For Jack Hershey
TGK
first of all thanks for the encouraging words, they really help especially because they come from
you. I traveled to Spain last week and so I had plenty of time to read all your messages which you
have posted in May and a few others from this and last year (there must be more than 1,000 messages
around, probably much more but DejaNews deleted older ones due their server updates).
As I can see reading your posts it seems that I am at least on the right track refer indicators and
time frames, which certainly means nothing and all at the same time depending if I can overcome the
last knots and hurdles regarding to place all the learned knowledge into action and start using it
in combination - I might have some clues but not the experience and many time I feel kind of lost
with all those opportunities, indicators and price moves!
If you don愒 mind I would like to ask you some questions refer previous posts and answers, please
forgive me when some of them got already answered last year or the year before but there are so many
messages and its not easy looking through all of them.
Jack Hershey wrote: 5/3/00 5:52 AM
To keep the time required at a minimu, I recommend that 30 minute bars be used. Start with the US
trading duration for S&P. Watch for a Breakout from one bar to another. As the bar breaks out,
enter and hold until there is a 50% retracement or a .612 or .38 retracement, i. e., use any of the
popular retracement values.
Question:
I suppose you are talking about the SP500 Futures? Breakout means when the next bar makes a higher
high than the previous bar preferably early enough to give the bar room to go higher? How would you
determinate when the next bar is a "Breakout Bar" or just making a higher high before reversing
and/or retracing too much? When I look at the 30min bars I ask myself what kind of StopLoss you
would suggest when entering because many bars are covering each other especially the SP500 does
that. So, is a 50% retracement of the previous bar tradeable or the low? In both cases someone has
to deal with a 5 to 10 points StopLoss from the breakout entry, especially when the move is fast and
the previous bar was a big one, right?
Jack Hershey wrote: 5/4/00 7:13 PM
The index is leading the cash usually and the moves are always contrarian until sync is reached. So
when the nornal spread of the prior day is achieved you will see they will move together with the
futures leading the cash somewhat.
Once it begins with the compression switching to divergence you know the breakout is coming with the
futures leading the cash by up to a minute sometimes two.
Question:
Thats something I never understood. I think I understand that an Cash Index is moved by all the
minor moves from the Stocks which are included in the Index and the percentage which is the factor
of their weights. How can be the futures leading the cash - that would mean that Stocks are bought
and sold just to keep track with the futures price?
Jack Hershey wrote: 5/6/00 7:14 PM
1.) It is a priori that no indicators work well when ythey are bridging a close- overnite shutdown-
and open. You wait until the end effects are eliminated and take the measures.
2.) Pull up the Friday 5 min chart this weekend and you can see how to do the evaluation. I
recommend that anyone who is optimizing their trading copy this chart as a reminder of how momentum
scoreing really works.
3.) On this one you optimised when the momentum took off, the MACD's started to diverge, and the
MACD histogram went long, and the stochastics came off 20% headed to 50%.
1.) Question:
I know that you don愒 give much about the 24h Globex session because there are not many traders and
volume is very thin (interesting to see price moving for 5 or 6 points with almost no volume) but
how long does it take till you trust your signals especially when Gaps are distracting the indicator
readings?
2.) Question:
I missed that thread - what contract or stock are you refering?
3.) Question:
What are your Stochastic Settings?
Jack Hershey wrote: 5/7/00 12:38 AM
I woulnd up being premature in exiting and i also got aced at the beginning of trends where my stops
were initially so low that I lost on trades.
Question:
Where do you place stops nowadays, based on what criterias. Do you also use specified ProfitLocks or
do you always wait till the market tells the story?
Jack Hershey wrote: 5/12/00 3:15 AM
- 8. Wrote books (27).
-12. designed and implemented a parallel paradigm for futures indexes that works at a pace that is
about 50 times faster than the stock approach
- I earnestly use fundamentals to find the highest quality universe (earnings and capital
appreciation and absolute leaders) of stocks. They are moving up and down by 10 to 20% five times
in 6 to 8 days every six months.
Question:
-What kind of books did you wrote?
-Where are the differences between your paradigms for Stocks and Futures? Why do you trade the
DowJones contract and not the SP500? Isn愒 the Emini a good instrument for trading multiple
contracts facing an impressive execution speed?
-When you enter a trade where is your original StopLoss? When the choosen stock is moving 10% to 20%
within 6 to 8 days and assuming that you jump in late and exit early when it goes in your favour
where is the correct StopLoss?
Jack Hershey wrote: 5/18/00 12:18 AM
I recently, one day traded 100,000 shares of a stock for a 17 point average profit during the one
day i made my exit. I traded into the high, across the high and down the side of the high that day.
I had to use 31 trades to not disturb the peaking that occurred because of the blocks being traded.
Question:
I looked at the Genzyme Transgen. chart trying to duplicate your readings of the price action and
indicators as an example but could not bring it to live especially because you扉e said you only
trade the long side!?
Jack Hershey wrote: 5/24/00 8:32 AM
1.) More cycles in more important than profit per cycle. Youcan use the compound interest formula
to do a spectrum of examples. Try as you work through them to optimize the variable pairings for
cycles and price. As you practice this optimizing by changing one of the two variables at a time
you will learn how the formula works.
2.) I mentioned to several people today that the market was not going to hold today about fifteen
minutes before the NYSE and the NASDAQ broke out down. A half hour before that I had said that the
testing time would be when it happened. Very Very few of them changed their positions.
3.) In an abstract way, these fractal pairings of one variable, price and price only based
indicators, do lead to a P/V duality with statistical significance.
1.) Question:
As english is a foreign language to me I just wanna make sure that I understand the terminology for
"CIF": FV = PV(1+i)^N where
PV = present value
FV = future value
i = interest rate in percent per period
N = number of periods
2.) Question:
What kind of signals where you watching at this time and which combination finally gave the signal
for the big drop?
3.) Question:
Due my own observation refer price and volume and as I mentioned in my original post there is no
doubt that volume leads the price, BUT how do we avoid those false signals where the volume dries up
but the market starts turning around, beginning a retracement or even starts reversing and cancels
the trend? And when do you start paying attention to lower volume bars - do you use a Moving Average
for the Volume, is it a fixed level or morelikely an eyeball thing?
Jack Hershey wrote: 5/29/00 10:34 PM
1.) Oh the signal isn't generated by most software, it is a volatility compression signal that is a
leading indicator. I have modified one of the indicators to fix it a little.
2.) The key quote.com indicators didn't get on the list here. if any one wants to raise them for an
application we can run though them.
Question:
1.) I never worked and observed Volatility Indicators yet. What kind is worth watching for a while
and which one do you use? Does Quote.Com provide a useful one or might it be possible to create one
that works?
2.) Count me in, please !
Right now I am reading some interesting articles about averaging which as a topic certainly can be
reduced to price averaging while buying/selling stocks but to me it seems that it is morelikely a
huge topic including the mixture of Futures, Stocks, Options, Mutual Funds and the opportunity to
become a "complete trader".
To cover at least a few of them I am thinking of buying TA software. As I live in Europe some of the
available programs might be limited to me and it愀 not easy to choose the right one refer all the
pros/cons and different experiences from others. Right now I am testing "FibonacciTrader" which is a
nice piece but the realtime feed is a big handicap and there are also no stock screening tools
included. I also started to test "RavenQuote" which looks promising but is still in beta testing. A
few weeks ago I tested "Byte Into The Market" which is very powerful but only EOD right now. The
stuff someone can read on several news and usergroups refer Omega products doesn愒 seem to be worth
to give it a try either - well I am stuck again :-)
Thanks
TGK
>>Von: Jack Hershey <jher...@primenet.com>
>>Betreff: Re: Useful Indicators - MA愀/MACD/Stochastic/Volume ?
>>Datum: Wednesday, May 24, 2000 3:47 AM
>>Terrific post. You have all the clues and the ingredients to develop a
>>terrific paradigm. Since I trade an allied futures index using the answers
>>you are requesting from experienced traders in your venue, I can suggest
>>that the answers are out there for everthing you want to know.
>>Best of luck to you.
>>>"TGK" <tgk...@lycosmail.com> wrote in message
>>> Hello all,
>>>
>>> I am trying to optimize buy & sell rules from my observations for the SP500
>>> and Emini market and would love to hear what kind of experiences the ex-
>>> perienced trader has made with some indicators.
snip>>>>
Jack Hershey
I looked over your post and I have divided in it the eight snips and the
last two paragraphs. When I lived in Switzerland, Canton Zurich it was good
to head south to get ahead of the seasons this time of year as well. The
sun is always a nice respite.
I am going to respond to your post in parts to begin the process more
promptly. For transference purposes either to people, markets or portions
of the world, I am always interested in the nuances that appear. Each
portion I respond to will be in the context of making money.
I am wrapping up the statement of what I do in two comprehensive freeware
renditions. Now that there are persons working in several ways on this
stuff, I find that I can produce a descriptive rendition that allows a young
person the do it manually instead of just operating off a software package
that does it automatically. I also am going to sort through my posts to
come up with the basis of TA from my point of view. My goal is to have
many people making a lot of money so they can contribute to dealing with
societies challenges as contributors of their talents. I am going to state
my views on this using an outline I made in 1982 following the publication
of John Naisbitt's Megatrends: ten new directions transforming our lives.
In 1982 I wrote down the 10 I thought were more on target. I find I was
really on the ball 18 years ago. In two steps, I am going to first create a
broad supply of money (by having lots of people make it) and secondly, I am
going to suggest where the left overs can be placed to get the jobs done
that need doing.
So from now on my responses will be better done and they will be more usable
by me to tuck into the stuff I am knocking off now. This post will be done
with 11 responses from me. Ten more to go.
Jack Hershey
I use simple mechanical systems to get people to understand the basic
concept of making money steadily and with little or no risk.
When you trade daily for 6 1/2 hours a key thing to consider is not doing
too much to make some money.
By choosing a futures index of any sort on any exchange in the world, you
have put yourself, for 6 1/2 hours a day in a place that is truly dull and
unexciting. Being there is fairly safe and not too demanding so you can
relax and repeat a few tasks over and over to make some money.
I work first with 30 minute bars to frankly eliminate any sense of urgency.
I use the prior days last bar to get the ball rolling, or I suggest you wait
until the second begins to eliminate the end effects of the market.
Here is a progression of four mechanical methods to illustrate making money
primarily and secondarily to illustrate that losses are neatly reduced more
and more as a little sophistication enters the picture. I also introduce
how in a trend you can switch to the most favorable side of the channel to
exit. Because this is very simple and mechanical there is no need to
clutter it with a stop system as yet mostly because it an index tied to the
performance of and aggregation of stocks. We can tuck stops in easily
though as a commitment to our ordinary discipline.
The four items in the progression are:
1. break out of prior bar.
2. slope pairs of bars.
3. overlapped pairs slopes
4. retracement.
Here is the progression:
1. set up a 30 bar display for a futures index.
2. enter on the breakout beyond (above or below) the prior days last bar
hi/lo.
3. hold until the current bar breaks out of the other end (from your long or
short entry) of the prior bar.
4. hold on inside bars.
5. hold on successive bar break outs in the same trend.
6. on breakout of 3., reverse so you can take on new trend trade.
7. repeat 3. through 6. for remaining bars of the day.
8. settle at end of day.
For this week, Tuesday was one trade enter long on bar 1 and exit at
settle. Wednesday was enter long bar 1, exit on 1 (loss), and reverse into
short; exit bar 2 (loss), and reverse into long; exit bar 4 (profit) and
reverse into short; exit bar 6 (profit) and reverse into long; exit bar 11
(profit) and reverse into short; exit bar 13 (profit). Thursday was enter
on bar 1 long; exit on bar 7 (profit) and reverse into short; exit on last
bar (profit) and reverse into long; settle (profit). Friday Enter on gap
open; exit on bar 4 (profit) and reverse short; exit on bar 6 (profit) and
reverse long; exit on bar 8 (loss) and reverse into short.; exit on bar 10
(scratch) and reverse into long; exit on bar; settle on last bar (profit).
2. slope on pairs of bars.
In 1. the breakouts were on a horizontal extension of the bar end. Lets
look at bar ends. In this rendition we use the last two completed bar ends
to do a projection on to the bar that is forming and note a value. If we
are long we project the lower ends, if we are short we project the upper
ends. the projection is a possible exit/reverse value to insure a longer
trend run and profit. It is an enhancement to improve the horizontal
projection of the single prior bar. Cuts losses to a good degree. Do it
for the week and see how the profits improve and the losses diminish.
3. This is an improvement on 2. I do this on a sheet that I keep on a flat
surface. You make the day (13 bars spread across a sheet and put between
each bar placement another vertical line. mentally plot each thirty minute
bar on the appointed vertical places. Good. Now the challenge comes. In
2. we improved the exits and reversals in price. Now we continue that and
also improve in time. To do this we plot thirty minute bars in between the
standard ones. Think Plot a standard one. when 15 minutes passes plot the
hi lo for the last 30 minutes. Cool you thought it through. the easy way is
to pop over to a 15 minute display and get the hi lo from the last two bars.
Okay when you have done this twice you now have a pair of 30 minute bars
in-between the standard pair of 30 minute bars. Do the projection of slope
onto the next in-between space. I use a higliner in orange for one set and
yellow for the other.
Use the projections to gain some more of the trend end and also exit and
reverse 15 minutes earlier at a superior value. this is a two for one deal
each time. there are few losses at this point and each profit is enlarged
too. we made a time enhancement here. in two we focused on a price
enhancement. Now you are beginning to see some steady profits all the time
and few losses. You also see we are watching the market so the risk of not
having a stop in is reduced too.
Oh put a stop in if you want. Slip it in outside the lines you are
drawing... Cool.
All of this is work on the right side and is neutral biased long and short.
It is a good exercise series to get a little relaxation into making money.
I mean there is a little to learn here right? Throughout this we have not
seen a nontrend...well that is because this is so insensitive that's all.
4. Retracements. after the above retracements seem a little naive..but they
do help because they may make more profits...BUT... you may go out early
too. That's the trade off. there are lots of posts here on trends.
Retracements are the biggest losers for getting you out too early. No cure
available for such strategies.
Your query contains a bunch of Q's that I will address collectively. Here
is the method. And as you use it there are no losses associated with it.
1.enter on bar 1 when Break out occurs. Log value.
2. log peak/trough value on entering bar when it and if it occurs.
3. take difference of 1. and 2. divide by two. log.
4. Get the midpoint value (50%) using the difference
5. exit and reverse at that value.
6. if price retraces to buy of 5. exit and reverse.
6. log break out value (current bar extends beyond an end of the prior bar)
if and when it occurs. repeat 2 through 6.
This is a left side of trend channel technique, i. .e., the extreme side of
the channel instead of the mean side of the channel (centering tending
side). Either you are making money here or scratching.
Final note: You can use 3. or 4. and earn a steady income usually in the
low six digits with a small # of contracts like one. to do more work in a
given day, just change from the 30 minute (middle one of seven) to the next
fastest one (15 minute). This is a doubling of income type move.
You can see by looking at what you are doing, you can do successive
enhancements to improve it. I use this with those I mentor to ease them out
of trading stocks and into day trading in a simple market. It is not an
optimum method at all; it is just to overcome the psychological barriers
that you can see abound here.
When you read the post posed, you see all the frightening unrelaxed
considerations jumbling upon oneanother. this makes looking at a chart very
unsettling. By imposing a few rules to be carried out where only specific
data is being considered, the market itself becomes understandable and
manageable. Coming up with hundreds of tools and enhancements doesn't just
pop into the space at all but if you just iteratively work from here to
there it does shape up after a while.
Jack Hershey
I did a post today that hits on this directly. It is entitled TA
milestone. There is a reference in it to a post I did that subtends some
greenfinch trivia as well.
The key thing to get straight here is that there are some skilled folks
around the financial industry. They are skilled enough you can count on
them. Today, as usual, they were leading the cash market with their futures
positions. what helps here is to understand that the futures market is open
sooner too. The hugely paid fund managers I think are totally out of the
running when it comes to anticipating the markets as well as the futures
folk do. I know the fund managers have so much money to work with that they
just can't get it done very well. But maybe there are some other ways to do
it. The traders of the 30 industrial stocks are not doing their thing to
match the leading futures index. The futures guys are just real sharp and
you can take your lead from them.
Jack Hershey
Hi TGK,
The index I referenced to was DJ0M and the best Stochastics write up and
stting I know of thats convenient the analyze is the Clearstation.com
setup. They use a 5 3 3 set up. Be sure to read their education part
because they do define how the set up the system.
Jack Hershey wrote: 5/6/00 7:14 PM
> 1.) It is a priori that no indicators work well when ythey are bridging a
close- overnite shutdown-
> and open. You wait until the end effects are eliminated and take the
measures.
>
> 2.) Pull up the Friday 5 min chart this weekend and you can see how to do
the evaluation. I
> recommend that anyone who is optimizing their trading copy this chart as a
reminder of how momentum
> scoreing really works.
>
> 3.) On this one you optimised when the momentum took off, the MACD's
started to diverge, and the
> MACD histogram went long, and the stochastics came off 20% headed to 50%.
>
>
> 1.) Question:
> I know that you don´t give much about the 24h Globex session because there
Jack Hershey
> I woulnd up being premature in exiting and i also got aced at the
beginning of trends where my stops
> were initially so low that I lost on trades.
>
> Question:
> Where do you place stops nowadays, based on what criterias. Do you also
use specified ProfitLocks or
> do you always wait till the market tells the story?
>
>
design I made look up table for the stop offset (one of six different
tables) to set the stop with respect to the current price on a periodic
basis (the interval is also from a look up table). For five market
conditions, in minutes and DJ0M points the pairs are: 12, 15-20; 9, 20-25;
7, 20-30; 6, 40; n/a,n/a. Add to this 7 DJ0M points for scalping. The five
corresponding market trading fractals are: 5; 5; 15; 15; and 30 minutes.
What I have gotten by interviewing several persons who trade other indexes
than DJ0M is that I can put sets of lookup tables in for each index easily.
I am also asking that a gismo be put in place to periodically refine the
look up tables in a quasi automatic way, i. e., if there is a change
proposed that is relatively different then the computer seeks a manual
review of this. I have used this in radiation therapy applications where
the medical prescription operating histories from the staff were monitored
to see if any mavericks were appearing. The most significant function was
to determine erroneous prescription combinations for individual clients and
to eliminate"more is better" thinking. Prescriptions were rejected in ten
possible categorical areas as a means of enabling the equipment operators to
have a basis for negotiating with the prescribing medical staff.
So here, the offset is determined by the sentiment character as determined
by four sentiment values ( volume fractal coefficient, MACD volatility, MACD
offset, and congestion status) and market pace as determined by six inputs
that yield a score with a five point range from 1 to 5.
I compiled he data for this manually over about two years to understand
trends and market sentiment and pace. So there is a way to do this manually
that is quite easy using a visual measurement and logging index values for
each micro formation that occurs on the 1 min fractal and simply staying
offset from the current micro formations a distance of several formations.
Here is the easy logging scheme and how to determine how many formation
values to lag behind. From fast paced activity to slow paced activity stay
back 3, 4, 5, 6, and n/a micro formation values. As you get mentally
calibrated you can shift one formation value one way or the other. A
formation is a hitch, stall , or retrace, or a no trade minute. Simply log,
in order, the value where the formation occurs. Some times you will have
to cross out some and then reenter them again. The other half is this.
Every 12 to 6 minutes (use 9 nominally you must change the stop. If you
can't, then you exit at market and sideline right then and there. The trend
has ended.and a nontrend has begun.
This stuff above makes you more money than anything else. And also cuts
losses more than anything else.
What you are trying to do is stay in trends and get out when the nontrend
commences. It is not a thing where you keep figuring out how it all looks.
Just stay offset with the stops the market gives you the values for. When
it stops giving you stops the trend is finishing..then you cant' change the
stop and you leave on the high side of the "noise" of the short swings that
define the volatility every few minutes. It a very smooth approach that you
share with the market.
Jack Hershey
I write in the environmental and energy field. Mostly scientific and
technical stuff in emerging fields of endeavor.
I known for my revision of the think tank Delphi Process into what is now
called the "Modified Delphi Process". A problem solving technique. My
first book on a Ford Foundation grant was a seed grant to deal with "what's
wrong out there". It became known as the environment and the first
legislation was PL 92-500, the clean water act.and subsequently the EPA was
established... I was a small part of that. I helped sew the first
environmental flag three years before earth day...
I am well know as a futurist and the econometric modeling have done has
been adopted by the world bank as its model for a developing nation. As an
architect,. I have focused on the harmony of the built and natural
environment, the AIA has published a two volume book on that subject which I
wrote. I have declined faculty positions in departments of psychiatry, but
I am recognized with respect to learning theory.
In the course of working for three Presidents, I have written several other
books mostly on energy ( the alcohol movement) and recoverable energy from
waste, and national energy self sufficiency.
The copyrights to all books I have written have been contributed to not for
profit 501 (C) (3)'s.
The paradigms have a core of common precepts. Both are strictly based on
the P,V relationship and the corollary I feel completes it. To whit:
If the Volume trend is increasing, then the Price trend will continue ..and
If the Volume is decreasing, then the Price trend will change. since this
neglected on e aspect of volume I added a corollary. If the volume is
relatively unchanging, then the Price will drift slowly downward ( 4 o'clock
drift.)
I have utilized Boolean algebra to formulate this in software for several
purposes. for stocks I automatically prepare short lists of quality stocks
that will soon be commencing their upward cycles. I believe prediction does
not work so I have turned to anticipation. I use my orientation to "know"
what is going on ahead of the opportunities that are coming. So my
paradigms proceed the market action and I trade ahead of the herd and the
herd drives me to optimum profits. One item that is nice to use is scoring
part of the stock paradigm. By scoring a cycle you know where you are,
what's next, and the pace of the cycle. It is referred to and copyrighted
as "getting tomorrows newspaper today" It was named by those I mentor.
The stock paradigm on a transference level (others using it whom I have not
met) yields 11.1% per cycle with and average of 6.6 days per cycle. This is
using self originated software from math I provided, and only monitoring the
market as much as 2 hours in the morning and being employed at a
professional job. Olivia and Natalie in other nearby posts is another
example.
The commodities paradigm operates at about 50 times the pace while day
trading.
I do look at the other indexes and I can suggest that when colleagues are
trading them and I am present, I am able to go with the flow quite
effectively. I followed the DJIA since 1957 and I can predict it absolutely
cold, absolutely on the money. So I feel its fun to do the DJ0M in an
anticipatory manner. they recognize my voice, I like the sound of the floor
and they let me use short hand on the C&R's and they call me if I forget an
update when guys are coattail trading me.
Trading on a machine requires more than punching the redial button and
working with a guy who has my drift on making money.
Stop loss on stocks. I pull up RT data for 5 min bares and check out the
institutional trading. I check the dumbest fund managers plays which are
causing drwa down. I use that draw down plus a 1/4 to 1/2 more depending
upon the price of the stock. use 1/4 in the 20 dollar range. I use EOD
charts and graphically check out how good I'm doing. when the day count on
the cycle is nearing its end I tighten up. finally I monitor the prorata
volume relative to the prior day and have three volumes on my log: DU, FRV,
and peak. when the prorata volume is fading I go out at market no later
than 2 hours into the trading day.
The cash goes into one of 7 stocks that is moving up at a rate faster than
that of the stock I am exiting. I want to get out to get into a better
stock. I am walking up the up escalator. I tell them to buy the stock at
market with the money. They tell me the # of shares and the price they got.
I set my commission at the desk. it is usually three times the standard rate
if I win (throwing a bone) and at paper cost if I am scratching for some
reason like I made a mistake.
Jack Hershey wrote: 5/12/00 3:15 AM
> - 8. Wrote books (27).
>
> -12. designed and implemented a parallel paradigm for futures indexes that
works at a pace that is
> about 50 times faster than the stock approach
>
> - I earnestly use fundamentals to find the highest quality universe
(earnings and capital
> appreciation and absolute leaders) of stocks. They are moving up and down
by 10 to 20% five times
> in 6 to 8 days every six months.
>
> Question:
> -What kind of books did you wrote?
>
> -Where are the differences between your paradigms for Stocks and Futures?
Why do you trade the
> DowJones contract and not the SP500? Isn´t the Emini a good instrument for
Jack Hershey
Rough it out this way. At any point in the day think of me as trading 10% of
the total volume. get the trades on a list and circle one and wait until
ten go buy of equal size and do another one. You will have bout half of it
done when the peak ends. Then just ease on throughout the rest of the day.
I determined it was going to be the peak day the night before so it was an
easy day. Picture 2.7 million dollars accumulating in the account with a no
new buy order. The market was peaking that day too. I did not want to buy
into a falling market.
My actions were to not cause a draw down in price. that is why I used a 10%
max at any given time and the requirement that the block size be equal or
less than the blocks showing.
> Jack Hershey wrote: 5/18/00 12:18 AM
> I recently, one day traded 100,000 shares of a stock for a 17 point
average profit during the one
> day i made my exit. I traded into the high, across the high and down the
side of the high that day.
> I had to use 31 trades to not disturb the peaking that occurred because of
the blocks being traded.
>
> Question:
> I looked at the Genzyme Transgen. chart trying to duplicate your readings
of the price action and
> indicators as an example but could not bring it to live especially because
you´ve said you only
> trade the long side!?
>
>
Jack Hershey
Take the trouble to use the compound interest formula on this stuff, then
write them a letter saying they need their heads examined. Don't get into
the sophomoric world where a bunch of guys operate who can't get the idea
that the market can easily be timed operate
.
Jack Hershey
See below in body of yourtext for the comments
jack Hershey wrote: 5/24/00 8:32 AM
> 1.) More cycles in more important than profit per cycle. Youcan use the
compound interest formula
> to do a spectrum of examples. Try as you work through them to optimize
the variable pairings for
> cycles and price. As you practice this optimizing by changing one of the
two variables at a time
> you will learn how the formula works.
>
> 2.) I mentioned to several people today that the market was not going to
hold today about fifteen
> minutes before the NYSE and the NASDAQ broke out down. A half hour before
that I had said that the
> testing time would be when it happened. Very Very few of them changed
their positions.
>
> 3.) In an abstract way, these fractal pairings of one variable, price and
price only based
> indicators, do lead to a P/V duality with statistical significance.
>
> 1.) Question:
> As english is a foreign language to me I just wanna make sure that I
understand the terminology for
> "CIF": FV = PV(1+i)^N where
> PV = starting capital value
> FV = results at end value
> i = interest rate in percent per period
> N = number of periods
>
See above I changed it a little.
> 2.) Question:
> What kind of signals where you watching at this time and which combination
finally gave the signal
> for the big drop?
Formation break outs with prior low volume. and then high volume on the
break.
>
> 3.) Question:
> Due my own observation refer price and volume and as I mentioned in my
original post there is no
> doubt that volume leads the price, BUT how do we avoid those false signals
where the volume dries up
> but the market starts turning around, beginning a retracement or even
starts reversing and cancels
> the trend? And when do you start paying attention to lower volume bars -
do you use a Moving Average
> for the Volume, is it a fixed level or morelikely an eyeball thing?
What I am suggesting takes all these things into account. What a person
does is develop a method of making money and use it... When little things
happen on the charts are just the flavor of the moment and they are all
taken into account. Sometimes people override their approach as they see
this and that. These are time when by deviating from a plan, you are
paying tuition to learn not to deviate from the plan.
>
>
> Jack Hershey wrote: 5/29/00 10:34 PM
> 1.) Oh the signal isn't generated by most software, it is a volatility
compression signal that is a
> leading indicator. I have modified one of the indicators to fix it a
little.
>
> 2.) The key quote.com indicators didn't get on the list here. if any one
wants to raise them for an
> application we can run though them.
>
>
> Question:
> 1.) I never worked and observed Volatility Indicators yet. What kind is
worth watching for a while
> and which one do you use? Does Quote.Com provide a useful one or might it
be possible to create one
> that works?
Work through the Conners-Haywood approach. i repaired it a little so it is
shifted and a better signal is generated.
Everyone does Chaikin's as well
Jack Hershey
Get something you can analyze with and that has a data feed. For this news
group I stick with quote.com mostly.
I am going to fix up my operation soon. I'll let you know what they set me
up with. I have some free spending hot shots who keep my machines sharp.
Focus on a european version maybe.
Jack Hershey
Use the symbol SPX.X by clicking at the upper right corner. Sorry there was
a reference in the post I reponded to about the S&P I thought.
Don Cameron
and possible ambiguity.
Consider the following situation for method 2. You are long and for
the current bar we have a higher high and a higher low that the
previous bar i.e. an uptrend situation. We draw a line connecting
the two lows and extrapolate to the next bar. If the price is a near
the bottom of the current bar when it is completed, it will be below
the level of the extrapolated trendline when the third bar starts.
My reading of your rules is that you would immediately cover and go
short. Is this correct?
Now assume that the uptrend continues. What would cause you to cover
the short and go long again? I assume it would be the beakout above
the top of the previous bar, as the extrapolated line connecting the
ascending tops of the two previous bars would lead to covering and
going long at a significantly higher price. Correct? I guess one
could also use a break back above the lower extrapolated trendline.
However, this might lead to lots of whipsaws.
This example suggests that using the trendline extrapolation technique
will cost you money on some trades. In this example (assuming I am
right about when you cover the short and go long again) it will cost
twice the difference between the cover long and go short price and the
top of the previous bar when you go long again + slippage. I will
have to asume based on your experience that stalls, such as this, are
less common that reversals on this time scale.
Jack Hershey
clutter up the new posts so much. Also it is my preference to respond to
questions here because I believe a lot of people probably have similar
questions. I often ask people to try stuff out to get them to make an
effort to actually understand what I am saying.
I detailed out some comments below to respond to your more specific
interests. *** begins each of my comments.
Your comments
Hi Jack,
I would guess a fundamental money making concept is what most of us
daytraders would need especially
in the higher time frames because exactly as you said a 30 min time frame
takes out the fear and
stress to react (instead of anticipating), being vulnerable to a string of
minor losses which might
add up including the commissions. Besides that it would provide a way for a
steady income to allow
someone to put time and efforts in optimizing and jump to study and
observing smaller time frames to increase profits and fun while trading.
*** Persons who are learning to make money usually make one of two choices:
1. To learn a method and make money; or 2. To forfeit the learning process
and thinl of a better way to make money on their own. You have in an a more
unusual way than most decided to try to do both.
As I really want to understand your way of daytrading, beginning with the
"warm-up" down to the
sophisticated stuff (which I have not really much clues about yet) I just
started to do what I
always do - going through charts and trying to visualize the given
instructions.
*** First, repeat what I did to get you started. In your first 3:00am
email to me you rushed to the wrong chart apparently and then sent me your
results to find out what either you or I did wrong. I just brought up the
quote.com 30 min chart for S&P the index you referred to in your post to me.
I did this to accommodate you. At this point, were I you I would pick an
index and try to repeat the task for a few days after close just to learn
what a break out is... how to extend horizontal lines. how to record profits
and losses, and how to find out what the daily profits is. doing it after
close will take you about fifteen minutes for a day.
To me it seems that the Cash Indexes are much smoother than the Futures
Indexes (so it might be a
good idea to take a closer look at the market movers for those who want to
trade stocks) and for the Futures Indexes the DowJones seems to be smoother
than the SP500 while the Emini got even more erratic moves and
retracements - overall they are mostly in sync and for a better signal on
the DJ you might get a better one for the SP later on.
****What I recommend is that you focus on making money. Your comparisons of
those indexes you looked at will probably fit into the subheading of trying
to figure out something better. Once you begin to understand that this
approach can be used on any index you will begin to focus on making money
instead.
The question to trade the first 30min bar is not clear for me - sometimes
your first trade is the
"trade of the day" because it initiates the following trend which might last
the whole day or
for hours but on the other hand and especially when the market gaps up or
down someone might be
better of to use a StopLoss or to avoid the first bar.
*****Okay I can under stand that the first bar is a challenge for you.
simply write down the high and low of the previous day and prepare to make
two decisions. Is the open inside or outside? And if it is decide whether
to place a market order that is either a "buy" or "sell" then you are on
your way. Most of your comments are those dealing with coming up with ideas
to trade in a better way. It is just another case of your getting side
tracked in a thought process that clouds using the first bar when it begins.
In this portion of the posts that I made to you, I refrained from getting
into the business of dealing with stops. What I was hoping for was to focus
on making money using a few exercises to make a modest beginning. These
four suggestions were to engender a little discipline in focusing on
something going slowly so there would be a beginning of understanding with
an actual experience. If you do go through what I suggest, you will get to
a place where there is some level of connection to trading indexes which
closely follow a cash market that is very very huge and moves ever so slowly
from place to place.
Well, even for the #1 warm up stage the most fancy enhancement is to watch
the next lower time frame by skipping the first 15min bar and start trading
the 2nd one from wherever it breaks through.
******It is discouraging to read your comments just above. what you are
pointing out to me is that you do not understand the idea of looking a bar
and drawing two horizontal lines, one across from the top and one caress
from the bottom. Try to invest a little time on this concept. I am trying
to get you to look at the market in pieces.. they are thirty minute
pieces...I am trying to get you to examine an interval of time that is 30
minutes long and use the information you have to make a decision bases upon
some uncomplicated rules embodied in a series of steps. You have missed the
point of my post in response to your request for information. I strayed
from just answering your original questions for a single big reason. I
wanted to clean the slate of a lot of miscellaneous information that could
be in the space so an excellent beginning could be made.
*****dkomo did a neat three fat cats post on 5/31. Read it. Some famous
guys are doing a buy and hold strategy and they made some bucks that way.
They, of course, aren't primary;y investors, they are all buy and hold
types... with respect to what they do for a living...run the companies
cited. They gave themselves loads of stock in each company and over ten
years that stock and the additional stock they gave themselves has gone up
some were and there. Say 60 times in ten years.
******We aren't doing that in your posts......we are making money instead.
We are not going to fool around running companies. So we have to focus on
using a paradigm. I am trying to get you to take a few bucks and make some
money not give yourself loads of stock and run a company.
Olivia and Natalie have three years in the stock stuff, you have or four day
week in. Both of our teams are doing okay compared to the fat cats.
Overall I would say (not backtested from my side though) that even stage #1
creates profits but
someone might have to deal with occasionally drawdowns in the 15 points
range (Emini).
*****I did ask you to do last week. Think of this as backtesting. I hope
you did it. I did it the on the first level to get people in gear here and
to get them to make a minimal effort to give themselves a break. What did
you learn? Answer: you made money. Few people who did it or are reading
this know in any way why it made money. I do. But that is not what is on
the table here. We are not having an academic discussion of why this makes
money. What we are talking about is just that there are futures indexes out
there and that you can make money on every one of them. Period.
To include StopLosses and ProfitLocks in this stage does´t make sense for me
because it would srew up the entry/exit rules and kill the nice runs when
the train heads in your direction.
*****Good you got that. Here is another thing. You definitely know that if
you did you skill level is such that you would have screwed it up. I hope
everyone really gets it straight that when you take over the markets job,
you are stepping over the line.
The results for #1 "break out of prior" have been 58.00 Emini points or
$2,900 for the week from
05/22 till 05/26 calculated with one contract and without commissions.
*****Cool you did it for the week before the one I suggested. If you do it
for the week I suggested and for the chart I suggested it will be a little
easier to discuss. At least everyone can see that we do much better than
the fat cats of 5/31
***** Now with three more ways to do this we ought to do even better. We
need to learn why it makes money a little too. These are just warm up drills
so they aren't optimum ways to make money but they do connect you to the
market.
79 3950 35
TGK
Von: Jack Hershey <jher...@primenet.com>
Betreff: 30 minute warmup bar trading.
Datum: Saturday, June 03, 2000 12:54 AM
The four items in the progression are:
1. break out of prior bar. <-------
Jack Hershey
I commented below again with the ***** inserts
Hi Jack,
#2 "Slope pairs of bars" - when I first added this improvement mentally to
the chart I thought this
will screw up everything and someone has to face lots of whipsaws which will
indeed cut the losses
but will kill the good runs as well - what I then did was to use the same
exit and entries from the
chart where I marked the buys and sells for #1 "break out of prior bar" and
added the line using the
last 2 bars and adjusted the former entries/exits.
****It is good to try to guess what will happen. You did and you learned
two things. 1. you are not a good guesser and 2. next time I will just do
it and save time and learn what is the purpose of what I am doing.
This will give an early exit which saves a few points but the most important
improvement is the fact that you immediately reverse the trade which gets
you on board early for the next trade and in either way you gain a few extra
points if right or save a few points if wrong because you start the trade
inside a bar and not at the start or end.
****Early is the key word here. Two reasons for this: Get more of the
trend you are in (maximizing). Get more of the new trend you reversed into.
For losses what happened. Look. Nothing. Thats okay.
*****You went from 2,900 to 3,950 what you say is a 35% increase in profits.
*****What was learned was this, we can use trend lines. No stops. Just stay
with the trend until the trend extinguishes in this market which moves very
very slowly with a spread off the cash indix made up of many individual
securities..
Even it´s a little bit tricky
****** Tipping a horizontal bar to touch a pair of ends is the least
tricky thing that could be done. It just states the trend. and that the
trend, if it weakens, is going to reverse in a period as slow as 30 minutes.
This is a slow action taking place. We aren't shooting for the maximum just
35% more. The next two parts will each give us more too.
its cool stuff because your entry/exit is within the bars range but
the next exit is always the between the low/high from the two previous ones
but you always have to
look what comes first - the breakout/breakthrough for the high/low or the
line and where the bar
opens! A strong trend will remain untouched till it gets week.
For the less experienced trader (like me) it always pays off to draw the
chart, mark entries and
exits, draw the lines and do a few days on paper to see and compare - just
looking at instructions
and rushing over a few bars doesn´t do the job.
***** So true. Now our attitudes are getting closer to being the same.
The results for #2 "Slope of pairs" have been 79.00 Emini points or $3,950
for the week from 05/22
till 05/26 calculated with one contract and without slippage & commissions.
That´s an improvement
for more than 35% compared to the #1 "Break out of prior bar" for the given
week.
**** okay is 4000 dollars on the margin of a single contract a good
return??? Well use the compound interest formula for a year. Always
remember to add contracts too when possible. I think in less than ten years
we are doing better than the fat cats if we can use this #2 level.
****Level #3 will make some more money it looks like and level #4 further
extends and makes the truns sooner and also takes the loss factor out of
the equation. Losses become scratches..
*****Next to lastly can you begin to see you are using a stop substitute
with the reversal technique.
****Lets see why it works. Then we can see the limiting case for its
application. some of you should hav a feeling here right now because I have
given you the clue as to how it works when I said "limiting". What I am
including in my posts I hope, is something new for everyone in each post.
There no point in going over anyone's head ever but there always should be
something that is leading to higher levels of understanding and sensitivity.
its the relationship[ of the bar length for a duration the the slope of the
trend. if you get too short with the bar duration the range of the bar
doesn't diminish correspondingly because of inherent "noise" in the market.
we will run into this in setting stops mathematically too. The "noise" of
the market is what sets the limits of optimization. If scalpers weren't
playing their game against low commissions the noise would be even lower.
*****We just have to be satisfied with a beginning that gives you say 4000 a
contract a week after doing half the four drills. It has taken the weekend
to get half of them done to start the next week.
Jack Hershey
"Don Cameron" <donald....@sympatico.ca> wrote in message
news:rv4jjsgbueco7a97o...@4ax.com...
> Method 1 is very clear but, as is common, refinement adds complexity
> and possible ambiguity.
>
> Consider the following situation for method 2. You are long and for
> the current bar we have a higher high and a higher low that the
> previous bar i.e. an uptrend situation. We draw a line connecting
> the two lows and extrapolate to the next bar. If the price is a near
> the bottom of the current bar when it is completed, it will be below
> the level of the extrapolated trendline when the third bar starts.
> My reading of your rules is that you would immediately cover and go
> short. Is this correct?
Yes, I would.
>
> Now assume that the uptrend continues. What would cause you to cover
> the short and go long again?
Break out on horizontal line at the top of the prior bar.
I assume it would be the beakout above
> the top of the previous bar, as the extrapolated line connecting the
> ascending tops of the two previous bars would lead to covering and
> going long at a significantly higher price.
Right, you only draw the sloping pair bars on one side of the pair.
Correct? I guess one
> could also use a break back above the lower extrapolated trendline.
> However, this might lead to lots of whipsaws.
Yes, we want to saty out of the "noise". so we would wait until we got to
the top (One whipsaw value)
>
> This example suggests that using the trendline extrapolation technique
> will cost you money on some trades. In this example (assuming I am
> right about when you cover the short and go long again) it will cost
> twice the difference between the cover long and go short price and the
> top of the previous bar when you go long again + slippage. I will
> have to asume based on your experience that stalls, such as this, are
> less common that reversals on this time scale.
Cool. thats right.
The #3 gets us to a good point on this issue as well.
Ultimately its just these issues you raise as we keep using #2 to see what
the shortest bar duration there is that can be used to make money. The
issue of bar length for a given duration is the one that is on the table
when we leave this starting evolution of techniques.
The first three don't eliminate loosing trades but the improvement in
winning trend length advances. The slow 30 minutes bars keep us steadied
down for learning to monitor between decisions...then we shorted bars to
step up to the line in decision making. laslty we switch to the optimum
side of the bar for exits.
We leave this series to begin to get the leading indicator concept in had
and begin to see that the market is a two variable system as well. we move
further and further into market psychology as we go.
Jack Hershey
the description below shows you know how to chose the correct pair of 15
minute bars to put between the regular 30 minbars. As I said I colore them
with yellow and orange highliners.
as you take a yellow pair you can project them as you did successfully in
#2. then you take the alternating orange pair and do the same. the
projections are shorter always because one side is sloped. I color the
resulting range as red for the yellow pair and green for the orange pair.
Now something will be soooo clear at this point. In the middle of trends
the red/green markings are longer. At the point where reversals are coming
about the length grows very short of is a negative length because the
projection overshoots .
You reverse as on #2 it may be on a green or a red projection though. So
now trends can last as long a any combination of 15 minute intervals. And
you are doing the reverses sooner and making more on each trend.
Once you see this phenomena occur, you become more and more relaxed with
the trading because you have become more and more anticipatory in your
charting efforts. what is so terrific is that your reversals push well up
into the bottoms of the peaks and well down into the troughs. there is also
a smoothness that you were looking to see and very few surprises from this
point on.
If you correlate this with 50% tracebacks you will see this is at least as
effective.
with respect to volatility, you see there is a volatility compression in the
bars leading into trends as well. The attribution for this is the decline
in volume that is preceeding the reversal. As you become more and more able
to just chart the #3 effort, you are much more able to anticipate the time
to make the phone calls and do your reversals at market be doubling down the
# of contracts.
Read Don Camerons comments because he is very astute at seeing the nuances
of what is going on here.
You can see at this point each enhancement makes you increased money. Many
of the enhancement increases are in the range of the monthly total profits
some are comparing here for other methods, etc. It, therefore, looks as if
we have exceeded the limits of credibility right off. what could be a
possibility is to reajust the limits of credibility.
TGK" <tgk...@lycosmail.com> wrote in message
news:393adef6$0$65...@SSP1NO17.highway.telekom.at...
> Hi Jack,
>
> on this one I could need a little bit advise - might be because I dont see
an example or my english
> ties me down on this one - maybe a combination of both...
>
> From my point of view it´s all about plotting 2 time frames on the same
chart - but not unlikely
> FibTrader where two 15min bars getting covered from one 30min bar instead
the 15min bars are
> displaced backwards by 1.
>
> So you start with an empty chart/sheet - it´s 09:30 and the first 30min
bar being formed - I am
> looking at the 05/22 Emini chart - after 30 minutes I have one down 30min
bar or two down 15min
> bars - the 2nd 30min bar and the 3rd 15min is starting........
>
> 30min bars = 1 2 3 4 5 6 7 8 9
10 11 12
> 13 14
> 15min bars = 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
24 25 26 27
>
> I do start my drawings when the 3rd 15min bar has ended and draw the
highest High and lowest Low of
> the 15min bar #2 and #3 by creating my first "in between" 30min bar - then
comes the next standard
> 30min followed by the highest High and lowest Low for the 15min bar #4 and
#5 and so on - when the
> day is over I will have 27/30min bars, correct?
>
> But how do I draw my lines from now on and get my entries and exits? Do I
include the "in-between"
> bars and view the standards and "in-betweens" together treating them as a
completely new chart? i.e
> going short when a "in-between bar" breaks through the low of the previous
"standard" one and start
> to use the highs from "standard" and "in-between" as exit?
>
> Including both sets of bars sometimes seems to produce a conflict between
the trendline and the
> breakout rule?!
>
> Thanks
> TGK
>
> P.S. A little to learn here? That´s pure understatement! :-)
>
>
>
>
Jack Hershey
The retracement does need a Numerical example to best be worked through.
No matter how many bars long a trend is, for me only the bar being formed is
subject to retracement. Here is the reasioning I use. One we arbitrarily
change bars, I believe the prior bar is "banked". That is the rules now
applly to the bar forming and that bar only. It's not elegant but we are
still at the stage where our money making doesn't take much skill either.
Lets assume we can enter on the first bar to get on with the day. We go in
as soon as the last bar of the prior day is broken out of. This is long at
1456 even. So we can just use the step from now on. I'll add some notes to
each step. We are long. We are in. The price is rising. We are monitoring
and looking for one thing: a peak.
1.enter on bar 1 when Break out occurs. Log value. ****we logged the
breakout value. We own it at that value. And we own it with a buy our
ticket is 1 contract #OW1 2045.
2. log peak/trough value on entering bar when it and if it occurs.
**** two bars occur, and we record the Break out value of each in turn.
they are 1458 and 1462 both even. the next rises to 1468 even and begins to
drift lower.
3. take difference of 1. and 2. divide by two. log.
******The answer is 3 even. 1468-1462=6. 6/2=3.
4. Get the midpoint value (50%) using the difference
**** As soon as the drift lower starts we do this step and write down 1465
even. This is the sell and reverse value for our 1 contract. 1468-3=1465
5. exit and reverse at that value.
**** Monitor, Call Do an order. Ticket OW1 2084 is assigned. 1 contract
becomes short and running.
6. if price retraces to buy of 5. exit and reverse.
******Step 5 put you short with1contract. If the short trend reverses
inside the present bar that was retracing and it goes back up to where you
entered short, scratch out this is a reversal failure and congestion has set
in.
6. log break out value (current bar extends beyond an end of the prior bar)
*****the trend goes along for a while and the price extends short below the
prior bar. Log this value. It is the reference for the possible next
retracement. If, in this case, a trough occurs, log the value and get set
up to exit.
if and when it occurs. repeat 2 through 6.
__________________
***** my prior notes
This is a left side of trend channel technique, i. .e., the extreme side of
the channel instead of the mean side of the channel (centering tending
side). Either you are making money here or scratching.
Your post began here:
Hi Jack,
Sorry, I tried all sorts of combinations but even after jumping on my
calculator I can´t bring this
to work!
1. ok
2. I don´t understand what you mean with peak/trough value?
3. should be: when 1. is X and 2. is Y it´s "1 - 2 / 2"
4. 50% from what? using the difference from above?
5. ok
6. I don´t understand "retraces to buy of 5. exit and reverse?
If you don´t mind - a sample day with numbers would help so much!
Thanks
TGK
TGK
Jack Hershey <jher...@primenet.com> schrieb in im Newsbeitrag: 8hetk2$47q$1...@nnrp03.primenet.com...
> *** Persons who are learning to make money usually make one of two choices:
> 1. To learn a method and make money; or 2. To forfeit the learning process
> and thinl of a better way to make money on their own. You have in an a more
> unusual way than most decided to try to do both.
##### Okay I have to admit that I was wrong thinking that I can trade my usual way and learning from
you and following your instruction & suggestions in realtime on the other hand - it´s not possible
especially with te speed I am picking up the things.
> *** First, repeat what I did to get you started. In your first 3:00am
> email to me you rushed to the wrong chart apparently and then sent me your
> results to find out what either you or I did wrong. I just brought up the
> quote.com 30 min chart for S&P the index you referred to in your post to me.
> I did this to accommodate you. At this point, were I you I would pick an
> index and try to repeat the task for a few days after close just to learn
> what a break out is... how to extend horizontal lines. how to record profits
> and losses, and how to find out what the daily profits is. doing it after
> close will take you about fifteen minutes for a day.
##### Let me say it this way - after reading all your May posts and 50 or 60 others from this and
last year it was obvious to me that there is tremendous knowledge and experience to gain - so I
simple got impatient, one of my real bad habits to force things to quick but I am working on it :-)
> *****Okay I can under stand that the first bar is a challenge for you.
> simply write down the high and low of the previous day and prepare to make
> two decisions. Is the open inside or outside? And if it is decide whether
> to place a market order that is either a "buy" or "sell" then you are on
> your way.
##### What I have problems with is the fact that if the market opens below the low of yesterday
(which the Emini did YD and today) I see three choices - going short at the open and placing a stop
at YD´s low (here I would hope that the market doesn´t gap to far below YD´s open) and reverse the
trade when the stop gets hit - going short at the open and placing the stop at YD´s last 30min bar
which would create a real big loser when wrong and screw up the setup for the next trades - or going
long as soon as YD´s low gets hit but where place a stop when this happens on the 1st 30min bar?
Choosing the right entry for the first trade seems to be very important because wrong decision here
and you put at least the next trade in the sand especially as I begin to understand that at this
level placing Stops besides those which the market tells us makes no sense.
> ******It is discouraging to read your comments just above. what you are
> pointing out to me is that you do not understand the idea of looking a bar
> and drawing two horizontal lines, one across from the top and one caress
> from the bottom. Try to invest a little time on this concept. I am trying
> to get you to look at the market in pieces.. they are thirty minute
> pieces...
##### Sorry, there was no intension to discourage you but as you mentioned before it seems to be
much harder to learn new things when you think you already know something and to get your head
free - therefor kids are much more open and adapt things much quicker and bring them to work. As I
was mostly dealing with 1-2-5 min time frames before and never met anybody before who shared and
explained things which DO work I had to step back a little getting out most of the things I have
learned to look at which besides I am beginning to question in depth more and more.
> *****Cool you did it for the week before the one I suggested. If you do it
> for the week I suggested and for the chart I suggested it will be a little
> easier to discuss. At least everyone can see that we do much better than
> the fat cats of 5/31
##### I am doing last week as well and will try to share the results and observations so anybody who
is interested (eveybody should be) can participate.
> *****
> 79 3950 35
##### Did I already miss something on those?
Thanks
TGK
TGK
-----Ursprüngliche Nachricht-----
Von: Jack Hershey
Newsgroups: misc.invest.technical
Gesendet: Monday, June 05, 2000 3:32 AM
Betreff: part#2
****Early is the key word here. Two reasons for this: Get more of the
trend you are in (maximizing). Get more of the new trend you reversed into.
For losses what happened. Look. Nothing. Thats okay.
##### When I ignore Inside bars what should I do when the next bar also closes above the original
trendline (ie. Emini from Monday - bar #5 to bar #9) Should I expand the orginal trendline
connecting low of bar #5 + 6 or start a new one connecting low of bar #6 and 8?
Okay, the 12 oclock Emini bar was an inside bar and it looks that expanding the original trendline
would have been a better idea for keeping you with the trend - but I already know that
#3"overlapping pairs" will provide the answer.
*****What was learned was this, we can use trend lines. No stops. Just stay
with the trend until the trend extinguishes in this market which moves very
very slowly with a spread off the cash indix made up of many individual
securities..
##### No stops - okay! But what when I see a very shallow trendline which happens when the original
stop at the previous bar does not get hit only whithin a few ticks so the entry bar is still valid
when it closes. Connecting the 1st and 2nd high (when short) will provide a shallow trendline which
when gets hit is producing a loss. Do we accept that fact while in stage #2 or could we place a
mentally stop right in the middle of that bar?
****Level #3 will make some more money it looks like and level #4 further
extends and makes the truns sooner and also takes the loss factor out of
the equation. Losses become scratches..
##### Besides working with #2 in realtime (paper) I draw the charts on an empty sheet inserting one
30min bar and two 15min bars but believe that I should fully understand #2 before using #3 in
realtime (paper) especially the point when and where to intitiate the first trade.
*****Next to lastly can you begin to see you are using a stop substitute
with the reversal technique.
##### That was and is the most astonishing thing for me - to go with the trend but overall stay on
the safe side - with my MA crossovers I can only survive by scalping with tight Stops and
ProfitLocks but we don´t wanna discuss this here and now - the key is to make money :-)
Thanks
TGK
Jack Hershey
I responded to your original post that began the thread with several items.
The first one involved using the Standard and Poors Symbol SPX.X and the
week including tuesday through Friday with dates including 30MAY through
2JUN, a four day week. I did that to partially respond to your S&P
orientaion. And I ask that four successive sets of rules be followed to
make several points about beginning to day trade as a way to see that there
are possibilities of looking at a market's price record to use simple rules
that work in trends.
You gave me an example for a day that apparently isn't on the chart. My
chart has Monday of that week as a Holiday. I know your prior effort chose
a different week to examine than the one I did. Thus i am making some
general comments to try to give you some help.
I did a four day sequence to illustrate most of the possibilities that could
occur. There are some that didn't.
Level #1 is the simplest break out format that may be considered. The
underlying principle that allows it to work is that the bar heights for a
long duration bar are lengths that well overcome the "noise" in the market.
The "noise" is the small fluctuations that are relatively insignifcant when
adjacent treansactions in time are linked together. this "noise" be comes
more and more significant when the bars duration is ever shortened to its
limit of just representing the duration of one trade.
#2 was intended to show that if there was a trend and adjacent bars were
paired (and only adjacent bars were paired) that there was an advantage
related to the trend and the trends strength. Thats all. Simply that you
could use the trend as your friend. What was desired was to make money.
This being the case the side of the pairs of bars that was important was
only one of the two sides. reasonably, it is the side the potental breakout
that would end the trade would occur on. So only those ends of the pair are
considered.
As a side comment, you needed to think about ignoring bars, stops, and
midpoints of bars. Also you wished to consider the #3 exercise. That is
fine and it also delays and thwarts you efforts to understand basic
fundamental cardinal points. for me to read about your intellectual trail
is wonderful for then I learn how to better try to emphasize what I believe
it is possible to learn. Today I watched my capital meander between 87 and
92 thousand dollars greater than before the market opened. and I noticed the
group of stocks that I own where I have 4000 shares each of selected stocks
in the high 50's were behaving differently than the group of stocks I own
that were making and additional 4% for the day. I have to think clearly
about my strategy in these matters. I absolutely do not bring into the
picture any extraneous thoughts whatsoever. Here is the reason: I believe
it lessens my effectivness in dealing with what I am trying to do.
Between #1 and #2 if you used any week with any market, and that is what you
did and I couldn't repeat your work and you didn't repeat mine), you learned
that you could improve your performance 35%. That is a lesson well learned.
To polish it a little try doing what I suggesteed and see that you get the
same results. As you do as I suggested you will not be thinking about the
things you ordinarily do vis a vis stops, and locks and MA crossovers and
stuff.
Subtly there is an opportunity to synthesize an approach from the very
beginning and bring into play the various tools that are available
"TGK" <tgk...@lycosmail.com> wrote in message
news:393d4ece$0$72...@SSP1NO17.highway.telekom.at...
> my comments ##### - sorry for the ones I´ve sent directly to you - wrong
"reply to" button
>
> -----Ursprüngliche Nachricht-----
> Von: Jack Hershey
> Newsgroups: misc.invest.technical
> Gesendet: Monday, June 05, 2000 3:32 AM
> Betreff: part#2
>
>
>
> ****Early is the key word here. Two reasons for this: Get more of the
> trend you are in (maximizing). Get more of the new trend you reversed
into.
> For losses what happened. Look. Nothing. Thats okay.
>
> ##### When I ignore Inside bars what should I do when the next bar also
closes above the original
> trendline (ie. Emini from Monday - bar #5 to bar #9) Should I expand the
orginal trendline
> connecting low of bar #5 + 6 or start a new one connecting low of bar #6
and 8?
>
> Okay, the 12 oclock Emini bar was an inside bar and it looks that
expanding the original trendline
> would have been a better idea for keeping you with the trend - but I
already know that
> #3"overlapping pairs" will provide the answer.
>
>
>
> *****What was learned was this, we can use trend lines. No stops. Just
stay
> with the trend until the trend extinguishes in this market which moves
very
> very slowly with a spread off the cash indix made up of many individual
> securities..
>
> ##### No stops - okay! But what when I see a very shallow trendline which
happens when the original
> stop at the previous bar does not get hit only whithin a few ticks so the
entry bar is still valid
> when it closes. Connecting the 1st and 2nd high (when short) will provide
a shallow trendline which
> when gets hit is producing a loss. Do we accept that fact while in stage
#2 or could we place a
> mentally stop right in the middle of that bar?
>
>
>
> ****Level #3 will make some more money it looks like and level #4 further
> extends and makes the truns sooner and also takes the loss factor out of
> the equation. Losses become scratches..
>
> ##### Besides working with #2 in realtime (paper) I draw the charts on an
empty sheet inserting one
> 30min bar and two 15min bars but believe that I should fully understand #2
before using #3 in
> realtime (paper) especially the point when and where to intitiate the
first trade.
>
>
>
> *****Next to lastly can you begin to see you are using a stop substitute
> with the reversal technique.
>
Don Cameron
<jher...@primenet.com> wrote:
>. At this point, were I you I would pick an
>index and try to repeat the task for a few days after close just to learn
>what a break out is... how to extend horizontal lines. how to record profits
>and losses, and how to find out what the daily profits is. doing it after
>close will take you about fifteen minutes for a day.
>
A word of caution here for anyone following with end of day data.
While reviewing historic trades is useful in getting a feel for the
method it may nor give reliable figures for profit and loss. Look at
today's NQ00M for example. The bar for the previous day's last 30
mins was exceeded twice on the low side and once on the high side in
the first 30 minutes. Using your method 1 (I am taking it slowly and
still studying this approach) this would have resulted in a short,
long, short sequence followed by going long on the next 30 min bar.
The two whipsaws would have resulted in a loss of about 80 points
before getting into the profitable long. This is indecipherable on a
30 minute chart and any outside bar (higher high and lower low than
the previous bar) should be studied using data from a finer timescale
to analyse the sequence of trades and detect possible whipsaws.
TGK
same on the SPX.X chart - 05/31/00
-long at 9:35 - the breakout of last 30min bars high
-cover at 9:40 - the breakthrough of last 30min bars low
-short at 9:40 - the breakout of last 30min bars low
-cover at 9:47 - the breakthrough of last 30min bars high
-long at 9:47 - the breakout of last 30min bars high
-this one did the job
Here the risk was not too bad because 05/30 last 30min bar had only a range of 4.63 and it was the
only day for the week when it happend if you have any TA software you could run a test to get a
ratio how often that occurs. For May the ratio is positive and around 3:1 and is #1 out of four
possible stages.
There also is the suggestion to draw in the day befores high and low although I couldn´t figure out
the enhancement.
Regards
TGK
Don Cameron <donald....@sympatico.ca> schrieb in im Newsbeitrag:
n24rjsghpclg6pas1...@4ax.com...
> On Sun, 4 Jun 2000 17:52:29 -0700, "Jack Hershey"
> <jher...@primenet.com> wrote:
>
> >. At this point, were I you I would pick an
> >index and try to repeat the task for a few days after close just to learn
> >what a break out is... how to extend horizontal lines. how to record profits
> >and losses, and how to find out what the daily profits is. doing it after
> >close will take you about fifteen minutes for a day.
> >
>
James Fowler & Harla Yesner
I don't mean to quibble, but I could never get an automated system to be
profitable using your simple bar top or bar slope methods. The problem arises
when the trade you take on retraces to a loss--sure you capture the trade going
the other way once you break out of the prior bar or bar slope, but in the mean
time you eat a pretty sizable loss because you were "hanging loose" waiting for
the retracement to reverse back in the direction of the initial break out. On
the other hand, if you don't take the loss because you have some other stop
strategy available then we are moving to a more complex system.
Let me be more specific.
Bar one is hi 100 lo 50
Bar two is hi 120 lo 70 -- we've taken on a long at 100 that is meandering
between a profit and a loss.
Bar three is hi 130 lo 80
Bar four is hi 120 lo 60 -- we're stopped out at 80 from previous bar for a loss
of 20+ and we take on a short at 80.
Bar five is hi 110 lo 70
Bar six is hi 100 lo 80
Bar seven is hi 120 lo 90 -- we're stopped out at 100 for another loss of 20+
Etc.
On slow days it is not unusual to produce five losers in a row on the order of
20 - 30 points each.
The slope method produces smaller but more numerous losers for an essentially
similar reason.
These bars are purely fictional. I can find a real world example if you like,
but in my statistical test of the simple bar tops method and the slope method, I
cannot get statistically significant profits. I also added various scalp values
without success. The problem is that the simple method as you have laid it out
is too easily whipsawed on slow trends.
Am I missing something? I can only assume that there is a more sophisticated
stop strategy that is being implicitly integrated in cases where you have taught
others to use the simple system successfully.
In fact, it is my frustration with the simple method that has led me to
investigate so intensively your 5-min and 1-min method, which has yielded better
(though not yet acceptable) results in historical data simulations.
Please help me figure out where my error is.
Thanks,
James
Jack Hershey
you on the automated trading. The four stages of the 30 min warm up are to
elucidate just exactly how a markets works from a fundamental and
theoretical TA point of view. Look for the "adequate volume" signal line
and how it is generated as an anticipatory signal.
Before these posts slip off the bottom of the list I am going to clean up
all the loose ends.
A lot of people have made some terrific points which establish a well
scoped point of departure to formulate, constructively, several of the next
set of enhancing elements.
What was of primary importance was to crudely establish that trends prevail
and when they sufficiently rise above market "noise" that comes from the
shape of a market. Once there is more than background "noise", there is the
element of change. Changes are successive and their is always the
possibility of sufficient disagreement in the market for it to return to its
"noise" level and mark time.
The next order of business is to set up differentiators of these phenomena
and to further expand our ability to anticipate the end effects of trends as
well as the end effects (beginning and end) of markets that have a standard
periodicity of opening and closing in a global continuing cycle.
The inane in ability of Greenspan to have to open his mouth when he does is
just one of those fortuitous ways to make money for some of us and
unfortunately for many many others to loose money. He is so abysmally
irresponsible in his incantations and his actions.
One of the ways we will deal with whipsawing is the move to the opposite
side of the envelope in which it occurs.
"James Fowler & Harla Yesner" <james_...@harvard.edu> wrote in message
news:3942A925...@harvard.edu...
> bzzz wrote:
>
> > > On slow days it is not unusual to produce five losers in a row on the
> > order of
> >
> > I think this might be the answer - "on slow days".
> >
> > The whipsawing you described usually happens on trendless low-volume
days,
> > which ideally are days when you would not be trading. As far I know,
the
> > easiest way to know of these days in *advance* is to monitor the news
> > (ie,"is Greenspan speaking today?")
> >
> > I had a jiu-jitsu teacher who had me in a figure-4 headlock, and I was
> > supposed to escape. After trying everything I could think of, I asked,
"how
> > the hell am I suppose to escape". He laughed and said, "there is no
escape!
> > the only escape is to not to get here!" Same thing with trying to trade
on
> > low-volume days.
> >
> > Bill
>
> Nice metaphor.
>
> The trick, then, is to add a simple volume anticipatory device that can be
> automated. But as it stands, the simple bar approach needs either
refinement or
> the human touch.
>
> James
>
>
Don Cameron
that trading on the basis of breakouts form 30-min bars or the slopes
of adjacent 30-min bars alone does not result in a trading method
which yields statistically significant profits?
If you do not agree, what has he been doing wrong? If you do agree,
why did you present them in a way which suggested that they do
represent profitable approaches, rather than as a first step in
understanding the development of a more sophisticated approach? Have
you subjected these approaches to a computerized trading simulation
and found them to be profitable? If so, over what period of time and
on which index futures?
I am finding this very confusing and a bit disconcerting. Perhaps you
could also respond to bzzz's post questioning the smooth transition in
scoring of a stock as it goes though a trading cycle. He asked for a
specific example.
I am not questioning your ability to trade futures or stocks in
similar ways to those you present here. However, I do wonder and
have for some time whether you unconciously apply additional filters
form your own internal neural network based on your years of
experience that perhaps even you do not explicitly recognize .
Would you please explain for the rest of us who did not receive a fax
what the adequate volume signal line is. As always, you are invited
to post illustrations to clarify your approach at my board discussing
your methods.
Jack Hershey
Sorry to be so slow to respond. I parsed it below, briefly. I agree lets
dwell where we are for a while to see if its possible to have a common
starting point. I used a four day week of stuff to illustrate what I was
trying to convey. Then as TGK did it along with me and was getting improved
results going from #1 to #2 to #3 and then delving into the left side #4. I
thought we had a beginning. Several people sent charts of their sequencing
and their were a range of successes.
"Don Cameron" <donald....@sympatico.ca> wrote in message
news:b5saks8jdqaeq0eri...@4ax.com...
> Before we move on, Jack, do you agree with James Fowler's contention
> that trading on the basis of breakouts form 30-min bars or the slopes
> of adjacent 30-min bars alone does not result in a trading method
> which yields statistically significant profits?
it produces profits to an given extent that is directly proportion to the
ratio of the signal to noise ratio of the market. When the volatility is
low and approaching the noise level there is no way anything will make
profits. James is preforming a basic test of market volatility. one of his
examples at the beginning of the day was a whiplash example where the
volatility slowly increased and the trade started on the side of the BO that
lead to several consecutive losses using 30 min bars.
>
> If you do not agree, what has he been doing wrong?
James isn't doing anything wrong in my opinion. I haven't gone over his
methodology and lets always assume that it is correct. The fact is I just
gave a four day example that is similar to most that I have going on at a
given time. It did work as I posted and James got a different result when he
applied a test of significance to it or to a sample he chose.
If you do agree,
> why did you present them in a way which suggested that they do
> represent profitable approaches, rather than as a first step in
> understanding the development of a more sophisticated approach?
With beginning persons, I do this series. I do move them to higher and
higher levels as they permit. What stops the process invariably is their
getting behind the beat. once they loose their composure I request that
they exit. And we spend alot of time getting through a debriefing of what
was going on with them. It is my opinion that the best traders, are using a
seemingly KISS oriented approach, but in reality they really have a firm and
strong grasp of the market's action. It may seem dumb at first but to not
use anything faster than 30 minute bars, is a good way to surface the notion
that the market is not erradic or unnatural.
Have
> you subjected these approaches to a computerized trading simulation
> and found them to be profitable? If so, over what period of time and
> on which index futures?
My answer is not going to please you very much. But it is a good one from
my perspective. i hope James is able to provide a lot of substantive backup
for trading in the market successfully. My interest is that of an amateur
who is focussed on making money primarily and sharing success with others
through transference of ideas that are collateral to the approaches they
use. and even actually getting someone to mimic my approaches. I have a guy
who isn't getting some homework done now because he has taken some profits
out to buy a home about a 400,000 dollar home he is paying cash for; the
closing execise and watching sunsets overshadows what the heck I need him
to do daily. We all know that is not prudent. But people do that
occassionally. If you take 1/4 of the profits on a stock turn in one of
your streams of trading who is going to really effectively criticize you.
>
> I am finding this very confusing and a bit disconcerting. Perhaps you
> could also respond to bzzz's post questioning the smooth transition in
> scoring of a stock as it goes though a trading cycle. He asked for a
> specific example.
Both of you are correct. Shortly, because I am enjoying a little confusion
here too, I will try to have at least four portfolios on clearstation , each
subtended by a "potentioal buy portfolio" and the "recenty owned portfolio".
These will be focussed on the scoring thing Billiam is very conscious of.
Whether it be the commodities or the stock paradigm, I believe in six months
I can get this stuff to a transferable state for most types of personality
and intelligence level.
> I am not questioning your ability to trade futures or stocks in
> similar ways to those you present here. However, I do wonder and
> have for some time whether you unconciously apply additional filters
> form your own internal neural network based on your years of
> experience that perhaps even you do not explicitly recognize .
Okay, I concede there is a serious Jokari type thing going on here. I am
trying to have all of us have as large a common reservoir of shared
knowledge. I don't know some things about others; they don't know some
things about me. That takes care of three of the four Jokari cells. You
raise the last one where neither of us knows whats cooking. I am
dilligently trying to uncover that stuff.
Billiam is a very cool cat in trading and James really is objective about
the model. Recently billiam said jack is sidelined when the model
performance you see whipsawing going on in is happening. there are 82 more
of those exchanges coming up I am sure. What is in hand is this. If we
were to actually start from scratch, a la, the simplistic 30 min thing or
the 30 points per day replacement for FDates, we would get to a screen that
has an array of items on it, and the array simply tells you what and when to
do stuff to just go on down the road very very effectively. at that time we
will all feel a little better about it all.
>
> Would you please explain for the rest of us who did not receive a fax
> what the adequate volume signal line is. As always, you are invited
> to post illustrations to clarify your approach at my board discussing
> your methods.
I want to do as you suggest. And even better I will begin to do that. The
reason is is that there are some guys around here that are good at getting
stuff out of my computer to yours and they are going to get into making that
happen regularly. I, at last, seem to be having luck at getting some guys
who are programer types to make enough money that they realize getting this
stuff out is a priority.
Strangely, I feel that we are on the brink of getting organized. I believe
that we can build from these two paradigms a way for anyone to be able to
have as his main resource flow an investment approach that fits his
temperament and life style. I expect that the most important single factor
to us all is its efficacy as clearly stated by the statistical significance
of the performance of the practice as related to the performance of the
market.
I will try to get the essentials of each software paradigm on the site you
have. The "adequate volume" thing has about twenty companions. As I
present these, I believe the math associated with them and how the realtime
data is related to them is explained, then there will be a better
understanding of how it all fits together under a fundamental set of
principles. Where I am now is that I believe everything can be continually,
iteratively, adjusted to optimize the paradigms. I am told it comes under
the heading of "fuzzy logic".
>
>
http://www.coolboard.com/boardshow.cfm?msgboard=586866385932493&page=1&retur
nURL=&idDispSub=659539442790969