my journal 2

Yamato · Feb 19, 2011

Yes, I see what you are saying. By the way, your name and how you used the term "mean reversion", as opposed to "trend", made me think that I may have finally understood once and for all what "mean reversion" means. I suppose that is what I call "bouncing". I do have both "trend" systems and "bouncing" systems, and other systems, and I trade them all (funds allowing). Here's my 9 groups of systems/strategies so far: ON bounce, S/R bounce, Opening Gap, overstretched, Range Breakout, Volat.Breakout, WeekDay Bias, WITH ID trend, WITH ON trend.

If we were to group them in terms of bounce vs trend systems, it would be as follows. Bouncing systems: ON bounce, S/R bounce, Opening Gap, overstretched, WITH ON trend (yes, this is "bounce": it's not a typo), WeekDay Bias. Trend systems: Range Breakout, Volat.Breakout, WITH ID trend. Most of my profit comes from bounce systems, because even if you enter with the trend, you want to enter on a bounce anyway rather than when the market has already gone a long way in a given direction. Those 3 categories left instead do just that: they follow wherever the market has already gone, even though "range breakout" systems do it with surgical precision. So ultimately the famous saying "the trend is your friend" only applies to 2 groups of systems I have. My idea is that the trend is your friend in moderation. It's best to have the trend on your side, but also to rely on other friends, such as hour of the day, support and resistance, and as many friends as possible.

Having all these systems looks very good, but I must add that, as I have mentioned repeatedly, only a minority of them are profitable, even less are very good, and I have only recently become profitable, and that may not even last, due to my past (and maybe future) as a compulsive gambler. Yet I am not a compulsive gambler in any other area of my life, so that is leading me to conclude that maybe I was just calling myself that, because i was desperately looking for a way to increase my capital and make money, but without succeeding. A hard-working trader may look like a compulsive gambler, but once he finds out how to make money, he will make money. And I think that may be my case, even though I did have many symptoms of compulsive gambling, such as having blown out my account dozens of times.

Regarding your combining of systems, I don't know if I got it right this time, too, but if you mean putting together two different edges, that is something I never do. Whatever edges I come up with, even if they produce just 3 trades per year, i keep them as they are and move on to a new system. I keep all edges separate.

Yamato · Feb 19, 2011

final S/R bounce code

Continuing from here:
http://www.trade2win.com/boards/trading-journals/85510-my-journal-2-a-207.html#post1429464

Here's, simplified, the final concept and code (no values or it would be too easy and unfair to all my work):

Inputs: test(0), test2(0), test3(0);
vars: sample(0),pre_SR_bars(0),post_SR_bars(0),entry_after(0);
sample = xx;
pre_SR_bars = xxx;
post_SR_bars = xxxx;
entry_after = xxxx;

If marketposition = 0
and low<=lowest(low,sample)[1]
and lowestBar(low,sample)[1] < pre_SR_bars
and lowestBar(low,sample)[1] > post_SR_bars
and time >= entry_after and time <= 1500 Then buy This Bar;
If marketposition <> 0 and time >= 1600 Then Exitlong This Bar;

If marketposition = 0
and high>=Highest(high,sample)[1]
and highestBar(high,sample)[1] < pre_SR_bars
and highestBar(high,sample)[1] > post_SR_bars
and time >= entry_after and time <= 1500 Then sell This Bar;
If marketposition <> 0 and time >= 1600 Then Exitshort This Bar;

Basically we take a sample over which we poll the highest high and lowest low but we also require them to have been in place for a while (first part of the sample) and to be in place for a while (last part of the sample).

Maybe I'll explain it one more time: we require the lowest low (support) and highest high (resistance) to neither be at the very start nor at the very end of the sample.

That is why we require a buffer of x bars before (pre_SR_bars) and after (post_SR_bars) the highest high or lowest low has taken place. If that buffer is not in place, the system does not work.

But since the lowestbar and highestbar functions are equivocal in that they return a value of "bars ago", I'll improve the code even more by writing it like this (in red I also added a "print to file" verification code):

Inputs: test(0), test2(0), test3(0);
vars: sample(0),maximum_bars_ago(0),minimum_bars_ago(0),entry_after(0);
sample = x;
maximum_bars_ago = x;
minimum_bars_ago = x;
entry_after = x;

If marketposition = 0
and low<=lowest(low,sample)[1]
and lowestBar(low,sample)[1] > minimum_bars_ago
and lowestBar(low,sample)[1] < maximum_bars_ago
and time >= entry_after and time <= 1500 Then begin
buy This Bar;
Print(File("E:\downloads\financial_software\tradestation\EURO\mydata.txt"),
" min,max,sample bars ago are days", (Datetojulian(date) - Datetojulian(date[minimum_bars_ago])):2:0, ",",
(Datetojulian(date) - Datetojulian(date[maximum_bars_ago])):2:0, ",",
(Datetojulian(date) - Datetojulian(date[sample])):2:0, ",",
" date entry: ", Date:7:0, Time:5:0,
" min date ago: ", Date[minimum_bars_ago]:7:0, Time[minimum_bars_ago]:5:0,
" max date ago: ", Date[maximum_bars_ago]:7:0, Time[maximum_bars_ago]:5:0,
" sample begins: ", Date[sample]:7:0, Time[sample]:5:0,
" lowestbar is ", lowestBar(low,sample)[1]:3:0);
end;
If marketposition <> 0 and time >= 1600 Then Exitlong This Bar;

If marketposition = 0
and high>=Highest(high,sample)[1]
and highestBar(high,sample)[1] > minimum_bars_ago
and highestBar(high,sample)[1] < maximum_bars_ago
and time >= entry_after and time <= 1500 Then begin
sell This Bar;
Print(File("E:\downloads\financial_software\tradestation\EURO\mydata.txt"),
" min,max,sample bars ago are days", (Datetojulian(date) - Datetojulian(date[minimum_bars_ago])):2:0, ",",
(Datetojulian(date) - Datetojulian(date[maximum_bars_ago])):2:0, ",",
(Datetojulian(date) - Datetojulian(date[sample])):2:0, ",",
" date entry: ", Date:7:0, Time:5:0,
" min date ago: ", Date[minimum_bars_ago]:7:0, Time[minimum_bars_ago]:5:0,
" max date ago: ", Date[maximum_bars_ago]:7:0, Time[maximum_bars_ago]:5:0,
" sample begins: ", Date[sample]:7:0, Time[sample]:5:0,
" highestbar is ", highestBar(high,sample)[1]:3:0);
end;
If marketposition <> 0 and time >= 1600 Then Exitshort This Bar;

Yes, this is how I work. This is the living trading legend travis. Yes, this is me, a goddamn marvel of modern science.

Was all this worth it for just 1000 dollars of profit per year per system (return on Investment of 25%)? Yes, it was. 90% of wins, profit factor above 7. It is a rudimentary system, it doesn't trade often, there's so much potential that is not being exploited, but... it basically never loses and so it should be traded.

Now comes the out-of-sample. I won't come back to tell you about it unless it fails really badly, in which case i will not be a trading legend anymore.

Three out of three passed the out-of-sample so far. The legend of travis keeps on building up...

Nope. The EUR failed with its settings, but it is safe to use the other forex systems' settings. So the EUR is all right, too, even though a bit disappointing.

Nope. GBP fails as well. No more travis the living legend.

EUR and GBP pretty much fail to produce profit. The others do great but trade once a year. Not good enough.

Not worth it to implement this whole thing for just one trade per system per year.

One last chance with gold and hopefully it will work so I can pass on the genes to the next generation of systems, maybe one day I'll use this concept again so it must not become extinct.

Perfect example where you save a year of hopes and waiting and hours of work because you used the out-of-sample. Very good experience in a way. More or less 50% of my systems fail at the out-of-sample stage. When there was no out-of-sample two thirds of my systems failed at the forward-testing phase.

Now only one third fails at the forward-testing phase.

No...!

Gold fails as well, and even worse. The worst one of all:

Dude, after all this work, the "S/R bounce" group of systems will not exist at all. I will not have 82 systems, but at best (if the other group will work) I will have 76.

Of course I have saved all this work for future reference, but for the moment the S/R bounce systems are being archived and no longer studied. At least I spared myself the trouble of automating them.

Yamato · Feb 20, 2011

still alive and kicking...

I am pretty pissed off about the "S/R bounce" systems not having worked.

Now I want to get to the other group right away. We're talking about the "WITH ON trend" systems.

I will get to the out-of-samples right away.

Ok, CL passes. Makes 5k per year.

NG passes. Makes 5k per year.

GC passes, but not good enough to be worth trading: makes 500 dollars per year.

And with these we're at a total of 73 systems. The rest (GBP, ZN and ZN TWIN) I will do tomorrow.

[...]

Resuming from last night.

GBP is like gold: profitable but not good enough to be traded, and thus discarded.

ZN and ZN twin will work, so I'll end up having 75 systems.

Ok, first I tried the reversed one (it belongs to the "overstretched" group), that says that if we've been falling for a few days in a row and at 8 EST there's a reversal, then you can bet it will continue for a day. It works all right:

Not as much as the others, because it only makes 1000 per year, but it also requires an investment which is one third, so it is almost as good.

Now I'll try the original system, that says that if we've been falling for days and today we rose but it's 1600 EST, then we'll just keep on falling for the next few days. You see what I mean? If we've been falling and we reverse and it's 8, then it lasts a while, but if the same thing happens and it 16, then it doesn't last. Strange, huh? But it works, so I'll use it. First I'll make money, then I'll wonder why. I don't want to lose money just because I need to find out something that makes sense to me.

Nope: it fails. Or rather: it makes very little money.

Too bad. I only came up with 3 more systems.

Out of 11 systems, despite 90% of them being profitable, I only came up with 3 tradable systems. Here's their out-of-sample charts (the out-of-sample starts at the vertical red line):

CL:

NG:

ZN:

74 systems and counting... however only 25, one third, are worth trading. Maybe even 30.

Yamato · Feb 20, 2011

A documentary means it is the truth. MM's movies are NOT the truth. What he does it take a few facts and then twists them to fit his agenda...

Nonetheless his movies are well made and he states his points clearly. It doesn't matter if he's rich and fat.

No, wait... in the second hour it has a lot to do with derivatives, literally.

1 hour 33 minutes: "...a financial coup d'état".

2 hour 1 minute: "... I refuse to live in a country like this...".

Yamato · Feb 23, 2011

I can't believe I haven't written for so long: 3 days.

I suppose the fact that Vito is not in my room any more is slowing my metabolism or whatever it's called.

I don't feel as much the need to complain, and that was half the fuel behind my posts.

This week there's big events in my life. I am meeting the investors. After blowing out my own account. The experiment of them holding a golden share did not work, because after a few weeks, I felt the urge to step it up and even gamble, and I wrote them "either let me do it or I will return your share to you and do it anyway". They let me do it, I did ok for a few weeks, bringing my account from 5000 to 8000, then blew it out a bit, wired more money, then brought it back to 8000, then totally blew it out with Oil on Monday (+10% in 2 days), by going short on it when it was a +2%, because I thought "hey, on Mondays Oil falls, and it can't go any higher than this". Then it went up 10%.

So I blew out my account, or close to it (had 2000 left), then wired to the investors whatever was left of it, returning their share to the day I blackmailed them and said "agree or I'll do it anyway".

Amazingly, back in gambling mode (in a state of despair), right before the wire went out from IB, I managed to make 700 dollars yesterday, so now I can actually still trade despite everything going wrong on Monday. If I had to invest whatever money is now left in my account, it will be on Natural Gas going up. I would buy one small mini contract (one quarter the leverage). It seems like my destiny in my own account is to go from 3000 to 8000 for the next few decades. I get to 8000, then get frustrated and for some reason need to bring it back down to 3000. Or maybe it's been happening for other reasons and it will stop happening.

So I'll meet the investors, and hopefully they won't judge me by my appearance, because I feel insecure - I mean I am a much better trader than male model. They'd be disappointed if they expected me to be as good-looking as my systems are.

Another risk, since we already get into arguments via email, is that we'll get into arguments by working together for so many hours (it will be 24 hours and I am taking a day off from work).

The systems are close to six thousand dollars of profit today, so we're getting into professional mode and that is why we're meeting I guess. I guess this meeting goes in the direction of me getting paid. And things getting big. Even though I fancied the idea of never meeting people in person and being able to run a multi-million dollar business via email. Just to be able to say: hey, I got paid thousands of dollars but I never met these people actually. After tomorrow, this won't be the case and the power of the internet will show its limits, in the sense that people are not ok with handling so much money via email and chat.

Actually it's a wonder that we already went so far, with me having access to their account, trading their capital, without ever meeting them and not even identifying myself univocally (with documents and so on). "Who's managing your account?". "This guy, travis, we met him on the internet. We don't even know what he looks like". Well, I liked that - never meeting people. But I guess I have become too successful with this automated trading to keep on being just a nickname, "travis the fund manager". The time has come when they want to meet me. For what? To see if I am good looking and if my face is trustworthy enough to be running their account. I think this is pretty funny, but most people would say it's funny the other way (not meeting people).

I am uncomfortable about meeting people, especially when I don't know them. I'd rather meet people whom I already know. So basically I don't like to meet anyone new, because you can't know them unless you first meet them.

I remember there was this guy when I was in the states, his name was fabio. He was considered "hot" by the american women in the nineties:
http://en.wikipedia.org/wiki/Fabio_Lanzoni

Now if I looked like him, I wonder if they'd trust me more:

Yeah, I am nervous. I would like it better if they met my parents, and if I could let my parents be in charge of the negotiations and interactions. I am only good at running my own journal, with the moderating powers, being able to block people and so on... in real life, things are more complex: I can't use the "undo" button, I can't block people. I can't even get disconnected in case of emergencies. You have to spend time with people non-stop, for hours, and they're staring at you and talking the whole time. It's a very embarassing situation to be around real people, I mean outside of the computer screen.

I guess this is the price I have to pay for scaling up. If everything goes according to my hopes, within a few days we'll be scaling up, and making about 4000 dollars per month, which would be enough for me to quit my job and live at the island.

Adamus · Feb 23, 2011

Plus ça change (plus c'est la même chose).

Yamato · Feb 25, 2011

everything went well with the investors

Less writing from now on about the details of our partnership
As the partnership with the investors grows bigger and more professional, I can write fewer and fewer details here, even though I owe this partnership to this very journal.

Let's just state a few general concepts then.

First of all, the investors heard with their ears the neighbour bitch slamming her door, and they agreed about her being a bitch.

Agreement on payouts and scaling up
We also reached an agreement about scaling up, and payouts. Very much a compromise, where I still contribute to the safety profit cushion (to cover drawdowns), but in a decreasing amount as the cushion increases, from the present maximum of 33.3% to a future minimum that will get closer and closer to 20%. When the cushion ceases to build up, then we all cease to contribute to it, even though it has to stay in place until "the end". The "end" would mean the end of the partnership, when everything gets dismantled and I get to take a piece of the cushion home, to help my sleeping habits.

Peace of mind mantra
I was recommended to somewhat find peace of mind in many ways. That was a constant leitmotif of our meeting. Peace of mind about the neighbour bitch, about the payouts, about what I want from trading. The typical trading mantra in fact, which I've often heard before, about how you can't make money unless you really are at peace with yourself. And I agree, but i want peace first of all because I want to sleep better. Then, secondly, because I don't want to blow out my account anymore.

No more insomnia and nore more people laughing in and at my face
Speaking of sleeping, today we managed to find a good family restaurant right near saint peter, despite it being on a street with nothing but tourist traps. I could go there again actually, even though it is a bit far from my house (10 minutes walk). Anyway, I mentioned this because there's something that is still bothering me: a waiter outside the previous restaurant, a tourist trap, as he was talking to me, was almost laughing in my face, by how screwed up my face looked. I guess it's my usual appearance, but you have to add to it that today I had slept 5 hours or less. And I don't want that to happen again and again, as it's been the case in recent years.

I always find good enough reasons to not sleep properly: today it was the anxiety about the business meetings with the investors, and the stomachache from the cheap wine we had yesterday (I removed an earlier post about this stomachache, which I wrote during the night, while sleepless). The previous day it was for another reason... I always seem to find a reason to suffer from insomnia. I need to be in control of my sleep, and therefore in control of my inner peace, which is what matters as far as sleeping. What do I want? How do I get it? Can I be happy about the day? Why should I keep myself from sleeping and how will it help me face my problems the next day? It will make matters worse.

Trading has always been a good reason to not sleep at night, and a valid one, but if this is true about overnight discretionary trading it is much less the case for both intraday discretionary trading (no overnight positions open) and automated trading (no involvement in the choices, no big bets). So I am definitely in a position to sleep at night even as a trader, as I don't need to do any overnight trades.

CL_ON_3 assessment problem
On a side note, we did not include CL_ON_3 in this phase of scaling up, and it is acceptable to hear a "no" among many "yes", but I don't really understand why and we didn't have time to debate it, because we had to focus on many other things. CL_ON_3 has a sharpe ratio of 3 for the past, 2002 to 2010, but, since the investors were considering only the years 2006 to now, the sharpe ratio got worse and they said they won't enable trading for it. I can't understand why we are excluding a system that seems so good to me. We'll have to debate this further in the future and double check if their data (which caused the dropping of the system) is right. But I am not going to bring it up now, because everything has been done either my way, or with my agreement (after i understood that it was a good idea), so there's no point in ruining the peace between the families by bringing this up right now.

Great week, and NG finally going my way
Other than this meeting with the investors, it's been a great week for the systems (see first chart below), and my NG trading recommendation is paying out finally (see second chart below). I am quite proud of having posted it on Adamus' journal once again just yesterday:
http://www.trade2win.com/boards/trading-journals/86996-deadline-june-82.html#post1440224

Yamato · Feb 26, 2011

beatles cover versions

Nice versions:

http://www.youtube.com/watch?v=_T8XfBVM-ug

Yamato · Feb 27, 2011

Hello, Goodbye

http://www.youtube.com/watch?v=fAaurWomRc8

Yamato · Feb 28, 2011

In trading as in life what matters is what you do. Not what could have done.

I was just thinking this: today I looked at the CL, knew it would fall, but did not want to take the risk. I would have made thousands. I felt awful about it, but then I realized that this is a crazy activity of mourning what could have been.

I should instead focus on what I will do. In trading as in life, what matters is what you do. Not what could have done.

In fact, if you trade properly you will be missing a lot of opportunities and this needs to be done in order to also avoid some risks. You cannot have a situation where you just seize all the good opportunities and avoid all the risks. If you never want to miss an opportunity, you will incur so many risks as to blow out your account endlessly. Now that i am trading properly, and playing it safe, I find myself missing many opportunities out of prudence.

You see, the risks are always there and the opportunities are always there, in other words volatility is always there, provided that you're monitoring enough markets.

Your profit is not given by how many opportunities you run after, but rather by how carefully you choose which opportunities to go after. By chasing opportunities you blow out accounts. By waiting and selecting opportunities you make money. An opportunity is always also a risk. You make money if you correctly evaluate how much of an opportunity and how much of a risk you have in making a trade.

Gambling, which is what I have a tendency to do, means only looking at the "opportunity" aspect of a trade. But a trade is not like a girl, where all you risk is being rejected and so you can try as often as you wish. In the markets, great opportunities turn to great risks, if things do not go according to your wishes.

This is what people summarize by saying "risk / reward".

To put it all together once more:

trading is not about grabbing as many opportunities as possible, but about discarding as many risks as possible.

This is like for my NG trade:
http://www.trade2win.com/boards/trading-journals/85510-my-journal-2-a-209.html#post1442236

It is on support, it's fallen for a month, it is ranging: big opportunity for a rise to the up side and very small risks. And so that is definitely one good trade, regardless of where it ends up going.

Yamato · Feb 28, 2011

Hello, Goodbye

http://www.youtube.com/watch?v=fup2KSfXTW8

http://en.wikipedia.org/wiki/Hello,_Goodbye

"I say high, you say low..."

Somehow this reminds me of trading. It sounds like "I sell high, you sell low...".

Yamato · Feb 28, 2011

a beautiful picture

Sorry for showing off again, but just as I've shown it when we were on the bottom of the equity line, I have the right to show it now that it looks so beautiful. It looks like an exponential growth. It's never looked so nice as it does now. Each dot is is a trade:

And this music is just right for my equity curve:

http://www.youtube.com/watch?v=3PIH649Hh1U

Mind you: this is just measuring our ability of selecting the right systems, because the systems traded during that equity line are not always the same ones. After getting rid of 4 bad choices, which had caused us 6500 dollars of losses, the equity line went up like a hot air balloon after dropping its sandbags.

Another interesting chart is the one showing the correlation between the markets and our selection of systems:

It might be partly a coincidence, but partly profit does seem to be affected by those markets going up.

This talented girl is telling me something wise: tomorrow may rain.

Well, actually we will see what happens to the systems if it rains (equivalent to eur and sp500 going down) and if there's a correlation even when it rains. For now, if I look at the forward-tested results, here's what I see for the systems we selected:

What i see is a drop that lasted less than 1 month, from the end of January to the middle of February, in 2010. Let's see what happened to euro and SP500 in that period.

EUR (to the USD):

SP500:

Yes, there must be some sort of correlation, because there, too, there was a sharp drop on both markets and that is why the systems suffered a month-long losses.

Now, since the EUR was falling throughout that period, we could just limit the correlation on the SP500 maybe, and find out what happens to my systems when that one falls.

Nope, wait. They both fall in May 2010 but my systems make money. Here's how it is: if they rise, the systems make money. If they fall, sometimes my systems lose, sometimes they do not. So the correlation is still there, but not enough to worry about it.

Besides, we're only trading half a contract on CL_ON_2 and CL_ID_3, so the actual drawdown goes from 4500 dollars lasting 25 days to... actually same time period and 5000 dollars. Damn. But CL is dangerous so it is safer this way.

Here's the exact combination of the systems we're trading with the exact contracts traded on a weekly basis (every dot is a week):

And, mind you, at least one third of the 12 systems traded now were added along the way, and do not show their trades from the start of the equity line in late July 2009.

At any rate, my guesstimate is that they would have made 100k in the roughly 20 months of forward-testing.

This would mean we're expecting, more or less, 5k per month, which will only be 4k because systems usually get worse - I mean you select the best ones based on past performance, so it is unlikely that exactly those that performed best will keep performing just the same way. It is more likely that some of the worst ones will get better and viceversa.

That is why I said to the investors that I am expecting about 4k per month in the coming months.

"Because the wind is high it blows my mind..."

Yamato · Feb 28, 2011

despite all this success...

Despite all this success... and the equity curve doing so well... or maybe precisely because of it, i feel restless and unable to sleep.

It's not like success will bring any normalcy into my life. My habit is going to work, being unhappy, not being able to do what I want to do.

This trading profit, a new event in my life, will disrupt my life. I want it, but it's going to disrupt my life. I don't know how much I can expect of it. If they told me Sharon Stone was going to be my wife, would I want her? I don't know. This is the situation now: I am going to have something very good in my life, coming from the sky almost. Something I've always worked hard to reach, but something I am not used to: profit.

I can't help being sleepless and scratching my head.

Luckily there's two other guys keeping me from losing my head and self-sabotage. Maybe I wasn't ready or maybe that is why I never reached profit on my own: I didn't want it. It made me nervous. Or maybe things just weren't good enough (money management and so on).

I would need a vacation or a part-time schedule ending at 1400 instead of 1500. This is hard on me. Having to stay there until 1500.

The investors act as auditors and keep me from tampering with the systems no matter what goes through my mind in these days of extreme success (see equity chart in above post).

The result is that if you invest 30k on your own you blow out your account, whereas if you invest the same amount or less with 2 other people you end up making money. It's funny, isn't it?

This is like eating things from the refrigerator. I have to keep it empty or I'll eat more than it's healthy. We don't always do what is in our best interest. We need little tricks to help us do it. Find me someone who does the rational thing all the time. I don't believe he exists. It's too boring to do the rational thing all the time.

You know what I mean? It would be like this: "You want an ice-cream?" "No, thanks: I am on a diet" or "No, thanks: it's not good for my health". "You want to have lunch with me?" "No, thanks: things get out of control when I have lunch with others, and I tend to overeat". "Do you want to have sex?" "No, thanks: there's a risk of getting diseases". Living rationally and in our best self-interest is like not living at all.

In this sense, trading does not allow us to live in a full and satisfactory way. That is why automated trading is easier. It forces to realize and accept that we cannot have fun while trading. Trading has to be boring and completely rational.

Luckily, thanks to investors I trade the systems, and thanks to systems being traded without tampering, when I am depressed or bored they trade just the same as when I am euphoric and happy.

In other words, 3 competent people, or even just 2, are better than 1 independent trader. It's quite unlikely that one trader will have the balance to make it through all the issues separating him from success.

I don't want to go to work. It happens every day. Every day, every night, every morning I say to myself "quanto manca?", meaning "how long more?". How long more do I have to go to work? I don't like to enslave my mind, and let it take orders from the bank. I feel the urge to rebel against the bank, against all authorities, and I've felt this way ever since i was a child, with my father, who repressed me too much. Then with teachers.

For now my need to rebel only plays out into insomnia. I stay up, I don't sleep... that is my way of rebelling. Can't quit my job yet. Can only increase my part-time one hour per year. Actually half an hour per year. Now I am down to six hours.

Because, the wind, is high... it blows my mind.

Yamato · Mar 1, 2011

latest discretionary trading attempt

This time I won't write the usual bible about it, nor try to turn it into a univocal method, because it would halt its development. I just want to note here that I've been noticing what may be a profitable discretionary method, which is close to the broadly used concept of "scalping", probably in the misused version of the term.

Characteristics:

1) it trades the CL
2) it trades on a 2-hour chart of 1-minute candles
3) it trades during intraday (half margin, more liquidity, more volatility)
4) you gather the most likely direction from any one or more elements from a variety of sources: the edges from my CL systems, static S/R, dynamic S/R, correlation with the ES and YM, correlation with the USD (I look at EUR), overbought/oversold situations, etc.
5) Once you gathered where it's likely to go, you wait for a big swing into one direction (opposite of where you will go), so you have overstretched levels on your side, on top of all the other edges you are using.
6) you then wait to see a 1-minute candle that, if you want to go long, is not red anymore, so either green or no color (it closes where it opens). You then enter long, in our example, (viceversa for short) at the end of that candle.
7) you exit when there's no more green candles, so that means at the end of a non-green candle, which also means a situation opposite of why you entered the trade: you see either red candle or no color (the open matches the close).

As a summary, the only univocal requirement for (let's say) a LONG entry is that, after a red candle, you have a neutral or green candle. Then you can stay in the trade until you see green candles (or neutral candles if it is at the start of the trade, e.g.: neutral candle makes you enter, then you get another neutral candle, and then another one... you can still stay).

An even shorter summary: enter with the first non-red candle, exit with the first non-green candle (viceversa for a short trade). The rest is up to you, in appraising what the most likely direction will be.

I believe I can make about 100 or 200 dollars per day with this method (more than my salary at the bank), with one trade per day, and have a success rate of at least 75%, with losers being on average about 50 dollars and winners being 150. This method takes a lot of waiting though: you can't just open a chart and jump in at the first opportunity. You need to discard a lot of opportunities which are not good enough.

Trades last only a few minutes on average. The least can be one minute. The most will probably be less than 10 minutes. I would say the average trade would last about 3 minutes.

This is potentially a 100% return per month, considering the margin needed is just a bit more than 2000 dollars. The drawdown is very small. It just takes a lot of waiting and observing.

It seems easy now, but it's not easy at all. If you don't wait long enough, you will end up taking 50 dollars losses every day for twenty days in a row.

Yamato · Mar 2, 2011

taking a short break

Taking a short break from work.

I am in my room all alone. Vito has been moved out a month ago already: excellent situation. And the lady, she's rarely here. I can focus like crazy and nothing is disturbing me. The door is shut, the lights are off. I am serene as far as my work situation.

Funny that my mind is wandering sometimes more, because when vito was here, I was extremely stressed out, but he forced me to work restlessly, because I worked in order to avoid talking to him. Now that I am able to relax, I do relax, but I am also less compelled to work nonstop.

What is lost to relaxation though is gained by being able to focus more. No one is there, three meters away from me, asking me bull**** work questions to try and interact with me (because he knows that if he asks for help, I will help him).

Actually, he still comes once a day, and tries to bother me, but it's all under control. He came today, pretending he was looking for the lady. I was working as usual and I said "she's not here" (obviously, since he could see it). But then he asked me how I was and I said "fine" and kept working. And this dick-head asked me "are you sure?". I said "yeah" and kept working and he left. That was a close call. If I had answered anything else, or hesitated, I would have given him the opportunity to bother me for an extra five minutes. But he taught me what to reply to make him leave fast and so I did, and he went on to another room to bother someone else.

The NG has been a big disappointment and did not go up the way I expected it to do. Everyone betting on it is now losing money and partly because of my trading tip. Hopefully today it will rise and recover from yesterday's loss. I am still at break-even.

I can't believe the NG can fall for that long, without ever bouncing. It's been falling for over one month. But of course if you look at the last 10 years of NG, you see that it is exactly what it does - going in one direction for weeks and weeks. It is the most volatile future I've ever traded, and the least zigzagging one.

Today, once I will get home, I will do one more of my tests of comparing the performance of a system with the performance of the underlying, to see how correlated they are.

Yesterday I compared ES_ON_2 and found that it's just a little correlated to the ES. It makes more money when it rises, but on average it still makes money if the ES falls.

Today I would like to look at the behaviour of GC_ON vs Gold, and YM_ON vs YM. Then I will give myself a haircut. Or maybe the other way around.

Now I'll go back to work, and keep doing it for another 2 and a half hours.

Yamato · Mar 2, 2011

going and going...

Gave myself a haircut.

Gathered all the data for the futures my 12 systems are trading live with real money. Now I will start comparing the performance of each system with the performance of the underlying future, to see how correlated they are.

http://www.youtube.com/watch?v=MSeGTrPBuOE

Getting there...

Happy. Happy so far:

1) ES_ON_2 correl 0.06: correlation causes no worries:

2) YM_ON correl 0.04: correlation causes no worries:

The bitch is still slamming her door, very violently. I can't believe that she doesn't do it on purpose. I hope she will plunge to her death in an elevator crash.

3) GC_ON correl 0.92: correlation cannot be detected but would seem to cause no worries. You see, all the GC does in the last 10 years is rise, but in the few periods that it falls, my system also reacts quite well, much like it does for ES_ON_2, in the sense that certainly falling by GC doesn't help it, but it doesn't always hurt it. So the correl function showing an awful 0.92 is misleading, because it's not my system's fault if it makes money all the time and gold happens to be rising all the time. I believe it will make money even with the gold falling. For example, look at the start of 2009 and also late 2008: my system makes money whereas gold is falling. This is what leads me to believe that it is mostly a coincidence that they both make money at the same time.

Mind you also that the two charts seem identical, but if you held 5000 dollars of gold, you would have made 500 percent, whereas with my system you would have made 1000%. Yes, because the two scales are different: the system's y-axis scale is to the left, and viceversa.

4) GBL_ID correl 0.56: worst one from the correlation point of view (gold is mostly a coincidence) but even more for its variability: we adopted it because it behaved very well lately. Because of it, I have started to calculate the correlations, using the correl function, and have added it to the previous studies as well, and have seen that it appraises the situation very very well.

The bitch is on the move. I can hear her heels walking nervously. I hope she gets into some sort of an accident with those high heels. The young neighbour bitch. As we say in italy, "the mother of the idiots is always pregnant". Similar to the point made at the start of the movie Idiocracy.

5) GBL_ID_2 correl 0.87: yet, here, too, I ain't worried about the correlation. The fact that GBL goes up and profit goes up is merely a coincidence. Like for GC_ON, in various places it can be seen how the system makes money despite the GBL going down.

Mind you, here I could use the whole data set I have available (before 2005), because the change in the schedule of GBL does not affect this specific GBL system (not true for the other two).

6) GBL_ON correl 0.80: here, too, just a coincidence. It's obvious that the system makes money also when the GBL falls, but excel can't know it, because we can't do anything about the fact that GBL basically rises non-stop for the period.

One more reason to believe this is the fact that both GBL_ID and GBL_ON are systems that go both LONG and SHORT so it would be even more unlikely that they're affected by a given type of market. In fact, GBL_ID is potentially a bad system but for its variability (that is why it has a low sharpe ratio) and not for its correlation with the GBL.

End of Part One and Summary

I can't do this whole thing in one day. I need to continue tomorrow, with the remaining 12 systems. Anyway, so far no worries from the correlations' point of view. We've just noticed one more time that the GBL_ID, if we look at its past, is not a very promising system.

Yamato · Mar 3, 2011

wait

It's been a long time... now I'm coming back home.

http://www.youtube.com/watch?v=4m4r17DFQLM

Coming up next, the analysis of the six remaining systems, in terms of comparing them to their underlying. Wait.

Continuing from previous post.

7) CL_ON_2 correl 0.89
The most correlated of all 7 tested so far, because its correlation does not come from the fact that both rise endlessly but to the fact that it falls at the same time the CL falls. Nonetheless, it is not so disgusting after all. Sometimes the rising and falling don't match one another, which is good and also, when the CL falls, the CL_ON_2 does not fall as much, so it is still far better than just buying the future and staying long on it, without ever closing it. In fact, if you are not fooled by the two y-axis scales (one on the right and one of the left), you'll notice that CL_ON_2 makes 500% while CL makes nothing at all (from 100 to 100). 500% is not peanuts, and that is why we're trading it live. Having said this, CL_ON_2 has the strongest correlation with its underlying of all 7 systems tested so far.

8) CL_ID_3 correl 0.21
Oh yes! This is an excellent one, for both overall profit (1200%) and lack of correlation. Just see it to believe it. In fact it is the best one we have and we should be allowing it 1 contract instead of just half (the small contract).

No wait, I was getting ahead of myself. The best ones remain the ES and YM. So far, all things considered, here's my ranking:
YM_ON
ES_ON_2
CL_ID_3
GC_ON
GBL_ON
CL_ON_2
GBL_ID_2
GBL_ID

9) EUR_ID_5 correl 0.79:
Here's another situation where the correlation is just a coincidence and I don't buy excel's appraisal of my system's correlation. You can easily see by looking at the chart that my system makes money even when the EUR is falling:

10) GBP_ID_5 correl 0.93:
Once again correlation does not mean much. If the value is low, then there is no correlation for sure. But if it's as high as this one, it just means that there is a little correlation and that when, like in this case GBP loses one third of its value, the system will lose a small part of its profit. So, crazy as it sounds, a correlation of 0.93, almost as high as it can get, is altogether acceptable. The chart and the results speak for themselves: while the GBP ends where it started, GBP_ID_5 made a profit of 1400%. Here's another system that we could double up.

11) NG_ID_2 correl 0.13:
And here's the masterpiece. No need to say anything about it.

12) ZN_ON_2 correl 0.69:
I should be worried about it, but after 11 systems, I realized that there's nothing to be concerned about as far as correlations, at least with the systems we're trading. Most likely they won't be affected by all markets coming down all at once or whatever will happen. Yes, there will a drawdown, but not a predictable one, not related to any way the markets will move. All right, the systems do behave slightly better if the EUR and ES are rising. Slightly better. But not enough to add an extra filter or anything like that.

The problem right now is not getting too excited. The systems almost have not had a losing week in two months. I bought some xanax because I know pretty soon I won't be able to sleep if it keeps going like this.

Yamato · Mar 4, 2011

WATCH IT: THERE IS A MISTAKE ABOVE: I WILL CORRECT IT 3 POSTS LATER, HERE: link

(Line added later as an alert - below is the regular post, previously written)

I am doing all right.

The systems are making money.

The systems are healthy, all of them, as can be seen from the two previous posts. The worst one is GBL_ID, which has been making nothing but money in the last 2 years. So that qualifies it as healthy.

Work is going great. No vito in my room, appreciation from my boss.

I also found a new sleeping recipe: a bottle (20 cl) of prosecco gancia, a melatonin pill, a xanax pill. It's not that heavy or dangerous and it will work for sure, for emergencies.

ayan · Mar 5, 2011

u r doing well travis.....best wishes to u....keep it up:clap:

Yamato · Mar 5, 2011

Yes, trading-wise something big is being achieved. Sleeping-wise I am still having a lot of problems. Actually trading keeps me awake, out of excitement.

Hopefully soon I'll quit my job so I can sleep late.

Something inside me awakes me before having slept 8 hours and keeps me awake, even when there's free time to sleep.

my journal 2

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Experienced member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Legendary member

Newbie

Legendary member

Similar threads