Twoblink and "The Secret Sauce"

So.. I might have the answer.. to the cow lick problem.

If 15M chart's 55EMA = 15 Minutes * 55 = 825 minutes
825 / 60 = 13.75 hours..

Rounded to 14 hours coverage.. that would mean I can put in a 14EMA on the 1H chart and get the same coverage. There should be fewer cowlicks, but it should be smoother at the same time..

So I didn't find the secret sauce today

So I might throw everything out the window, and start over with a blank sheet.

I need a solid concept, one that I can stand behind..

I need the Forex equivalent of fat-free cookies that don't taste fat free. I need something that tastes good, but is good for you with no harmful side effects or hangovers the next day..

Or am I asking too much? I would love to be right 100% of the time, but that's not going to happen. What about 80%? I always loved being a B- student!!

So let's dig into the psyche of the speculators:

1) If I think the market is headed to the moon, I want to get in low, buy low, sell high. I would love to be first in line, so I get in at the basement and sell at the penthouse, but everybody knows that won't happen..

So the evolution of news.. Would be: News...Reaction. BUT, this is SOOO contrary to the Dow theory.. Or is it? No movement is news right? It's not just the notes you play, but the rests you skipped, as my music teacher use to say..

So when I see the price "compacting" into a tight little ball, does that mean it's going to move?? If so, then the Bollinger Squeeze is correct.. BUT.. I don't believe so..

Why?

Swimming Pools.

There are two types of people, whose who stick in their toes to check the water temp, scream a little, then stick their toes in again, and then SLOWLY, one step at a time, screaming all the way, get into the water. And then there are those that belly flop in. They get a charging run, and then PLOP. But both exist in the market, and so... I think that's what is happening with a headfake in the BB Squeeze.. The people tip-toeing in, move the market in one direction, but those that cannon ball in, picked the other side, thus the whipsaw.

So I guess it's my ambivalence..

So I'm going to run down the list again just to flush out what I already know.

Two types of traders..

1) Fundamental ~ Those who trade based on news
2) Technical ~ Those who trade based solely on charts

Of the Technical Traders there are 3 types:

A) The Contrarian ~ They think since everybody loses, then just do the opposite of what everybody is doing and they'll win.
B) Top and Bottom Pickers ~ Those who try to pick tops and bottoms.. Or as they say, those who will soon be cotton pickers..
C) Trend Traders ~ Those who get in and ride the trend..

Of all of these, only the trend traders make money in the long run..

So, now it's a multiple choice. A,B,or C. I can try to find out WHY trend traders make money, or I can try to find out why Contrarians and Cotton pickers don't make money..

Like taking the SAT, let's start by eliminating what we know to be untrue.

Contrarians don't make money because... If the market is in a bull run, in theory minds, they are looking for the bear already.. That pretty much would then, make them PRE-Top Pickers wouldn't it? (Or I guess Pre-bottom pickers if it was on the short side).

Why don't Cotton Pickers make money?? Well, TECHNICALLY, if they had a crystal ball and could pick the tops and bottoms, they'd be the richest out of all the groups.

Why don't fundamentalists make money? Because we have no clue if news was leaked out first, or how the market will react to the news.. and as forex is global, how can you possibly keep track of all the news and how it effects the price? Assuming it DOES effect the price, then wouldn't the news be reflected in the price already? If the news __IS__ reflected in the price already, then wouldn't you then become a technical trader?

So..

On to TrendTrading.

So I know the secret sauce has to do with Trends.. I knew that from the first 2 posts in this journal.

Characteristics of good Trend Trading:

1) ID the Trend
2) ID the End of the Trend
3) MM MM MM MM MM MM

So I'm going to start tackling these one at a time..

How do I ID a Trend??

First, let's define a Trend.. Webster's Dictionary..

So, general direction or general course.

So a trend has a prereq of DIRECTION.

to veer in a NEW DIRECTION. So, again, a Trend requires the identification of a direction, and of a change in direction.

I will assume that the market has 3 directions:

1) Up
2) Down
3) Sideways

Or as my friend says, the 4th is seasickness direction..

I think the way to deal with the 4th is to just not be in, or to have tight stops with good money management.

We have to have a system that gives 3 ID:

Up, Down, and Sideways.

Up is down, and down is up, in forex.. so let's deal with sideways first..

So we are at pattern recognition and trends.. Just like the first two authors said! I've come full circle!

So my system will have a narrow range.. a tunnel. GOT IT.

It will have something that either points up, or down..

So I need a system that has a tunnel and 2 arrows.

GREAT!! My system is starting to take shape.. I just need to get everything ready, and then add the secret sauce...

I have to say, the ability to program is one of the greatest freedoms on the world..

So.. about the tunnel..

I just programmed in SART. The Self Adjusting Ranging Tunnel.

I took the 34EMA.

I took the 3 previous bars of the current bar, and I averaged them. I then plotted a % based tunnel around it..

The theory behind SART is; if you are within a range, then you are probably close to the previous 3 bars.. if you are moving and groovin' then you are probably outside of the average of those 3 bars.

Pic attached for you curious folk, and the VT indicator is attached as well..

I wanted it to be self-adjusting.. which it is!

I "stole" the idea from
http://www.traders.com/Documentation...e/Rae9906.html

Yes, it talked about bonds, but I stole the self adjusting tunnel idea from here. Didn't steal actually, it just inspired me to think a little differently..

Now I's gots mes a tunnel y'all..

Hi twoblink, Great thread. Thanks.

As you have mentioned time on several occasions and seem very scientifically inclined I thought you might find some of the ideas at

http://www.mesasoftware.com/technicalpapers.htm (advanced price filters including non -linear price filters) and
http://www.cycle-trader.com (non linear time concepts, 4D geometry, tunnels)

interesting if you are not already aware of them. They seem right up your alley. Good luck in your search for the "secret sauce"!

Hello Enroth, thanks for the links, I will take a look at them, and steal whatever ideas I find that are worthwhile.

So I am now at another dilemma (geez.. how many can a guy have a day??)

Two routes:

Route 1 ~ What I deem the EKG route) I keep going in a certain direction as long as I keep getting a signal. Once my signal "flatlines", then I'm dead, and get out of the trade.

Route 2 ~ What I deem the "Forrest Gump" route) I get a signal to run, and I just run my arse in that direction until someone tells me to stop.

Of the two, I think Route 2 is more profitable. The reason is, a lack of signal does not equate to a signal in the opposite direction. Of course nobody said it did, but they are different.. Exit when an Exit signal appears..


So I'm going to pick route two..

Woke up really late today.. Upset at myself for that, but at least my skin is good today!!

Today might be "paper day". I had worked on SART all last night, and at the end of the night, I was unhappy with it. Oh, it was a system that made money, but it was one that gave up too many pips.. I am the type of guy that eats EVERYTHING on his plate, no exceptions, and I can't stand wasted food, let alone wasted pips..

I'm going to print out 4 graphs:

1) One that shows a clear trend
2) One that shows a sideways (BBand Squeeze like) movement
3) Choppy market
4) Trend and Reversal

I am going to work these guys backwards. On these paper charts, I will DRAW the signal lines myself, and THEN generate the equations which would produce the said signal lines.

Start with the desired outcome, and work backwards.

As for the links Enroth gave me.. WOW.. those guys are nuts, I guess I'm "sane" compared to them!

I am however, searching for simplicity...

Well.. drawing the desired signals when given a blank chart, is more difficult than I thought!!

However, today, I am engrossed in the concept of the tunnel.

The tunnel should do two things at least:

1) Prevent you from getting in during a sideways market
2) Confirm (by the crossing of the tunnel) of the trend direction.

Now, I have officially decided that the tunnel size and width, should vary, and not in equal amounts... The top and bottom of the tunnel should be unbalanced. Balanced tunnels don't tell me anything more about the market, but unbalanced tunnels do; sort like Bollinger Bands..

So today is officially tunnel day..

To have an unbalanced tunnel; basically we can use Open vs Close prices.. They are almost always different, and the difference is bigger on the moves. The problem is, how do I apply them differently on each side of the tunnel?

The reason the tunnel can't be even, is because the price isn't.. Bulls vs Bears..

The tunnel needs to be uneven and self-adjusting..

I am currently having, what I call "The V" problem..

When the market moves in a downward trend, and you are in, and then the trend goes "V" on you.. the problem is, my current SART doesn't really exit until the top of the right side of the "V". I need it to stop me out on the turn around.. as close to the bottom of the V as possible. The PROBLEM is that MA's are lagging.. and so if you were deep like 55EMA, then a 5 bar move constitutes only 10% ~ 18% of your movement. If you were doing down fast, and then coming back equally fast; it's not going to change directions until like 7 bars later on the EMA. That's 6 bars too late IMHO..

So I'm thinking, I gotta give weight to the directionality of it all... But also the length of the bar.. a uber long bar, is different than a short doji..

The other problem is, the width of the tunnel.. or I should say "the height" of the tunnel..

I smell ATR variant coming up..

How do we tell the difference between an up bar vs a down bar? The open and close.. but if we have a up down up down, but within a range, it might have easily been an up up up up, but for when we drew when a bar opens and when a bar closes...

So it isn't the open and close that tells us the most, it's the high low.. Or I should say, it's the high vs the low..

Middleline := (Open+Close)/2;
abs(high-Middleline) vs abs(low-Middleline). That should tell me if the low is biased or the high is biased..

I then replot the bias as a tunnel.. gonna give her a try...

Well, that gave me interesting results, but not the results that I had hoped for..

The bias was erratic.. Visual OPR didn't see anything of value..

So let us go to the good book for inspiration...

God, what is the answer??

Well.. The Bible doesn't talk about Moving Averages, and the closest I can find is Ezekiel's wheel..

All 4 wheels looked alike. Wheel intersecting a wheel.

I don't have a wheel on the charts, but I've got some MA's.. What if based on the same concept as OCC, I stuck a 4 different types of MA's on a chart and see where they intersect? Why not?

That was interesting.. of the Simple, Triangular, Exponential, and Weighted, (all at 13 Close) I thought the weighted gave the most accurate movement..

I am officially unhappy with the tunnel concept. Having to cross the tunnel, you pay a toll. It makes you get in late, and get out late, and if the market is trending but dribbling, you are in the tunnel the entire time and you miss out on the pips..

Drawing on the charts isn't helping... Too many options..

I need to limit those options..

Well, not yet.. But gonna get there.. Every journey starts with the first step right?

The "V" problem though, is basically the problem everybody faces.. If I can solve the V problem, I'm 90% home. (where did I leave those Ruby slippers..)

So Basically, the system's gotta move, and have a follower.

Like I said before, a Vodka and a Chaser.. Now I think about it, OCC is pretty damn good!

So it's the relationship between the two though that's going to be the key.

I have a theory.. I'm now going to go program my theory...

Yep.. just as I suspected.. OCC is a difficult bastard to beat :

I've made a few modifications to OCC, and every iteration is a slight improvement...

OCC is actually built on a very sound foundation.. Bulls vs Bears. reflected in the open vs close.

But I still got that choppiness. I'm wondering if that's something I can get out with a little club soda and some rubbing...

I've been using the Heiken Ashi bars. I find them to be superior over the regular bars. I gain a Trend confirmation of sorts from them. I'm ALMOST ready to say, let's trade with just 1 MA and HA bars!!

I think it's actually quite possible.. But I still haven't nailed down the correct timeframe.

I (however) think the 15M timeframe is not the answer. I think the correct timeframe is somewhere between 1H and 1Day. You gotta give it time to gather some juice..

I'm waiting however, for the price to whisper sweet nothings into my ear. Well, don't whisper sweet nothings, please whisper "sweet somethings"...

What are you trying to tell me? Timmy fell down the well? Grandma's fallen and can't get up? What boy? What are you trying to tell me???

<sigh>

I might be living with OCC for a while.. (Don't I make it sound like herpies??)

OCC it solid and will generate 1000 pips a month on 4 pairs. That's the "backup" plan..

But that doesn't end my quest for the sauce.. In fact, it brings me more freedom... It's like you know you got a free pass, or no matter how badly you do on the final, you can't get worse than a B.

But I don't want to be a B student.. I want to be an A student..

So let's study some crossovers..

the equilibrium.. Everybody's -------------- here. If you are ___ instead, then you are short. So..

the question is, are you short because you are short relative to the previous bars, or are you short overall compared to the rest of the world? This then means, to answer that question, you gotta look once again at the longer timeframes..

Tide-Wave-Ripple. If I'm trading the wave, watching the tide so I don't go against it, and entering at the ripple.. there's a problem..

Oh.. I now see the problem that I never saw before..

If the TIDE was going to reverse.. it would start with the ripple reversing.. and so if you are waiting for the ripple to retrench, and the move, how do you tell a retrenchment from the start of a tide reversal?

Deduction vs Induction problem..

It's smarter then, since Deduction is more accurate to start with the Tide, and work your way down?? But indicators lag, and so if you did that, you might not be aware of the Tide change, since the tide change always begins with a ripple change..

So is the correct conclusion:

Watch the Tide for the trend, watch the ripple for the trend change?

I think it is..

So here's the perfect world indicator..

I need my indicator to do that..

So forget the tunnel for a moment.. and let's see.. I have just drawn the ideal line. I just need to now, dissect it, and see what it does at certain points, and what it doesn't do..

I'm going to start putting MA's and tweaking stuff until I can find curves that total up to what I just drew..

Sum of curves, average of curves etc... there's gotta be a way..

"Your system sucks. I tried OCC today and I lost 20 pips"

"What parameters do I use? I haven't read the thread so can you tell me?"

"I'm losing money with your $hitty system, what are you going to do about it?"

OK people, I have to say, that the fact that you morons are still alive, is a testiment to the fact that the world is too safe and that Darwinism is in fact, a lie. The thought that you people have kids makes me lock my doors at night.

The total lack of responsibility and though is just utterly amazing to me. If I yelled "hello" into one of your ears, will it echo like the Grand Canyon?

You know what.. I'm just going to come out and say it. Don't use anything I write here. In fact, I'm actually writing so you all would lose money. There happier now? Yes, I'm evil, yes I'm stupid, yes yes yes and yes.

Now will you all just leave me alone?

Go watch the tv show "IT Crowd". You people are the stupid ones that call, and force me to answer "Have you tried turning it off and on?"

If you are stupid, that's ok. . . it's not a sin. But the fact that you try to run your stupidity on me __IS__ a sin. And a big one at that. My IQ drops when I read your PM's and IM's.

Like I said in my OCC thread. If you are illiterate, that's not a crime. Buy Hooked on Phonics! Hooked on Phonics, worked for me!! Watch Sesame Street. Watch old reruns of Mr. Roger's Neighborhood. Whatever you do, do it at least 100 yards away from me and my journal.

And please, don't post in my journal that I'm a sour puss or something like that. I will just ask the admins to delete them.

This is my journal. If you don't like it... START YOUR OWN...

If you don't like my system, guess what??? Stop using it. Start your own.

Start your own people, start your own...

You all have turned a productive day into a non-productive one..

Hull's Moving Average, averaged with Moving Average = HMAMA..

So I've got HMAMA going, and I kind of like it. It's not so sticky like HMA, but it's not so laggy like MA.

HMAMA= (HMA+MA) / 2

Why HMAMA is effective:

One of the problems with MA crossover methods is that the MA's really lag. And when you have a sideways movement, you get the flip up/flip down problem.. and get thrown around along the line. HMA moves really smoothly and moves quickly, and so HMA doesn't have the same problem. But HMA moves too fast and sticks too tight IMHO for it to be an effective xover indicator. Also, it tends to overshoot on the big moves.. MA doesn't.

So when you combine the two, you seem to get the best characteristics of both.. (or maybe you will end up with the worst characteristics of both, not sure yet).

But so far, I LIKE IT.

If I traded it like a regular MA xover.. it seems to work well..

I consider an entry signal to be when the body of an entire bar has crossed over the HMAMA. The exit signal, the same.

HMAMA 34 on the 1 Day charts seems to work beautifully.