Using time targets/Bar Counts (EW if practical) and applying fibo price levels, with experience, one can trade with these simple tools applied to any timeframe, defining the trend, end, beginning, changes when the 3 lines meet/intersect.
that's the exactly the issue - they work for me so well but i cannot explain what is the logic behind them. i am forced to use them consistently because of their immense value to my trading, but i cannot explain why this tool i use works. and that bugs me.
Ahhh Grasshopper... they work because other technical traders see them. If *enough* players trade on them with enough volume, the market turns. Simple as that. Isn't it amazing how your financial success depends so heavily on how well you play this *game*?
I have tried to stay out of this discussion but have to jump in here. Drawing trendlines can be turned into a mechanical exercise. I've heard several people say they are open to interpretation, user discretion, etc., however I have coded up autotrendlines based on definite rules. They draw the same way each time based on the rules that I have given. There is no user discretion in this program. If you can describe how you draw your trendlines a competent programmer should be able to turn that into a set of rules a computer can interpret. I think the main problem is that many have not been able to fully define the way they draw trendlines down to a mechanical equation, or may be unaware of how they are doing it. There may be better rules than others in this regard. I think that John Merchant brings up a perfectly valid point. The trendline rules he has tested don't make any money. Ditto for the ones I've coded up. Someone may have profitable trendline rules, but I doubt anyone on here has really coded up profitable trendline rules because (and this is only my opinion) technical analysis tools like trendlines don't do a very good job of telling us if or to what extent the market's current trendiness is going to continue in the future. However, it is possible that someone has coded up a set of rules that do test profitably, and is using those rules as the basis for a trading system, and is very intelligently either openly discouraging traders from using trendlines or keeping his mouth shut on the subject. My own guess is that such a system would need to have a filter to prevent it from trading during times when the market has a low probability of bouncing off from the trendlines. But, I'm also not spending my time searching for the golden trendline rules that would test profitably because I think there are other things that work better and that's where I'm spending my time. I'm now going to say some things that are probably unpopular. To those who say they require discretion or don't quite know how they draw their trendlines but that say they use them profitably, I'd say first if you are able to draw your own trendlines then it is possible to mechanically describe them in a computer program. If it isn't possible to describe what you're doing then your trendlines are inconsistent and as such are not going to be able to produce consistent results. As far as I'm concerned the easy way is to say "it can't be done" and "it requires discretion" and to leave it at that. Coding, testing and evaluating results is the hard way to go about things. And now to the point that is so often made with regards to a tool like trendlines. The same has been said about moving averages, pivot points, candlestick formations and on and on. It is typically said something along the lines of, "X isn't profitable on its own but it will probably work if combined with Y and Z." And of course the person who makes that statement almost always hasn't tested their intuition. The basic assumption of this type of statement is that if you take one thing that doesn't work on its own and combine it with other things that also don't work on their own, you can somehow come up with a combination that will miraculously produce profits. And then they wonder why their optimized parameters turn to mush after two weeks of live trading? This is a loser's approach, and it is so common that it is easy to see how there will never be any lack of losers in the market. I admit that I started off down that path. But fortunately on further reflection I realized that two edgeless tools don't in fact make an edge. I would say that starting at minimum with an idea that provides even a very small advantage, and combining that small advantage with other small advantages is miles ahead of the loser's approach. That is one lesson I have learned from my testing that I wished I had realized or stumbled on earlier. And one final point - I'm not saying that all the technical sacred cows are unprofitable, I'm just saying that for most of them I've not yet found a set of rules that tests profitably. If you have, then more power to you. There is virtually an infinite number of ways to describe things and dozens of rules that can be created to trade a MACD cross, for instance. But my own preference is in favor of the low lying fruit - I'll leave the heavy lifting and the endless iterations to others.
I use AmiBroker and formulas to detect Trend and Patterns are available freely . This not only draws trendlines but also looks for Triangle formations, Wedges, etc. I can get the results through a Scan too. But at the end of the day, to trade the signal, I prefer to have to have a look at the chart since I belive that a eye can detect far more than what any formula can. Cheers
I think most would say to establish a channel of some sort. I think there's probably alot of variance in how people like to draw trendlines though, beyond the 'norm' of lowest high to highest high for uptrends, etc. I know I draw my slightly different from that. I've heard that Vic Spirandeno, in "trader vic" gives some pretty good details on how to draw a proper trendline....how to tell if a test of a trendline confirms it etc... I have not read that but am looking forward to reading it soon. Anybody read that and have some insights?
Nothing, just wanted to show that all kinds of Permutations & Combinations some of which may be untradable can be got from Automated Formulas for detecting Trendlines. Hence my reasoning that its always better to view the chart with the naked eye rather than go for Automatic Scan and Identification of the same. But what the chart shows IMO is that Lu has broken out on the lower side of a formation of a Dowward sloping Channel. How one trades this is left to the traders discretion. Cheers