The Stochastic Indicator

Status
Not open for further replies.
Had to leave at 3:30 for weekend traveling. Here is Friday's stuff:

Well, after the two retart days i tried to figure just what the heck went wrong....basically dealing with the open was killing me. i was drawing trendlines where there was just chop and trying to enter on any liitle break of a tiny trendline...instead of waiting for a rocket or rocket failure i was looking so closely i was jumping in at anything...the biggest mistakes was trying to enter to early....no rocket signal at all...no good macd....(stoch BELOW 50 for a long one time) just a trend line break....and in reality i was just getting whipped by the opening chop....in short i was NOT doing what i had been having success with...anyways went back to basics on friday and it offered a big 'ol meatball reversal...so a tough week with a good ending.

1. rocket failure--->macd xo reversal

ok. up a lot into the # at 9:45 i did not have a position on long b/c just seemed like a lot of risk to enter as the # was going to hit.....anyways...9:50 big fat bearish engulfing candle (nq's was REAL big) and trend line break...i waited for the macd to xo...10:00 macd xo short 878.00....held until 10:35 bar with sideways action exit @ 871.50...stoch and macd were still giving good readings, but i wanted to get a winning trade in my pocket after the tough week. if you held you made even more.

short @ 878.00
cover @ 871.50 +6.50

day: +6.50 19 days: 29.25

2. short rocket

1:25 s(5.26, 9.53) macd hist: -.47 v: 5678

also broke the low of day...1:35 bar reverses into the previos range and rocket fails.

short @ 866.00
cover @ 866.75 -0.75

day: +5.75 19 days: 28.50
 

Attachments

I used a simple trick to BT inside the 5m bars - used 1m data, and multiplied indicators parameters by 5:
Stoch(70,5,15)
MACD(25,65,30). This is a gross approximation by if you look at 5m and 1m charts side by side, indicator values for a 5 min bar don't differ too much from values of the 1m bar that corresponds to the close of the 5m bar (for instance, 9:54 1m bar for 9:50 5m bar). So signals are generally identical.

Settings:
-Entry on fast line + MACDHist, on close of 1m bar
-Exit on fast line, on close of 1m bar
-Between 9:30-12:00 and 2:00-4:00pm
-Stop at 2 pt.

Results for fast line only were abominable, too many whipsaws. So I tested:
-entry on fast and slow line + MACDHist
-exit on fast line.

Results vary from -$3k/contract to +$7k/contract, with better results if you use the 20/80 trigger (+$5.7k vs +$2.4k cumulated over 2 years).

Optimization: combined trigger (entry 20/80 for fast line, 30/70 slow line; exit 25/75 fast line) gives better performance, +$8.4k over 2 years. Translated: you enter earlier (fast line above 80) but avoid whipsaws (slow line above 70), and you stay in the trade longer and avoid whipsaws (exit below 75).

So what does this mean? The results are not impressive at all ($8.4k/2 years means 0.34 ES points/day).

Skeptics can gloat all they want :p, because, as I pointed out before, I don't see this as a mechanical methodology. The stochs/MACD are merely the primary signals, and you act on them based on other secondary signals. The way you interpret the secondary signals is discretionary, and so will be the results. I believe that if you follow Jack's guidance you'll do very well with this method.

A simple example where discretion improves results are morning gaps. The system takes lots of losing trades after gaps (I didn't feel like coding for eliminating this): enters in the direction of the gap on every 2nd 1min bar, and gets stopped out on the next as long as the signal is there, although mkt is filling the gap.

You have the 5m bar backtesting attesting for it - if you avoid the intrabar signals you would make $97k/2 years, or 3.9 ES pt/day (see attached).

This concludes my backtesting efforts for a while, it's time for forward testing now. I hope this is helpful.

Special thanks to Volker for allowing me to finish this.
 

Attachments

Quote from jack hershey:



Sorry that was humor.

I posted on this two posts back this am.

I think the effort to do the work to get sequences makes more money per hour invested than any other effort.

It is a terrific use of time when ever you are bored looking in from the sidelines.
More...

Thanks for your posts Jack. The humor too :p

Let's see if I got it right.
Here are 2 sequences that were played out more than once:

Sequence#1: 5m MACD convergence -> fast line leaves zone -> 5M MACD cross -> channel BO = FLAW so exit

Sequence#2: 5m MACD convergence -> channel BO -> second point 3? -> 5M MACD cross -> fast line leaves zone = FLAW so exit

#1 and #2 are different, but they have common elements. So you can write the following decision trees:

Channel BO
#1 you have the trendline broken and you exit because all other indicators are pointing that way too.

#2 the trendline is broken BUT the other indicators are still good so you wait for a second point 3, but exit when the other indicators give exit signals.

Fast line leaves zone
#1 channel still intact but it may be a whipsaw - you wait for 5m MACD cross and channel BO = flaws so exit.

#2 channel is already broken with 5m MACD cross so you have your flaw and exit.
You can move your decision point further up and exit on MACD cross. Or even further, you can add a new decision point, such as failure to traverse to the left side with peaking volume, and exit there if it occurs.

Did I get it right Jack? How would you translate this into a diagram?
 
Quote from vorzo:

I used a simple trick to BT inside the 5m bars - used 1m data, and multiplied indicators parameters by 5:
Stoch(70,5,15)
MACD(25,65,30). This is a gross approximation by if you look at 5m and 1m charts side by side, indicator values for a 5 min bar don't differ too much from values of the 1m bar that corresponds to the close of the 5m bar (for instance, 9:54 1m bar for 9:50 5m bar). So signals are generally identical.

Settings:
-Entry on fast line + MACDHist, on close of 1m bar
-Exit on fast line, on close of 1m bar
-Between 9:30-12:00 and 2:00-4:00pm
-Stop at 2 pt.

Results for fast line only were abominable, too many whipsaws. So I tested:
-entry on fast and slow line + MACDHist
-exit on fast line.

Results vary from -$3k/contract to +$7k/contract, with better results if you use the 20/80 trigger (+$5.7k vs +$2.4k cumulated over 2 years).

Optimization: combined trigger (entry 20/80 for fast line, 30/70 slow line; exit 25/75 fast line) gives better performance, +$8.4k over 2 years. Translated: you enter earlier (fast line above 80) but avoid whipsaws (slow line above 70), and you stay in the trade longer and avoid whipsaws (exit below 75).

So what does this mean? The results are not impressive at all ($8.4k/2 years means 0.34 ES points/day).

Skeptics can gloat all they want :p, because, as I pointed out before, I don't see this as a mechanical methodology. The stochs/MACD are merely the primary signals, and you act on them based on other secondary signals. The way you interpret the secondary signals is discretionary, and so will be the results. I believe that if you follow Jack's guidance you'll do very well with this method.

A simple example where discretion improves results are morning gaps. The system takes lots of losing trades after gaps (I didn't feel like coding for eliminating this): enters in the direction of the gap on every 2nd 1min bar, and gets stopped out on the next as long as the signal is there, although mkt is filling the gap.

You have the 5m bar backtesting attesting for it - if you avoid the intrabar signals you would make $97k/2 years, or 3.9 ES pt/day (see attached).

This concludes my backtesting efforts for a while, it's time for forward testing now. I hope this is helpful.

Special thanks to Volker for allowing me to finish this.
More...

VORZO,

You really should take the results of the system testing for what they are. Jack's method really gives you NO edge.

Now, you are at the point of applying a layer of discretion which you think will improve the results... So in reality it would be your discretion not the system that would make you $.

I dont get it.. why dont you just trade purely discretionary.. and avoid all indicators.. except for price and volume. In reality you are proabably in the same boat. Indicators alone dont work mecahnically.. so who says you are better off combining discretion with something that doesnt stand well alone??

As I have stated many times.. indicators like stoch's macd.. wahtever.. are all BS. They give you absolutely no edge. If you want to become a pro trader.. you have to realize this. There is no method using indicators that will give you a mechanical edge.. Once you start adding discretion.. then you are just making an excuse not to use the indicators..

The only reason you would become profitable trading similar to DAWG's performance thus far.. is because of you experience and ability to trade disectionary... I looked over Dawg's charts every night.. and noticed.. a lot of times he avoided many mechanical trades part of Jack's system.. I bet if he took each signal Jack's system put out.. he would be losing.

In reality its DAWG's discretion thats making him the $. I also bet that if DAWG continues his good performance and gets the hang of trading.. he will more and more use less of Jack's methods and use more of his discretionary ability.

You look to Jack as if he knows all the answers to the market.. in reality he is a fraud. I have never seen one trade of Jacks in realtime.. I have seen his system backtested many times yielding unprofitability.. And also I have seen Jack answer hundreds of questions with complete mombo jumbo crap... Jack claims you can make thousands of dollars as a novice with his bullshit...

I am just getting sick of people like Jack... playing with all the vulnerable newbies... and taking them for a ride...

Vorzo.. after your experience with jack.. which I bet will fail... you will be back to where you started... looking for the holy grail or some set of magical indicators... The answer is far away..

I dont have the answer to your questions.. but I do know that if you will make $ trading it will not be because of someone elses system or set of rules... It will be because you have found an edge and will apply your experience.




--MIKE
 
Quote from vorzo:



Thanks for your posts Jack. The humor too :p

Let's see if I got it right.
Here are 2 sequences that were played out more than once:

Sequence#1: 5m MACD convergence -> fast line leaves zone -> 5M MACD cross -> channel BO = FLAW so exit

Sequence#2: 5m MACD convergence -> channel BO -> second point 3? -> 5M MACD cross -> fast line leaves zone = FLAW so exit

#1 and #2 are different, but they have common elements. So you can write the following decision trees:

Channel BO
#1 you have the trendline broken and you exit because all other indicators are pointing that way too.

#2 the trendline is broken BUT the other indicators are still good so you wait for a second point 3, but exit when the other indicators give exit signals.

Fast line leaves zone
#1 channel still intact but it may be a whipsaw - you wait for 5m MACD cross and channel BO = flaws so exit.

#2 channel is already broken with 5m MACD cross so you have your flaw and exit.
You can move your decision point further up and exit on MACD cross. Or even further, you can add a new decision point, such as failure to traverse to the left side with peaking volume, and exit there if it occurs.

Did I get it right Jack? How would you translate this into a diagram?
More...


VORZO..

You are trying to make sense of the market... as if they really care about MACD and crossovers.. your post sounds as if Jack knows the rythem and rhyme of the markets...

"#2 channel is already broken with 5m MACD cross so you have your flaw and exit."

This sounds like the market is your friend and you know its personalities.. as if it really cares what all of these indicators are doing...

You have done the testing yourself.. market couldnt care less what Jack's indicators say.. nor any indicators at all.

I hope all the newbies are reading my post.. so I can save them lots of time.

 
What i see from Jack Hershey's methods is a mechanical system that is part discretionary. It is discretionary until you can recognize all the flaws that happen in certain sequences. Once you get to this point, it becomes a mechanical system at every level.

The big picture that i think Jack is trying to convey is that you can take money out of the market at all times of the day. It depends on your level of conciousness to see things.


This is my interpretation(what i see) of his system.
AM: Get in the trend using stochastics, 20,80, macd 5M Xover, Stoch cruise through 50%. Draw channels using points 1,2,3 which you can get using the P,V relationship. Draw second point 3. IF price doesn't reach the left channel or right channel with volume pooping out, then you reverse(usually around 11ish). Stay in the trade using P,V. you are either gonna get a wash trade or new points and channel.
MIDDAY: IF after the reversal, you have a wash trade, see volume decreasing, then you must come to the conclusion that you are probably in CCC(horizontal trend) and trade using the 1 min macd xovers for rapid reversal trading. Once you see centering which is VDU and macd, stoch on 50%, then do a break out bracket trade. This might take you to the new trend with channels and pts(look for volume increasing compared to CCC period.).
PM: Ride the trend using points 1,2,3. Look for second point 3. Do reversals based on price not reaching channel line and then do points 1,2,3 again.


This is how i interpret his Methods. As you can see, you will find that you really can take money out of the market at all times. It all depends on what you see(Level of consciousness). I am sure i am still not seeing a bunch of things.

I also wanted to ask Mr. Hershey a question. I sometimes get whipped around in the morning looking for the trend to materialize. What is the sequence that you look for for the initial AM trend? I think Dawg might have the same question. thanks.


Jc
 
Quote from nwbprop:



The big picture that i think Jack is trying to convey is that you can take money out of the market at all times of the day. It depends on your level of conciousness to see things.


Jc
More...

C'mon lets be real... all times of the day?

You are telling me Jack can trade the AM. the midday chop and closing all using his methods successfully everyday?

There is no method universal to all types of trading conditions.. and all parts of the day.

This is completely bogus. His methods are supposed to do all these wonderfull things.. yet when backtested they are completely garbage... Now he is gonna give you the song and dance.. of how you can not backtest his methods.. there needs to be a layer of discretion applied...

Just notice his first post in ET. He states that a beginner can make 6 figures a year easily trading his rocket methods...


--MIKE
 
Quote from Trend Fader:



I dont get it.. why dont you just trade purely discretionary.. and avoid all indicators.. except for price and volume. In reality you are proabably in the same boat. Indicators alone dont work mecahnically.. so who says you are better off combining discretion with something that doesnt stand well alone??

I am just getting sick of people like Jack... playing with all the vulnerable newbies... and taking them for a ride...

Vorzo.. after your experience with jack.. which I bet will fail... you will be back to where you started... looking for the holy grail or some set of magical indicators... The answer is far away..

--MIKE
More...

Trend Fader,

I think you mean well, although in a non-constructive manner, but you're not paying attention to what I'm saying. I said in an earlier post that I look at Jack's method as training wheels to help me acquire market consciousness and find my edge, not as a mechanical system. BT results gave me an idea of the number of occurrences of the mechanical setup, what I can expect if I take all setups, and how entering earlier on the bar can make a major difference in results.

So what I'm trying to do is discretionary, and I use the indicators as signals that point out that something, not visible in PV to the untrained eye, is happening. If I were to trade purely discretionary, as you suggest, it would be so discretionary I wouldn't know where to start. So again I would need setups, more or less defined (S/R levels, trendlines, PV action, patterns)?

You seem to have a real problem with indicators, as if you got hurt using them and now you hate the very idea. They have been used successfully for decades and now you come along and tell us newbies that they are shit.

And yes, I may be a newbie, but I've been around just long enough to realize that there is no holy grail and I'm not looking for it either. There is the EDGE, and each trader has to discover his own. That's what I'm looking for. And that's what Jack is trying to teach people: how to understand the market so they can find their edges. You may have had the chance to learn your trading directly form other people, and you may have gotten to the level where only discretionary makes sense anymore.

As far as Jack having an answer to all my questions - I'm sure he doesn't but he's showing me how to find my own answers. Too bad you can't see this, you seem to have a personal vendetta instead.
There are at least a few people, including dawg and myself, that have benefited greatly from Jack's coaching and from this thread. If Jack's method doesn't pan out, what I'll be getting out of the journey is priceless: better understanding of the market, and hopefully an edge.

vorzo
 
VORZO..

I sincerely wish you all the best... and I realize that you are a clever person.

Here is my problem... Jack came along from another site claiming to teach people how to make thousands of dollars as a beginner.. using somewhat esoteric methods.. that many people still dont fully understand even today...

There is absolutely no proof to his success nor that his methods actually work in real trading... I just caution people to the reality Jack is presenting... Look at the facts.. thats all.

UNtil I see some proof that Jacks methods work I will doubt every thing he says.. and consider him just another snake oil salesman. Altough his motives are not financial they might be based on a cult following he tends to create.

If someone comes to ET.. and claims these types of returns.. before even making their first few posts.. a background credibility check should be warranted. I think its only due dilligence.

Why doesnt he make a Journal in realtime and show us how well he trades.. why doesnt anyone of his true students tell us how they make 1-3x their capital trading rockets???

WHy should anyone care what Jack has to say.. if there is no proof in the pudding. You must take these types of claims with a grain of salt.. because you know that in this business 95% of the gurus are bull. And one thing for sure.. Jack does claim that he is a guru... look at this first few posts on ET>. stating how much people are expected to make trading his methods...


--MIKE
 
Status
Not open for further replies.
Back
Top