my journal 2

[Yamato]

reasoning on discretionary trading CL system

Continuing from here:
http://www.trade2win.com/boards/trading-journals/86996-deadline-june-85.html#post1474172

As I was writing on "Deadline June":

1) first I need to make enough money to live for the rest of our life OR AT LEAST I need to prove that I am consistently profitable on a monthly basis (producing the equivalent of a regular salary) and keep it up for 12 straight months.

2) then we can quit our job.

Anything else is reckless in my opinion. On the other hand it is ok to gradually increase your part-time schedule as your income from trading increases. Of course, it would be ideal to quit your job and then be able to go back to it in case your trading fails, but in most cases like in mine, that is almost impossible. My bank won't take me back if I quit my job, won't let me take a whole sabbatical year, and I suck at selling myself to any other employers. This is it: I either succeed for a whole year, or I'll have to work for the rest of my life. So far my combination of systems has worked only for 3 straight months, and my discretionary trading has never worked. For me the time has not come yet to quit my job.

For me an acceptable salary is 2000 dollars per month. From the systems, I could make somewhere around 1000 right now. Probably more in a few months, probably enough to consider it a salary, but as I said it has been going this way only for 3 months (break-even before). I would feel safer if I could add to it profitable discretionary trading on the CL (the discretionary method I was talking about). Once I will have a stable income of >2000 per month from the systems, and a stable income of >1000 per month from discretionary trading, then I will be able to quit my job: only if that keeps going for six straight months. Six times 4000 is about 25k set aside, which would make me feel safe. Also because then I can trade the systems on my own, so that would mean a salary from investors, salary from discretionary trading, salary from my own automated trading. Yeah, with 5000 per month i could quit my job. By the time that happens, I will have reduced my regular job schedule to as little as 5 hours per day. I can't ask for more part-time because they would never let me.

I need profitable discretionary trading not just as an income but because I want it as a personal satisfaction. I can't accept after looking at the markets for almost 14 years that I still have not figured out a way to make money via discretionary trading.

My 1-minute candlesticks method described here...
http://www.trade2win.com/boards/trading-journals/85510-my-journal-2-a-210.html#post1446762

should be improved because so far it has produced a very low edge. I don't want 100 signals with an edge of 55% every day, because, given spread and commissions costs, the daily profit will be zero, as it almost was the case yesterday.

I just need 3 awesome 80% opportunities per day, and then I will make money every single day, which is my objective. The setup of my 1-minute candlestick system produces several signals per hour, and I don't need to take them all.

Here's the questions I have to answer in order to pick the 80% opportunities:

1) which is the most likely direction for the day? LONG or SHORT? Then I need to focus just on that side of the market and prepare my entry with a market order.

2) I need to keep asking myself for every opportunity that I spot:
is this an opportunity like one in 5 minutes, like one in half an hour, or like one in an hour? Then I should only seize those rare opportunities that arise only once in an hour. That's right, because I only need 3 of those to absolutely make sure that every day I will make money.

With a 1-minute timeframe the opportunities are several hundreds a day. Can I afford to discard the bad ones and spot the best 20%? Yes, for sure. At least I will be able to spot the "potential" opportunities - I am not saying I will spot the 20% of winners, but just what look like the best 20% of opportunities.

But I will add a new rule.

Until now the rule has been

ENTRY when you see a reversal candle (opposite color or neutral)
EXIT when you see a reversal candle (opposite color or neutral)

I will now allow a larger requirement.

ENTRY at any time, beginning when you see a reversal candle to any time during the ensuing trend. So basically the only requirement is that you cannot be against the 1-minute candles. The rest is all allowed. Your required skill is in spotting the right opportunities.

I will now show plenty of examples for entries and exits.

Entries are marked with a green arrow and exits with a red one.

Let's first analyze the chart in a LONG-only perspective:




Let's now analyze the SHORT-only trades:



Maybe moving averages would be better. I am thinking about it... but less accurate. I don't think I can afford them.

I could also be looking at price alone or bid and ask alone, or the three of them.

I'm wondering.

No.

No point in slowing myself down with moving averages, and no point in looking at too much detail. All I have to perceive is price. Candlesticks are good.

I've thought about it for hours by now and here's my final answer. It has to be 1-minute candlesticks. You can enter WITH the trend at any time (or neutral), and you can only exit if you see a candlestick against you (or neutral).

By "neutral" I mean those candlesticks that open and close at the same price. So the rules have only changed compared to a few days ago as far as the entry, which does not have to happen necessarily AT the reversal point, but can happen during the trend.

I realized that candlesticks are better than price alone in that they make you see time, which is something that gets lost if you only plot price (which means trades).

I think that if i can focus on an intraday chart 2 hours per day i can get really good at this thing, provided that i follow these guidelines, that I pick with precision with opportunities, and that I don't trade endlessly like I did on Friday.

After all these years, I feel like I am going to ace this thing. Not that i haven't said it or felt it before, but each time I get closer and closer to making it happen and this might be the right time. You know, once this happens, there's no looking back - but I don't know if I want to teach this to anyone. It took me too much time to learn it.

Now that i have this thing to learn, I am looking forward to the weekdays, if it weren't for having to go to work.

Besides, this week might suck because the investors are implemented the multiple accounts thing on their account, and we might come across some technical problems.

 

[Yamato]

feeling worried without worries, as if i were addicted to worrying

As the older worries disappear I find myself creating new and simpler worries to keep my own mind busy.

Vito is out of my room, and now I am worried about a guy who's in another room who seemed rude to me on one occasion several days ago.

I haven't been hearing the bitch slam her door in a while, and I find myself getting mad at the maid who didn't show up on Friday, to the point of hating her.

And finally, I've identified a good 14 systems that keep working week after week, and now I find myself worried about how simple they are and fearing that such simple systems cannot keep working for much longer.

Also, I am worrying and wondering if I could and if I shouldn't actually work on improving them.

This all boils down to self-sabotage in practice, because I am doubting what actually made me successful as far as systems, as far as solving my problems.

Or rather, maybe it's different. I solved problems by worrying and now I just want to keep on worrying and solve more problems even if there's none, so I can only undo what I have done right.

I've been scratching my head for hours out of boredom and a lack of problems to solve.

I am actually worried about not having any immediate problems to work on. I am worried about not having any worries. As i said before, I was brought up to always stay alert and to feel guilty about celebrating things and relaxing.

However, once I'll quit my job and move to the island, I will focus on swimming, which is a far better activity than worrying about a colleague or a bitch next door. I am longing for the kind of worrying that doesn't involve other human beings.

 

[Yamato]

adding ma filter to CL 1-minute candle system

As in this picture:



Changing it. You can do any trading you want, provided the 1-minute candlestick closes are within the 9-period moving average.

Entirely forget the green candle and red candle thing.

Anything you want - provided the candles' closes have the 9-period ma in their favor.

 

[Yamato]

adding more to my discretionary system

I've been practicing on the edemo account today and have more changes.

The 1-minute candles remain. The 9 period ma remains.

Now the 33 period ma also has to be in favor, which pretty much tells me exactly what I have to do and when and will give me profitable signals provided there's volatility and not just ranges, otherwise it will be a disaster.



On the example above it would be mostly short signals, with short entries being allowed starting at the red circles, but not limited to them (anywhere after them, as long as both mas are in favor).

Basically I can do anything provided that those 2 mas are in favor, which force me to stay within the trend.

How do I know there's volatility? That could be measured as well, but there's no need to measure it since on the CL there's almost always a trend during the intraday period.

The moving average is not 33. Make it a 44 period moving average.

Anything allowed as long as the candles' closes have 9 and 44 mas in favor. That's my system right there.

Wait. this is not enough... the 9 period has to be in favor and the the 44 has to be in favor of the 9.

 

[Yamato]

new change

It will be 44 in favor of 9 and that is all as far as averages.

The other rule is 1-minute candles as I devised the method in its first creation. As soon as I see a reversal I get in and then out again in the opposite situation.

So 9x44 crossover in favor, and for the rest it's the usual 1-minute candle method.

 

[Adamus]

Re: new change

It will be 44 in favor of 9 and that is all as far as averages.

The other rule is 1-minute candles as I devised the method in its first creation. As soon as I see a reversal I get in and then out again in the opposite situation.

So 9x44 crossover in favor, and for the rest it's the usual 1-minute candle method.

I think you need to be careful to differentiate between developing your discretionary trading skills and inventing new automated systems. A computer can deal with MA cross-overs better than you.

I know you need to give yourself rules so that you don't get stuck in a bad trade unable to kill it off and take the loss, but you don't want to suppress your ability to detect opportunities.

I think discretionary rules have to be rules that are difficult to program.

 

[Yamato]

Yeah, with those simple and univocal rules i listed there will still be dozens of signals, so I still need my discretionary intuition to choose the right ones. But let's say that at least I have gotten rid of the bad 75% of signals and I have the market on my side, provided it's not a ranging day (which on CL is rare). With the crossover rule I get rid of the bad 50%. With the candlestick rule I get rid of another 50% of situations, so the outcome is 50% of 50%, which is still quite a bit of signals. I've practiced it all day long while watching movies and so far this seems the best set of rules.

Good movie here:
http://www.putlocker.com/file/O44N632NJ3KSG80O

 

[Yamato]

No more 1-minute candle method. Now it's the first time we fail to make subsequent low which is less or equal... well, a general feel for it. I will write it down next week, once I get more practice with it.

So far here's the outline.

1) first I get the idea for the trade, only when I see that it really wants to go up or it really wants to go down.

2) then I make my trade within the mentioned framework. If the guidelines don't allow a trade in the direction I am forecasting, it means it is not safe to do so. If the guidelines prescribe to close my trade, regardless of what I am predict, then I have to do so.

The guidelines are clear. It has to have 9x44 in favor, and the candles have to be on its side as well. Each single one of them. That is why I can only hope to make 100 dollars per trade: because there won't likely be a move with more than 5 candles in a row in the same direction.

 

[meanreversion]

On the issue of correlation - it's an interesting point, either correlation between daily returns of S&P and your system, or in fact correlation of daily returns between systems themselves.

It's a common misconception that if two systems are perfectly negatively correlated, they will cancel each other out.

Correlation is the deviation from the average return.. thus if two systems both have positive expectancy, and 100% negative correlation, you have a money press.

E.g. system A makes 2 on day 1, 0 on day 2, 2 on day 3, 0 on day 4 (i.e. +2,0,+2,0 ...)

System B makes 0 then 2 then 0 then 2, i.e. 0,+2,0,+2 ...

Both systems have daily expectancy of +1, but the correlation is -100%.

If you add the two together, you get +2,+2,+2,+2 which is of course perfection.

 

[Yamato]

Thank you for the information. Yes, correlation is something I keep looking at. As I said, I use excel's correl function and I plot it on that chart. I am not that much of a scientist but I try to look at all ingredients, while keeping things simple, because I am not very gifted at formulas.

 

[poobah]

Re: correlation between my systems and sp500 and eur/usd

I am on my island and you're off it, and all around me there's space, silence, loneliness.

It's a beautiful island and I see no correlation between your presence on the island and the performance of your system. Congratulations!

 

[Yamato]

Yes, thank you. I created that island, too, among the other things.

 

[Yamato]

After pondering for almost 10 years, I have decided that this song wasn't that bad after all:

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

You know, this song seems so simple and yet if you try to sing it, you can't sing it, or it doesn't sound as nice. It's all in the arranging or whatever you call it. The melody is not that good but the rest of the work was excellent, that's what i am trying to convey.

The chick is ugly, the melody sucks, the lyrics are meaningless, but the song kicks ass. Why? There was some makeup perfectionist at work there, the video, the musical arrangement particularly. These guys managed to make the whole thing sound and look appealing, pleasant and profitable.

There's great art here, but it's not in the raw materials: it's in the details.

If I had to make a comparison to my systems, they're the opposite. All about substance and no makeup whatsoever. The infrastructure and the makeup is zero. They're entirely about the melody. Among the over 100 melodies (systems) I created, about 20 of them are worth listening to (trading) without the need for any makeup (advanced infrastructure, details, advanced programming skills). I mean the good systems I have are all about the melody and nothing about the dressing and the makeup. I suppose a lot could be improved if we focused on the details, but for me, with my skills, it's more effective to focus on creating more melodies, rather than on the details. So i'll keep writing and playing my systems-melodies on my one excel workbook, which is under two megabytes.

 

[Yamato]

sleepless as usual

Took xanax, took melatonin, drank some wine. Still sleepless.

The countdown continues, the countdown to the island.

Equity is now at 11,500.

Just another 500 and I'll start getting paid.

Now I am probably going to start making money from discretionary trading, too. We'll know by the end of this week.

If that goes through, then I'll be able to do my own automated trading as well.

Then I will have 4 sources of income, my dream come true: 1) job 2) investors automated 3) discretionary 4) my account automated.

As time will go by, the income from 2) 3) and 4) will keep on increasing. But i still don't know at which point i will be able to entirely let go of 1), which is the most reliable source of income. Pretty soon 2, 3, 4 will surpass 1 in terms of income, but they'll still not be as reliable, ever. To really be wise, I'd need a million set aside before quitting my job.

But in theory i could increase the part-time schedule all the way to just 4 hours per day. I don't know if they'll let me. That would be an ideal way to hang on to my job, without being too tired from it.

All this sucks. Why didn't i know what i know now when i was 20? Hell no, from 14 till 18 I was being taught ancient greek. That was supposedly more important for my future. My parents are intellectuals, and catholic ones, so you have to give your life for the world and you're not supposed to be happy in life - that's a sin.

 

[Yamato]

"new" discreationary rules



9 ma to detect overbought/oversold levels and old candlestick rules to not anticipate and not stay in denial if you are already in a trade

the rest is all up to me

9 ma is just an indicator but not a requirement - the only requirement is waiting for the candles to close and enter and exit upon reversal - color of closed candles signal it: i get in LONG at the first non-red closed candle and get out at the first non-green closed candle, and viceversa.

Oh, and, i was forgetting this... there's also the usual old golden rule, which I've mentioned for years by now, but which i rarely follow: one trade per day. When i followed it, i made money. After 10 trades with that rule and making money, i got tired and switched back into junkie mode. Lost money ever since. Would we rather be bored and make 300 per week? Or lose 2000 per month and get to have fun and make lots of trades every day?

Tell you what!

Let's even abolish the almost useless rule of the 9 ma. From now on the only rule is to have a fixed rule for entries and exits, which is the 1-minute closed candle's color, mentioned above. And the other rule, until I get good at this is to only make 1 trade per day.

And now enough of today's trading disasters and this unhealthy lifestyle. For today I am done. Now xanax, melatonin, alcohol, and some healthy tv, Big Brother.

Oh, I know I know I know I know I know I know
About your kind
And so and so and so and so and so and so
I'll have to play

I am still trying to figure out how many times she sings "i know" and "and so". She sings it so fast that I lose count. I presume six, like the lyrics say. But I would also bet something on her singing it 7 times.

But the best thing of this video and song is the guy saying "four, three, two, one" and the camera zooming in on his face as he says it.

 

[Yamato]

still thinking it over: discretionary challenge

It's a matter of pride. I have to succeed.

One trade per day.

You can go LONG after any closed non-red candle (even after a series of green candles). You have to exit after a non-green candle just closed - or sooner if you want.

Viceversa for SHORT.

Everything else is discretionary.

This way I have devised a univocal entry/exit method, which will protect me from anticipating a move, going against the trend, and will protect me from huge losses and blowing out my account.

 

[Yamato]

latest discretionary rules

The only requirement is you can't have the latest closed candle against you.

Say you want to go long: the latest closed 1-minute candle can't be red.

Say you are long: you have to exit as soon as you get a closed 1-minute red candle (or sooner if you want).

Viceversa for short.

Everything else is discretionary and trading is halted for the day as soon as, once a trade is closed, you have an overall negative balance for the day.

This thing of allowing more trades if the finished trades show an overall positive balance is meant as an incentive to myself to make good trades - if i pick my trades well, i am allowed to keep on trading all day, otherwise not. This way, the most i can lose in one day is 100 dollars or so, but i have unlimited opportunities in terms of profit.

 

[Yamato]

things are going great

Dude. No comment. Check it out:



Actually, comment.

We're 30 dollars above 12k. According to our own schedule, I will be getting paid on high water mark profit on a monthly basis, so it looks like this month I will be getting my first serious and stable salary ever since I started trading in 1997.

Instead of just blowing out my account endlessly, like I've been doing in the past 13 years or so, despite the promising systems I've been having for years, now I will finally start getting paid real money, month after month, because on the investors' account there's no way for me to blow out accounts, tampering or anything like that.

I am precisely entitled to being paid one sixth of the profit made in excess of 12k.

30 dollars divided by 6 is only 5 dollars. Yes, it's laughable, but listen to this.

We make about 4000 per month. And, as we scale up, this will keep on increasing.

So this means, starting this very month, I will be paid from 500 to 1000.

My salary is less than 2000 dollars per month at the bank. This means that pretty soon I will make more from this than from my bank. On top of this, the other sixth of profit, which is still my share, will be used to build up the cushion, so, if we ever stop doing this, I still get all that part of the cushion, according to a precisely defined schedule and formulas.

But let's not dream too much. Let's say that 500 dollars a month right now is almost guaranteed, and it's not peanuts. I mean, yes, it's almost peanuts, but it's better than losing money, which is what I've been doing all by myself. And it's not peanuts.

And that equity curve above, that doesn't even surprise me. I knew my systems were capable of this. I've been expectedly that kind of curve ever since we started trading. The reason it didn't happen earlier is that I chose systems light-heartedly and picked the wrong ones.

 

[Adamus]

Congratulations on the system results. Long may it last.

re your discretionary system, I wrote a system a bit like that once. There was an indicator which measured the average length of consecutive bars in the same direction, and when it went above 2, it went in on the next bar in the same direction. It didn't work just like that though and I didn't experiment much more with it because I couldn't understand why it wasn't profitable.

 

[Yamato]

Congratulations on the system results. Long may it last.

re your discretionary system, I wrote a system a bit like that once. There was an indicator which measured the average length of consecutive bars in the same direction, and when it went above 2, it went in on the next bar in the same direction. It didn't work just like that though and I didn't experiment much more with it because I couldn't understand why it wasn't profitable.

Thank you.

Regarding the discretionary method, it is definitely discretionary and cannot be automated. The potential signals being produced are hundreds per day. It's all up to me to pick the right ones. I do not expect to automate it at the moment, because I still don't know exactly what I am thinking as I pick the right trades - I don't even know if I am capable of consistently picking a majority of good trades. Automating it would be too good to be true, because an automated system that makes money every day would be the ultimate dream.

However, I think I may be able to make it happen at least on a discretionary level. Today there were some more changes, very important ones. The rule of the closed 1-minute candlestick stays, but now I added this requirement: i have to be paper trading at the same time as I am trading live. This way I keep myself busy with paper trading, and I only make the trades I really believe in (the rest of the trades I make on the paper trading account). Today it worked and I made 200 dollars. After 1 hour i was quite tired so i stopped. I have made 20 trades on the paper trading account, where I broke even, and 2 or 3 trades on the real account.

Here's the latest rules:

1) for both entries and exits you can't have the latest closed 1-minute candle against you: no color is ok and you can exit earlier.

2) halt trading if you show a negative balance for the day

3) paper trade non-stop on laptop and go place the trade on pc when you spot a great opportunity