Not interested in a debate of Fibonacci's usefulness, but I'd like some input about a specific question I have. Considering this scenerio; You identify a long term trend, it begins to pull back. You apply your fibonacci retracement lines and the underlying bounces perfectly off the 38% line and continues the trend to new highs. After trending up new highs it begins to retrace again. (here comes the question) Would one consider this an extention of the original trend and simply elevate the 0% line up to the new high, OR would one readjust the entire scale to begin at the lows of the last retracement and apply the fibonacci lines to this last leg of the trend? If it's the former, it seems you could go back for eternity on some markets that have been trending upwards for years or decades. A good example would be the current gold trend.
I think that's one of the issues with things like Fibs and Elliot Waves.... No one can seem to agree on which method of interpretation is best. I say decide for yourself and stick with whatever you decide. Regardless of whether a retracement is an exact Fibonacci number, 35% and 50% retracements are valid in my opinion. Check out Stan Kroll's books he was a big advocate of using 35% and 50% retracements...hard to argue with his success.
Therein lies the challenge with fibs, I use them too, from time to time. If the market begins to stair step again within your original pattern I may draw a new fib grid and see how PA is respecting it. Consider getting Timing the Trade by Tom O'brien ( and listen to his free radio show tfnn.com ). He use volume to filter fib trades for much higher accuracy. Since fibs are just a form of S/R you want to see a pullback to a S/R line on decreasing volume. Yes this takes time and practice, I am justing starting to get the hang of it. Some use EW to look at market structure and this can be helpful too, although I find it even more subjective. Good luck.
the problem with fibs are...they are useless if a % level is broken through?...what if it does not retrace?....you are left with a huge loss...QUESTION: how would you know that a 'retrace' is not going to happen and now a huge, 'break through all %' levels is now happening?
I would say delete and redraw from the last retracement low (in your example 38%) to the new high when price starts to retrace again, that way you're trading the retracement of the last part of the trend othewise your 38/50/61 would be a retracement of the whole move and like you say could go on for years without ever getting you in. Perhaps think of it like climbing a stairway? If you had to put your foot on the bottom step every time you would do yourself an injury!
Correlation, convergence across different time frames and as dandxg mentioned, volume. You would never be left with a 'huge loss' because as an intelligent trader you would know you're only dealing in probabilities and would therefore have excellent money management and be using a stop! As ess1096 specifically stated he's not interested in a debate on the usefulness of Fibonacci it's probably not fair discussing it in his thread.
ess1096....something else which just came to mind, consider the theory behind Fibonacci, the natural ebb and flow and the reasons behind it and how that relates to the behavior of market participants. A percentage of money already long takes profit at an extension therefore price retraces, they re-enter and additional money comes in from traders looking to get into the trend on a retracement so price goes higher, percentage take profit, price retraces etc etc etc. Seems logical, and it works reasonably well in a lot of markets (I trade forex using Fibonacci exclusively, have done for years).
I see Fib retracements as "just another setup to play"... not the Holy Grail nor even essential. I only consider the 50% and 61% ones and in the context of (1) "looks like a 50 Fib (or 61) has just held".. and (2) "market is approaching a 50 Fib (or 61), and I'm going to bet it holds for a fade play". If it does, that's fine. If not, just another busted support or resistance. We had one just this morning in the Dow @12665.... a 61 Fib of the prior bounce + there was a 3X bottom on the 1-minute chart. It initially held, then failed.... however it could have held then rallied back to the highs. Had it done that and in retrospect, one would have to conclude that it had "worked". There has just been another one... a 50% Fib in the NQ @ 1934... if it holds and the NQ bounces back to the top of the pattern, we can say "it held for a playable pop".
Very interesting replies, thanks. If I had to answer my own post I'd say that I do consider the major trend when using fibonacci but when looking for an entry point (or stop) I tend to reset to the current leg of the trend. Many times you'll see that a 24% retracement of the major trend might be very close to a 38 or 50%etc.. retracement of the last leg (current trend). I agree, as with any study tool, there is no holy grail. But I find that fibonacci retracements happen too often to ignore. The problem is that it only appears to be very clear in hindsight. But as it's happening it should be used just as any other support/resistance play. Gnome, if you only considered the 50% and 61% you would have missed a great entry in Gold and Oil considering the 38% retracement from the August to November tops in both. See the attached chart. Good luck all.
I would GLADLY trade all I've made for all I've missed. Lamenting "what could have been" is a waste of energy and emotion.