I agree 100% but this wasn't my point. I only meant that if someone is going to consider fibonacci lines at all they should consider all of the lines.
Sometimes the 23 and 38's hold... if it "looks " like that, OK to play 'em.... but it's been my observation that the 50 and 61's work with more regularity regardless of what might have worked in one instance. (For every 50 & 61 which worked, a 23 AND 38/50 had already failed...)
You guys seem to just draw a low to a high (or high/low) and then look at 38/50/61 levels, don't you find other Fibonacci correlations for a reason why price should be more likely to bounce off a particular level? Looking at that Gold chart 38.2 was the most probable level, not so easy to see on a candle chart but take a look at a bar chart. There's more to Fib levels than a lot of people realize, they dismiss them too quickly without understanding how significant these levels are.
Then stop starting threads about them. They were submitted to studies. They don't have anymore value than other retracements. Therefore, you are starting your own wild goose chase.
I agree with you about looking @ charts & patterns using different chart types. From a historical perspective I understand that the earliest western chart types were End of Day Line charts of closing prices. If that is true then most probably market technicians and chartist of those earlier times would have based their observations upon closing prices. That would include Dow, Elliott, Babson, Gann, Andrews and others. For quite some time I have had a problem deciding what a wave count / structure was until I used a simple line chart then it became easy. Using Andrews pitchfork on closing prices started producing more accurate and recognizable S&R. I agree with others that Fib numbers should be considered as areas of "Alert" that a possible trend change or continuation may occur. Whether you use this % number or that one is totally immaterial so long as you use it consistently. I currently own & use several different Fib tools and am running a test of them to check their accuracy on both closing prices & high to low prices (daily bars) of the G/U. I subscribe to a service that on Jan 9 posted possible price targets for Comex Gold pit cash ($959.20) and Amex Gold Bugs ($501.09). They use daily, weekly and monthly bars. It will be interesting to see how their numbers play out and which of my Fib tools is more accurate on which pricing method.
I would be interested in reading the results of those studies, do you have a link or details of who carried them out?
On both sides of the argument, you are ALL missing the point. Of COURSE fibs have no predictive value. You do NOT bet on a reversal at one. But fibs have REACTIVE value. You ACT on a reversal at one. "Nothing works all the time. Everything works some of the time." When it works, it WORKS. Fucking stupid thread!