|
PureBytes Links
Trading Reference Links
|
To my 11/24 post about divergence I should add the following.
It is essential to know what to expect from divergence. Chances for
reversal are high on double top/bottom or W.5.
Otherwise, most often, expect a correction to MA, e.g. 60 period
SMA in any time frame tickbar charts. Once market retraced to that
MA, the requirement for direction change has been satisfied and the
trend can be resumed.
A very strong, trending market makes an exception. It will ignore
divergence, will form double and triple divergence, most often making
double top/bottom before retracing to that MA or reversing.
With these qualifications, divergence is a great tool to use, if, however,
after divergence, you always expect trend reversal, no wonder you'll say
divergence doesn't work.
Jan Philipp
----- Original Message -----
From: "Vince Heiker" <tachyonv@xxxxxxxxxxx>
To: "mike ball" <thinkpad600e@xxxxxxxxx>; "Omega List"
<omega-list@xxxxxxxxxx>
Sent: Monday, November 24, 2003 7:35 PM
Subject: RE: My Take on Diversion (edit: divergence)
>
> Tom Bierovic studied divergences extensively a few year back. He found
> divergences to be no better than random at predicting price direction. I
> think, but am not as sure, that Larry Williams found the same thing in his
> research, done before or about the same time as Tom's, if I am recalling
> correctly.
>
> For my part, have found divergences worthless over the last decade+.
>
> Am not saying that the following is the case on your part, Mike, but
> sometimes each of us see things in charts or too-small samples of charts
> that do not cover all kinds of markets - that are only what we are looking
> for - we each tend to not see things that disagree. E.g., I was recently
on
> the hot trail of shortable patterns that I was sure that I saw before
during
> and after earnings projections/announcements, following 52 week highs -
that
> no one else seemed to have seen....for a couple months...only to discover
> that the patterns were too unreliable, were trumped by other events &
market
> conditions too often to be tradable.
>
> So, be careful...Tom Bierovic does good research.
>
>
> Vince Heiker
> Flower Mound, Texas
>
>
> -----Original Message-----
> From: mike ball [mailto:thinkpad600e@xxxxxxxxx]
> Sent: Sunday, November 23, 2003 8:39 PM
> To: DH; Omega List
> Subject: Re: My Take on Diversion (edit: divergence)
>
>
> on Divergence and Reverse Divergence...
>
> There are many metrics within which to consider when
> trading this setup in real time. Yes, as many have
> pointed out it is not an easy task to recognize
> validity of these.
>
> However, to those that have simply dismissed them as
> useless, I most definitely can raise my hand as one
> that is in opposition to your opinion.
>
> I don't discount the validity of your research
> methodology as I have no way of discerning the
> appropriateness or usefulness of the
> "number-crunching" that I'm sure you've done.
>
> The realizations that trading the markets intraday
> as I'm sure most on this list will agree is that No
> ONE indicator alone seems to get the job done.
> However, a "very good" indicator with a few "other"
> parameters defined by filters or secondary indicators
> validating the primary can be a trading weapon of
> considerable validity and repetitive usefulness.
>
> A very powerful setup that I use on a daily basis and
> is quite pronoucned at least a couple of times
> throughout the trading day is the concept of reverse
> divergence in a HRTBC(Higher Reference Tick Bar Chart)
>
> Most precisely validated by the dynamics that are
> occuring in lowest reference Tick bar chart I view to
> take my entries from.(100TB)
>
> Whereas to say that a trader may be considering the
> 200TB chart as the HRTBC and waiting for a resumption
> in trend(up), finds that price nose dives
> creating the onset of what looks to be the initial
> forming of Rev. Dvg. in your Stoch/RSI indicator(s)
> accompanying "Filters"; High Tick Vol., potential
> piercing the 3rd std. dev. on the BBs with acute force
> directly into a "significant" 2B and/or Trendline,
> price congestion level take your pick, only to be met
> with instantenouse reversal bars
> or extremely evident "market hesistation" never
> allowing the primary moving average of the HRTBC to
> be violated, could perhaps deduce that a resumption in
> trend could likely be in the making.
>
> Codifying a Divergence Engine correctly or
> appropriately within a given market may be something
> that is of on-going neccessity and perhaps not worth
> the maintainence costs involved if the given market
> from which the attempts at the recognition of
> divergences are ultimately producing trades that are
> statistically ending in very little derived point
> value. A.K.A. the trade isn't giving you enough Bang
> for your Buck.
>
> However, what I find as quite useful in my daily
> trials and tribulations at trading this particular
> setup of mine is simply this.
>
> If in fact the reverse divergence signals that I'm
> playing off of fails. Then what I do expect from
> market and happens quite often enough to extract the
> realization of profits is that the market WILL go the
> other way, once the divergence is "taken out"
>
> Some might say that an indicator that is Overbought or
> Oversold stays Overbought or Oversold. Yes, that is
> true is some cases, and in those that it is Not...
>
> Well there you have my setup.
>
>
> cheers-
>
>
> mike ball
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> --- DH <catapult@xxxxxxxxxxxxxxxxxx> wrote:
> > > However the
> > > one type of divergence that seems to have some
> > predictive value is
> > > intermarket divergences.
> > <snip>
> > > I found it was much more difficult to implement
> > > real time than in hindsight
> >
> > Bingo! As Mark correctly points out, even in John's
> > cherry-picked
> > charts, he conveniently ignores several turns of the
> > McOsc and draws his
> > trend lines on the "real" turns. The eye is very
> > good at recognizing
> > these patterns after the fact (too late to trade)
> > but we don't have that
> > luxury when trading the right-hand edge of the
> > chart. There are too many
> > fake-outs realtime to make it a very useful tool.
> >
> > And yeah, been there, done that, lost money, got the
> > "I survived"
> > t-shirt, moved on.
> >
> > "If it works, code it and prove it."
> > (Paraphrase of the philosophy of my friend and my
> > first mentor, Mark
> > Brown. I forgive him for accusing me a being a
> > closet astrologer. :-)
> >
> > --
> > Dennis
> >
>
>
|