[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[RT] Validity of triangle breakouts.



PureBytes Links

Trading Reference Links


In a previous report on triangles and the 
behavior of
prices on breakouts from triangles I did a 
"half-assed" job
and I apologize for it.
 
This post is a detailed and thorough 
examination of ALL the
patterns that have occurred in the DJIA 
since 1921 that can
be defined by a detection scheme using 8 
bar (day) highest
high and lowest low points to select 
"turning points". 
There is much detail on this approach 
at:  
<A 
href="http://www.theswingmachine.com";>www.theswingmachine.com
and the enclosed results derive from the 
approach outlined
on that site. 
 
The following diagrams outline two 
patterns
which (if the upper turning points are 
connected
by a line and the lower turning points 
are
connected by a line) form the classic 
triangle
patterns from which we might expect 
certain types
of breakouts.
 
 
<FONT face="Courier New" 
size=2>...A.............         
.........B........./\.............         
......../\...........\........C...         
......./..\....D......\....../\...         
....../....\../\.......\..../..\/.         
...../......\/..........\../....D.         
..../........C...........\/.......         
..\/......................B.......         
..A..............Pattern:..#2.....         
Pattern:..#8.....A>B B<C C<A D<C 
D>B       A<B B>C C>A D>C 
D<B                   
A line from A to C        A line from A 
to Cwould slope downward      would slope 
upwardand the next move         and 
the next move from TP-D would be        
from TP-D would bean UPSIDE breakout.       an 
DOWNSIDE breakout.
 
 
You can examine the  Junk1.gif  
attachment for a more
"picturesque" display of these same two 
patterns.
 
Now, what are the statistics on the 
behavior of prices 
after such patterns occur.
 
This study considers ALL turning points for 
the Dow-Jones 
Industrial index from the first of 1921 up 
to May 25, 2001.
 
If we don't really care whether we can make 
a profit on a
"breakout" but just simply measure success 
or failure to
occur if the next turning point in the 
price sequence is
on the PROFIT side of the appropriate 
triangle line then:
 
For UPSIDE breakouts the success rate 
is:  92.5%  for 200
triangles that have appeared from 1921 
until 2001.
 
For DOWNSIDE breakouts the success rate 
is:  92.5%  for 213
triangles that have appeared from 1921 
until 2001.
 
HOWEVER, if we ask a bit more out of such 
breakouts:
 
If the breakout must achieve 2% profit 
then:
UPSIDE breakout percentage 
is:     67%   and the
DOWNSIDE breakout percentage 
is:   58%   .
 
NOW, if we are realistic and want to insist 
that the breakout
must achieve a 5% move over the appropriate 
triangle line
then:
 
UPSIDE breakout percentage 
is:     36%
Downside breakout percentage 
is:   28%  .
 
The attached  .csv  file (import 
it into Excel) is all the 
raw data and the results of the 
computations.  I did not
bother with the full spreadsheet as it is 
much larger than
most of us want to fool with.
 
I now understand where people come up with 
figures like 
95% successful for triangle breakouts BUT 
even if we 
limit it to realistic breakout amounts it 
still appears
that such things as this do represent a 
significant lowering
of risk of trading when we can find 
them.
 
I would state that for such an advantage to 
exist the method
of computation of "patterns" for triangle 
would need to be
as rigorous as what is used 
here.
 
Clyde
 
 
- - - - - - - - - - - - - - - - - - - - 
-  - - - - - - -Clyde Lee   
Chairman/CEO          (Home of 
SwingMachine)SYTECH 
Corporation          email: <A 
href="mailto:clydelee@xxxxxxxxxxxx";>clydelee@xxxxxxxxxxxx  7910 
Westglen, Suite 105       
Office:    (713) 783-9540Houston,  TX  
77063               
Fax:    (713) 783-1092Details 
at:                      
www.theswingmachine.com- - - - 
- - - - - - - - - - - - - - - -  - - - - - - - 
-






Yahoo! Groups Sponsor


Yahoo! Website Services- Click Here!









To unsubscribe from this group, send an email to:
realtraders-unsubscribe@xxxxxxxxxxxxxxx





Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.



Attachment: Description: "INDU2108_UpTriang.csv"

Attachment: Description: "junk1.gif"