As of 03/21/2023
Indus: 32,561 +316.02 +1.0%
Trans: 13,994 +229.26 +1.7%
Utils: 909 22.36 2.4%
Nasdaq: 11,860 +184.57 +1.6%
S&P 500: 4,003 +51.30 +1.3%

YTD
1.8%
+4.5%
6.0%
+13.3%
+4.3%

30,900 or 33,100 by 04/01/2023
15,350 or 13,400 by 04/01/2023
940 or 860 by 04/01/2023
12,450 or 11,000 by 04/15/2023
4,100 or 3,800 by 04/01/2023

As of 03/21/2023
Indus: 32,561 +316.02 +1.0%
Trans: 13,994 +229.26 +1.7%
Utils: 909 22.36 2.4%
Nasdaq: 11,860 +184.57 +1.6%
S&P 500: 4,003 +51.30 +1.3%

YTD
1.8%
+4.5%
6.0%
+13.3%
+4.3%
 
30,900 or 33,100 by 04/01/2023
15,350 or 13,400 by 04/01/2023
940 or 860 by 04/01/2023
12,450 or 11,000 by 04/15/2023
4,100 or 3,800 by 04/01/2023
 
My book, Trading Basics, shown on the left, discusses Fibonacci retracements starting on page 50, in the section titled, "Fibonacci Retrace Stop: Deal or Dud?"
If you click on the above link and then buy the book (or anything) while at Amazon.com, the referral will help support this site. Thanks.
$ $ $
In a rising price trend, price moves up in a rise  fall pattern, often retracing a portion of the prior climb. The median retrace is 59%. The most frequent retrace values are, in order, 61%, 56%, 50% and 55%. On a cumulative basis, a third (33%) of the samples stopped declining on or before retracing half (50%) of the prior up move. Two thirds retraced less than 67%.
The numbers suggest that a stop placed no closer than 67% of the prior up move will protect your position in two out of every three trades.
New research suggests that Fibonacci retracements offer no benefit in swing trading and that probably holds true for stop placement.
I found usable patterns in 766 stocks (but additional stocks were not used because they trended downward) and found 1,956 samples. About a quarter (535 samples) came from 1994 to 2003, which includes the 2000 to 2002 bear market. The rest came from 471 stocks from July 2005 to August 2006 (a bull market).
In the test, I used about 525 samples from inverted and ascending scallop chart patterns along with 471 stocks that gave additional patterns found manually.
I measured the decline from B to C as a percentage of the rise from A to B (see the above figure). Point A is the start of the uptrend, B is the uptrend peak, and C is the retrace low.
The top five most frequently occurring percentage retraces are:
You will notice that the first (61%) and third (50%), are also found in the Fibonacci retrace list of 38.2%, 50% and 61.8%. There was no spike at the 38% retrace value in my data.
The median retrace is 59%. That means half the samples retraced less than 59% and half retraced more.
On a cumulative basis, the following list shows how often a retrace occurs.
Frequency  Retrace amount 
33%  50% 
40%  54% 
45%  56% 
50%  59% 
55%  61% 
60%  64% 
66%  67% 
70%  69% 
75%  72% 
80%  75% 
90%  81% 
In words, a third of the samples (33%) will decline less than half (50%) the distance from B to A (see above chart). Two thirds will retrace less than 67%. Nearly all, 90%, will retrace less than 81% of the B to A move.
Measure the recent swing move from A to B. Take 67% of that move and subtract it from B. Place a stop no closer than the result. That should prevent you from being stopped out in two out of every three trades. For example, if point A is at 20, B is 25, then place a stop below 21.65. The move from A to B is 5 points and 67% of this is 3.35. Subtracted from B gives the stop price of 21.65.
As a sanity check, a 3.35 point drop from 25 represents a decline of 13.4%, which is quite high. You might want to narrow your stop, but the risk of being stopped will rise if you do.
 Thomas Bulkowski
Support this site! Clicking any of the books (below) takes you to
Amazon.com If you buy ANYTHING while there, they pay for the referral.
Legal notice for paid links: "As an Amazon Associate I earn from qualifying purchases."
Washing your car to make it rain doesn't work.