Re: [dlevels] Re: HRC, identify the pattern...

[Neal Hughes <neal@xxxxxxxxxxxxx>]

Hi everyone, see notes below... This is really important. It contains two
key points
that can improve your trading:

1) Statistics can be misleading. We do not use the most common 
risk/return statistics to decide whether to take a trade. To know our

true risk we need to know our true risk, not the possible maximum
risk... The most common risk-return statistics will cause you to
choose some of the worst trades, because they encourage you to
have your stops too close. Ouch! This is not DiNapoli trading..
(This
comment is a generality, unrelated to what Kevin is doing, it's a comment
on
how many other traders use statistics). 
2) How to handle a B&B when it's not working right (advanced exit
strategy).

see below....

At 07:30 AM 10/26/00 -0400, you wrote:
 

There is a logic fault in the above statment. I challenge members 
to
find it.. Hint: yes the statement is true, but it doesn't reflect
what actually happens... 
 
Hi Neal,
 
 
One logical flaw is that it only represents the perfect situation of
having the B&B turn precisely on a .382.  The further it turns
beyond the .382, the smaller the ultimate profit and the smaller the
risk/reward ratio.  
 
 
 


Nope!

The logic fault is that we do not get out at the .618 support in most
cases. More often, when a 
B&B fails, you get out near break-even. See the PAYX discussion
below. So generating risk
statistics based on a losing exit at .618 are invalid. Another
interesting option is to actually
ENTER A NEW TRADE (which is what I did) at the .618 support, so that you
actually make
a profit in those minority cases where a B&B .382 support doesn't
hold.



Another hint: see the failed B&B on the
daily PAYX chart earlier this month.. Anyone jump in here, any
ideas... Look at the PAYX chart, how would you have made money
on the failed B&B?? 
 
I'm trying to dig a failure of a flag or pennant out of the
formation, but I can't find one for which I couldn't shoot holes
in.  A "failed grail" type trade comes to mind.  I
know you've mentioned reversing on an apparent B&B failure (I guess
failure would have to be defined).  
 
It didn't quite make an OP vs. the 10/6 low, but I guess a bit of
profit could have been taken via a COP.

No, that's not what I was looking at... The PAYX chart was a B&B
Look-Alike (thrust was poor). The .382
support didn't hold within 3 bars etc (see approximately 10/05/00)... So
if it were traded as a B&B and it
failed.. Where was your exit?? How does this change your risk-reward
statistics?



 
 
 
I'm not going to say how I did on PAYX, 
and I know another DiNapoli trader who said he did very well
on that trade too... And yet the B&B failed...


The result is that B&B's should be low risk and very profitable
on
average... This is a very important issue.. Figure it out... 
>>>>


Been working on this for months.  Page 90 shows a stop quite
distant from entry relative to the profit target.  Maybe I just need
a better defintion of "low risk".
 
I'll keep figuring...


Yes, your stop is not your measure of risk most often. Your stop is the
measure of
your POSSIBLE risk but not your actual most common risk. 

Your actual risk when a B&B doesn't hold at your entry is usually on
the rebound back
towards break-even. A (usually) low-risk trade, provided the leading
THRUST is good
as the B&B forms.. This is what is meant by
"low-risk".

Also... The PAYX chart had CONFLUENCE and AGREEMENT near the .618 (from
memory
I think it was near $48 area). If we want to use statistics, we need to
take into account the
presence of these strong support areas, to calculate our real risk/reward
ratios..  Another
factor is whether a "minesweeper" entry is used... 

I'd love to see some good statistics, but with the variables involved it
would take a LOT of work..

Heck, I don't need the stats, if I just trade the rules... Once you have
confidence, the statistics
aren't needed? But they are interesting I suppose. Entertainment.. 

Thanks, great email topic.. 

-Neal.



---
DiNapoli Fibonacci techniques -
http://www.fibtrader.com









eGroups Sponsor




<img width="468" height="60"
  border="0"
  alt=""
  src="http://adimg.egroups.com/img/9620/11/_/_/_/972956248/funfreasyCompute468x60.gif";>









To unsubscribe from this group, send an email to:
dlevels-unsubscribe@xxxxxxxxxxx