Let's say that for reasons of simplification you decide to trade a single pattern of 13 counting points. Let's say you're scanning 10 resolutions of AUD/USD looking for your count. Assuming that counting pattern does appear somewhere within your field of vision (at 1 of those 10 resolutions) how will you know where? If you guess, you have a 10% chance of being right. This is why most people using EWT have had only limited success and it is why any counting scheme is doomed to failure.
Complexity moves over resolution and time scale in the market like water flowing over rocks in a stream, and there is no endpoint (the endpoint is relative to resolution). What Elliott saw was just a slice of a certain level of detail. He was wrong to create a market model out of an incomplete picture. It would be like saying that visible light is all there is, that ultraviolet, gamma, and x-ray don't exist.
There is no magic number(s) in the market. At the resolution where a finished trend appears as a line, zooming in will reveal an infinite amount of complexity (assuming no technological limitations). Zooming out will reveal complexity outside of the line, and eventually that line will be assimulated into a higher scale line.
It is therefore the course of wisdom to avoid being taken in by groups like Elliott Wave International, or for that matter anyone claiming runaway success with 'strategy X' or 'market model Y'.
For years I was convinced that the market could be cracked, and by cracked I mean successfully traded. Now I believe the opposite. The fact that counting strategies do not work does not bode well for other types of strategies including technical ones. Even Mandelbrot, who discovered that volatility clusters, could not help us when it came to forecasting any direction for that volatility.
How do I respond when asked 'how then, do some make a living and even get rich doing this?' I point out that the great majority of traders are losing, not making money. I think this fact was born out well by the response to the April 25th post "Anyone making money using MCFX?" Only one person responded in the affirmative out of at least a dozen or so traders in this forum. Success in trading seems to fluctuate just as the market does. Some experience greater highs than others due to fortunate timing or more capital. And even when you look at the famous traders, not all is as it appears. Gann was a fake and Jesse Livermore, though he made a fortune trading, lost his fortune doing the same thing...
I wish that I was simply being overly negative here. I would love to make a living trading. If there's anyone out there who can change my mind, I'm all ears, but it won't be easy.
Thanks for reading my rant.
PS. My thanks to TSS and MCFX for helping me discover the futility and complete my studies. Your Point charts are good and they will be great once you impliment the 168 hour threshold. Still, an untradable market is an impenetrable wall. Please cancel my membership.
The conclusion is that all markets are untradable using count-based strategies such as EWT - and probably untradable using the more mainstream technical analysis strategies as well. When I say "untradable" I mean 'incapable of being consistently profitable.' Every trader will of course have periods of profitability and loss - my experience has been that 'loss' wins out in the long run. Here's why (at least from the perspective of using a count-based strategy):
Count-based strategies such as EWT are 'resolution dependant'. All of these strategies are based on some sort of counting pattern. With EWT it's all based on the numbers 5 and 3. The pattern in turn is tied to a specific resolution or level of detail (if you show me a 5 point count on your screen I can turn it into a 7 or 13 or 21 point count just by changing the resolution). Any trend can be as complex or simple as you like. It all depends on how close you want to look at it.
Per any given resolution you will find a hodge podge of trends with varying complexity (and that means varying counts). So trends vary their complexity from trend to trend within resolution A and from resolution A to resolution B. The conundrum for the trader and analyst is that there is no way of matching up these patterns with their respective resolution ahead of time. You cannot know that the current trend will appear as pattern X in resolution Y and you therefore will not be able to locate the terminal for that trend.
Let's say that for reasons of simplification you decide to trade a single pattern of 13 counting points. Let's say you're scanning 10 resolutions of AUD/USD looking for your count. Assuming that counting pattern does appear somewhere within your field of vision (at 1 of those 10 resolutions) how will you know where? If you guess, you have a 10% chance of being right. This is why most people using EWT have had only limited success and it is why any counting scheme is doomed to failure.
Complexity moves over resolution and time scale in the market like water flowing over rocks in a stream, and there is no endpoint (the endpoint is relative to resolution). What Elliott saw was just a slice of a certain level of detail. He was wrong to create a market model out of an incomplete picture. It would be like saying that visible light is all there is, that ultraviolet, gamma, and x-ray don't exist.
There is no magic number(s) in the market. At the resolution where a finished trend appears as a line, zooming in will reveal an infinite amount of complexity (assuming no technological limitations). Zooming out will reveal complexity outside of the line, and eventually that line will be assimulated into a higher scale line.
It is therefore the course of wisdom to avoid being taken in by groups like Elliott Wave International, or for that matter anyone claiming runaway success with 'strategy X' or 'market model Y'.
For years I was convinced that the market could be cracked, and by cracked I mean successfully traded. Now I believe the opposite. The fact that counting strategies do not work does not bode well for other types of strategies including technical ones. Even Mandelbrot, who discovered that volatility clusters, could not help us when it came to forecasting any direction for that volatility.
How do I respond when asked 'how then, do some make a living and even get rich doing this?' I point out that the great majority of traders are losing, not making money. I think this fact was born out well by the response to the April 25th post "Anyone making money using MCFX?" Only one person responded in the affirmative out of at least a dozen or so traders in this forum. Success in trading seems to fluctuate just as the market does. Some experience greater highs than others due to fortunate timing or more capital. And even when you look at the famous traders, not all is as it appears. Gann was a fake and Jesse Livermore, though he made a fortune trading, lost his fortune doing the same thing...
I wish that I was simply being overly negative here. I would love to make a living trading. If there's anyone out there who can change my mind, I'm all ears, but it won't be easy.
Thanks for reading my rant.
PS. My thanks to TSS and MCFX for helping me discover the futility and complete my studies. Your Point charts are good and they will be great once you impliment the 168 hour threshold. Still, an untradable market is an impenetrable wall. Please cancel my membership.