There seems to be an apparent contradiction between the first and second drawings of Fib Retracements lesson, or is this my not grasping the logic? (sorry cannot post a link yet as haven’t earned my pips).
The first is an uptrend, so draw at swing low to swing high. So far so good. Draw from the bottom of the bullish candle to the top of the bearish candle… Ok… That all makes sense. The second of each pair is the actual swing in price…
Second drawing for downtrend… The retracement is drawn from BEFORE the swing actually occurs in the first of the ‘swing pair’? Line from Bullish candle to bearish again. Why is this? I would have drawn from 1.4180 to a little above 1.3854 (candle before the bearish) or even the bullish candle after?
Please explain this to me as it has a bearing on support and resistance. Thanks d:56:
If it is, you might need to rethink your understanding of fibs.
You would use the highest point in a swing high or low.
All other factors of confluence notwithstanding.
From your post, it sounds to me like you are trying to get a fib line, and an S/R point to line up. Sometimes they do, sometimes they don’t.
S/R is not really a hard line usually. It’s a zone. Both sides will probe around, and there can be breaches, but usually it’s a daily, or weekly close above or below the S/R point that confirms it.
Oh! OK So that’s how it works! Thank you for clearing that up.
I wasn’t consciously trying to get fib and S/R to ‘line up’. my understanding was that fibs could act as S/R, in conjunction with other indicators such as SMA, stochastics all backing up the ‘evidence’ as it were. I understand that they are more zeroing in zones. But they do help to set stops and limits, right - “all other factors of confluence notwithstanding” (by which I think you mean what I have said here, but please correct me if I am wrong!)
Thanks for your reply. d
It’s true, 50 is not a number in the Fibonacci sequence but weather or not it works in trading is a different issue. BP should mention that in the school so people don’t misunderstand…I don’t know if they did or not.
I do not deny 50% is surely not to be taken for granted as it is a powerful level and why it is on all fib tools. But I just thought I would point that out since the topic was about contradictions in drawing fibs. as thats the only contradiction I see in that chart.
Just to clear it up a little further. Actually its not phi or 1.618 or 61.8%. It is phi is equal to 1.618 and the inverse of 1.618 is .618. 1/1.618 = .618. The math is below. This may be confusing for some but explains the numbers.
Hey this is the ‘Newbie Island’ thread here remember! OK there may some mathematical genius newbies I guess, but I couldn’t help blaspheming in incredulity at the geekishness of this response! ; D Obviously genius but are you a little mad too?! lol
This may not be the right place to ask this question but I can’t find a suitable thread.
I have written the fibonacci objects into a ColorZigZag indicator and have come up against the scaling problem of the fibonacci arc.
The information says that this is a ratio between the number of bars and number of pips. If any of you have used this tool you have experienced having to continuously adjust the scale for different timeframes and pairs.
I have put in a big array worked out by trial and error to keep the scale right but would like to know how to calculate it for all scenarios.
Huh? What is hard to understand or ‘geeky’ about that? This is simple high school maths (inversion), something you should have learned when you were 15 or 16 years old.
No need to tell them anything … the 50% level is simply the mean of the 38.2% level and the 61.8% level, similar to the sub-levels or mini levels many Pivot Point tools use.
Note that 38.2 + [B]11.8[/B] = 50 + [B]11.8[/B] = 61.8.
In addition, 11.8 is exactly half of 23.6%, which is the first Fib level.
Notice anything there? ^^
No, they didnt mention it, as far as I know; and again, no need to.
23, 38, 61, 76, 138 and 161 (the major Fib levels) aren’t numbers in the Fibonacci sequence either … the Fibonacci retracement and extension levels used for charting aren’t sequence numbers, but are calculated using numbers of the original sequence.
“It’s true, 50 is not a number in the Fibonacci sequence but [U]weather[/U] or not it works in trading is a diferent issue.”
I’m going to be a right clever **** here - sorry in advance.
Whether the weather be fine,
Or whether the weather be not,
Whether the weather be cold,
Or whether the weather be hot,
We’ll weather the weather
Whatever the weather,
Whether we like it or not!
To be strictly correct the Fibonacci sequence doesn’t have to be 0-1-1-2-3-5-8-13… or its inverses but any squence of numbers that are the sum of the previous 2 (or their muliples, inverses, additives, derivatives etc.)
Yes, you are correct, the 50% level certainly works for all sorts of reasons. Just accept it.
The squence (or sequence!) of numbers provides a closer and closer approximation to the fibonacci ratio. This may explain why the fibo time object doesn’t appear to work very well; it starts from 0-1-1-2 which does not give the 0.618 ratio that we are trying to approximate. 13/21 and upwards is more accurate.
It may be worth writing an indicator that works out the subsequent fibonacci sequence of bars from the selected ones. For example, if the selected range was 13 bars then the next point would be 21 bars. If we selected 18 bars say then the next point would be 18 *1.618 = 29 etc.
Come to think of it I’m a bit dissapointed in myself at this point i.e. I never ever thought to make the distinction or the observation that the NUMBERS IN AND OF THEMSELVES mean nothing i.e. it’s the RATIO.
This link is pretty cool … I’ve bookmarked it for further reference.
While it is not essential for trading purposes to fully understand how indicators are actually calculated, it surely doesn’t hurt to know as much about it as possible; this way, tweaking them becomes easier, as one realizes what the change in parameters actually effects.
Thanks, Dale.
I have similar question–in order to be able to mix price values (y axis) with date/time values (x axis) we need to be able to convert price movements to number of days or vice vesa. One old fashion way is to find and use slope of major trends. I did that and for AUD/USD is around 0.00085-0.00083 per day, but I need to know how the experts do it and how the X/Y scale is choosen or set in charting software which have Fibinacci Arc option that can be added as drawing. Did you get any response?, or anybothey please can help me in this matter. By the way I have Ph.D./Masters degrees in Civil Eng. and Computer Science and I am very good in Math. Thanks.
