Fibonacci Retracement on Chart appears to be incorrect
Author: thodder
Creation Date: 8/3/2009 11:53 AM
profile picture

thodder

#1
Using a chart of .SPX in WLP 5.4, I snapped a Fib Retracement from 1,576.09 (10/11/2007) to 666.79 (3/6/2009). The Fibonacci Retracement lines displayed by WLP are: 38.2% = 927.21; 50.0% = 1,025.69; and 61.8% = 1,134.64. I recently got an email from a friend mentioning that we were close to the 38.2% level at 1,014.14. This surprised me as it differed greatly from the WLP value I had -- 927.21?! When I did the calculations manually, I got: 38.20% = 1,014.14; 50.00% = 1,121.44; and 61.80% = 1,228.74. This matched my friend's calculations.

Why are the Fibonacci Retracement calculations off in WLP 5.4? When I put the cursor on the Fibonacci Retracement line, I can see the values of the end points are 1,576.09 and 666.79, so the line is correct; but the retracement lines are wrong.
profile picture

Eugene

#2
Switch to Linear Axis from the Semi-Log view.
profile picture

Cone

#3
It's a good find though. Please create a Support Ticket for follow up. The lines should be drawn w.r.t. the log or arithmetic scale selected.

Edit:
Now I see what's going on.. the values are not calculated arithmetically when using the log scale. It could be by design, but we need to at least document.
profile picture

thodder

#4
That's closer but it is still off.

My results with Linear Axis (instead of Semi-Log):
38.2% = 1,014.72; 50.0% = 1,122.68; and 61.8% = 1,229.41.

The values should be:
38.20% = 1,014.14; 50.00% = 1,121.44; and 61.80% = 1,228.74.

Why should the scale matter when calculating the Fibonacci Retracement lines?
profile picture

thodder

#5
Sorry crossing posts. I'll create a support ticket.

(P.S. It would be nice if a user post could be converted to a support ticket.)
profile picture

Cone

#6
That's a good thought, but we still need the wizard info filled out. For the text of a ticket like this one, just copy and paste the url from above.

QUOTE:
Why should the scale matter when calculating the Fibonacci Retracement lines?
It shouldn't (as far as I'm concerned), but whoever programmed the drawing tool appears to have been basing the calculations on the actual distances - maybe even based on a pixel reference (it seems) - instead of always calculating the arithmetic values and then plotting.
profile picture

thodder

#7
For creating a support ticket from a user post, have a "Create Support Ticket" button from this post. The functionality would populate the description info in the wizard with a link back to the user forum discussion, and the user can fill in the other support ticket stuff.

I created a support ticket suggesting this. Some one can close it if they disapprove of the suggestion.
This website uses cookies to improve your experience. We'll assume you're ok with that, but you can opt-out if you wish (Read more).