zupcon

I agree, there are some truly shocking examples of this, particularly in the area of artificial intelligence research. I suppose its not surprising, the people writing this stuff are not traders. As a general rule, any academic paper that defines success or failure as a function of how much money a "system" gains or loses is probably not worth taking the time to read, and its a bit of a give away that the authors don't really get it. Most of the better quality papers ask a question such as do reversals tend to occur at fib levels, or are markets equally as likely to reverse at arbitrarily selected levels. Even that approach is rather simplistic. I read quite an interesting paper a few years ago where a couple of academic types tried to identify swing points algorithmically as the first stage of calculating fib levels. These guys had access to "traders" working within the same company, and of course, the traders always tended to disagree with the location of the mechanically identified swing points. Eventually, these guys came to the conclusion that designing a mechanical method of identifying swing points was far more complex than they'd anticipated, and so they decided to skip that step and ask the traders to manually mark the swing points on the chart. So a trader marks the swing points, they do the analysis and conclude fibs are no more significant than a randomly chosen level. At this point, the other traders claim that the swing points picked by the first guy where wrong. So they get another trader to pick where he thinks the swing points are. They repeat the experiment, and get the same results No matter which human trader selected the swing points, they always got the same result Lets not forget, these guys where retrospectively cherry picking with the benefit of hindsight and they still couldn't make it work. In practice, the firm was using fib levels in their analysis and making money. In the highly unlikely event you where a trader in this firms employment, and you'd developed an edge based on something as trivial as fib levels, I suspect you wouldn't be too keen on disclosing details to someone whose job was to encapsulate that knowledge into a few lines of code.

I have to disagree. I do know people who make money everyday, or most days, but they are not traders staring at charts. Unless your engaged in exploiting some fairly complex arbitrage opportunities, I doubt there's anyone out there consistently taking pips over such a small period of time It doesn't take much imagination to design a method of making money everyday, and I'm sure we've all been there, but generally there's another side to that particular coin. Trading is a fairly difficult undertaking because your opponents edge is a lot bigger than most people imagine. Any post that attempts to reinforce the crazy idea that you get rich by making money every day needs to be challenged. I'm no advocate of martingale position sizing, and its somewhat of a shock to see it being pimped by a moderator (well actually it isn't, but that's because I'm a cynical old git who knows how the industry works) but the adoption of the mindset of a professional gambler is in my opinion at least, almost a pre requisite to trading successfully. I'm very clear on my position, I gamble, and I manage that process, and I make money by capitalizing on good luck.

Discussing anything on a typical trading forum is always going to be problematic because there's a significant number of people who are looking for simple rule based solutions. In order for this group of people to accept that "fibs work" you"ll need to come up with some fairly well defined back testable method, but of course, the moment you do that, you have the same curve fitting bias that I described in my moving average cross example, and the more degrees of freedom you introduce the worse its going to get. On the next level, you have a smaller number of people who have started the process of trying to figure things out for themselves. These are the kind of people who might ask the question "do markets turn at fib retracement levels more frequently than for example levels chosen at random" or "do swings terminate at fib extension levels" and they might research this stuff at various levels of sophistication. This is the sort of stuff you'd typically see in academic papers, inevitably the stats indicate that markets are no more likely to turn at fib levels than any other level, (despite the fact that I could immediately look at a chart and pull out hundreds of examples where they did). Although its interesting, its not really got a great deal to do with the subject of making money, its just one tiny part of a much larger whole There's an academic paper floating around written by a couple of HSBC researchers who looked at fibs, and another looking at the accuracy of support and resistance levels quoted by various banks. If you look at those papers, the research methodology is perfectly well laid out, unambiguous, no room for misunderstanding etc. the results are crystal clear too. Those results arnt opinion, they are verifiable facts, and if you ran the same experiment you'd get the same results too. You might not believe it, but I trade using completely random entries, but I could do the same thing using fib levels, and I'd get the same results, and I'd get the same results whatever I chose to use to select levels. I'm not trying to discount fibs, they have the same strengths and weaknesses as any other TA,

The problem is we can generate 2 numbers in the range 1 to 100 and use those numbers as the basis of a moving average cross system, and I'm pretty confident that I could find a financial instrument somewhere that could be traded profitably using such a system Of course I could equally well find you hundreds of markets where the same system didnt work. There are so many variables that fib traders simply will not address in a systematic manner. Charts themselves are simply an arbitrary construction, and that's before you even start to apply any sort of TA. What exactly constitutes a swing high, swing low etc. I'm not saying fibs are not useful, I've got mountains of statistical data from years of studying these things, but the numbers and levels themselves have very little to do with profitable trading,

http://www.lhup.edu/~dsimanek/pseudo/fibonacc.htm should get you started

Most things that occur in nature that are commonly used as examples by these charlatans don't actually conform to Fibonacci ratios either !

Thats a question that only Rob Booker can answer. Clearly rforexdad has an issue with whats being taught, as do other posting on other forums. However, I do think he has a valid point, IF a vendor is encouraging students to verify the method by backtesting over a particular time period and/instrument, then that is highly unethical. Most people with limited experience and a copy of trade station could find a simple strategy that worked well historically in backtesting. The important point of course is that most new trades dont understand these issues, nor do they understand how simple it is to curve fit. Therefore they are easily fooled by the limitations of backtests. In the early days, I'd say that Rob Booker was most definately a victim of being "fooled by randomness". I wouldnt classify Rob as a trader, but he's comes across as someone who at one point at least tried, and more importantly someone whose done a reasonable amount of basic backtesting, and I would have hoped that he'd seen the light by now. For all I know maybe Rob is providing the system to his students simply as an exercise in backetesting, and an illustration of the limitations of designing systems using these sort of methods, but if he's claing the system is profitable, and selectively backtesting to verify those claims, thats unethical, and he should be called on it. I believe Rob's a CTA these days, and his performance is presumably a matter of public record so anyone whose thinking of paying for training at least has the opportunity to check him out and determine if he walks the walk.

I'd probably argue that most people find it extremely difficult to identify an edge. There's no shortage of people attempting to develop mechanical strategies, and by and large they fail. The other issue of course is that on the very few occassions that someone reveals information that might help them identify an edge, they tend to completely ignore it. At a push I might even go as far as to argue that an experienced trader is able to trade without an edge

Conversely I'd argue that you can learn a great deal by taking random entries. I strongly advise that people try it.

Because there's absolutely no benefit in doing so.

He does not really have a choice since he's become a CTA.

At the end of the rainbow, next to the pot of gold.

Positive thinking will not help you in the slightest. It will completely destroy you. In this game you need industrial quantities of objectivity and sceptisism. Luck is important, never ever underestimate the power of luck. When you are lucky, learn to recognise that you are lucky and capitalise on it. When you are unlucky, learn to recognise it, and do something about it. Develop that attitude and you can literally make money tossing coins for entry.

If you want to make some money, its generally a good idea to ignore most of what other people say !

Easiest way is to buy it direct from TTM. Its $498. They could probably do with the sale.

The timescales look a little on the short side, but if your coming at this with a strong poker background, and an understanding of how an edge can be exploited, your already 90% of the way there. You can probably achieve something worthwhile in 2-3 years. I'd stay away from the books and seminars.

Cynthia Kase is one of the few technical analysts that I have any respect for (she's probably the only analyst I respect). I highly reccommend her material.