TA self fullfilled prophecies... what a joke.

Does it matter? Regardless of who swears by TA and who resents it, this game comes down to one single factor...profit. As for my opinion, there is no analysis which can work generically for the lifetime of a trader; but a true understanding of the numbers/charts/news which the trader employs. Any method of analysis can be successful, but only one level of understanding can prevail.

I find it bizarre that one can read the charts or the news and gleen something out of it. Maybe NFP if played right may be the only exception or if you have deep pockets and have the swami guru next to you. ie...."Any method of analysis can be successful, but only one level of understanding can prevail."
What the heck does that mean? The only level I understand is to buy when your getting crushed.

Congratulations on starting your 160 th thread.

I would like to know if you opened an account and started trading your own money yet ( not clients ) like you said you were going to do about 4 weeks ago when ATR returned to normal?

If you believe that the markets will return to "traditional" ranges, then you must believe there is order, therefore not random.

Just to add one more thing to this thread....

I think Tdions problem (and many others) is that the markets do not work acording to rules, math, I assume this is the way that his mind works, many people who program are similar. (as Im sure he knows) Markets are purely moved by their participents psychology, this is why I beleive he finds problems.

I started a journal afew weeks ago. I had been spending alot of time thinking after the thread I started('Dead end') and got an idea for a trading method. Since then I have increased a Demo account balance by nearly 40%. Yesterday I received an Email asking what turned things around for me(I have yet to reply), But I think the answer is that....... I stoped trading market swings and started looking at market levels instead. I cant be sure, but im hopefull my success will continue.

At the end of the day, where theres a will..... theres a way.......
And Tdion is still here.

I understand you completely. Unfortunately, you could have saved yourself a lot of time by not writing any statistical inference code until you actually discover what it is that you are trying to teach the machine learners to infer. There is an infinite number of ways (literally) to wire these things, and depending on the "wiring", they are capable of predicting different functions. Anybody with a programming background can write one.. but what are you trying to predict? It is actually very difficult to define, and it's a hell of a lot more than "down after I sell and up after I buy". Timeseries prediction is an unsolved case in the field of data mining.

If you are saying that nobody knows where the precise tops and bottoms of the market are, you are absolutely correct. Nobody who is allowed to speculate fairly can do that. The tops and bottoms are random - like everything else - but more importantly they are random with such a wide variety of values that predictions are not profitable with them given the current market history. Like you said, no two days are the same. No two hours are the same. If the earth lived for a hundred thousand years and we had market data for every hour of that, then maybe the models built would fair slightly better at say, the H4 level of local minima and maxima, but still wouldn't be useful. Tops and bottoms are only 2 of the random things that are difficult to predict. I have used them here as en example because people think about making money in the markets in terms of them.

But there are other random things that have a lesser range of possibilities that are more practical to play with. Simply because they are also random does not mean they are the same task as, say, predicting the price at one tick given the previous 2 ticks. Or the change in price one day given the past three days. That's all I am trying to say here - not all random things are created equal. It's such an obvious statement that I almost wish not to say it, and all the market speculators derive their hope from it, and all the consistent winners (some billionaires) owe their wealth to it, and you as someone who has bet your money already know it. Forex factory arguably would not exist if there was nothing useful that anyone had to say about the market.

So what are these "higher probability statements"? I wish I could write them down, but they are learned almost subconsciously by the wonderful human brain, which supresses things and keeps others. That's why even people with advanced algorithms usually want to be in front of the screen before placing an order. It's just hard to remember years of action.. you simply become accustomed to things. You learn to see the events/shapes that are uncommon, and take bites of the movements that counter them. Or bites out of the common ones, as recognized by other shapes (cue 'price action' thread links). The smaller the bite the larger the chance of success, but the lower reward per success - it's the well known risk to reward thing. This is a completely different challenge from trying to model price itself as a function of <insert known list of factors>.

I'm sure you already know most of this - how could you trade otherwise? And with third party money, no less? I use to want to do that, by the way, but never now. In fact now that I know how hard and time consuming this whole thing is, I am under constant pressure to use everything else I have to amass wealth rather than play this game. I encourage everyone not to trade their money. I am not entitled to tell anyone how to trade, and as you can see the above is not trade advice, as in all my posts, it is merely a discussin about numbers. If I ever make it to a few million dollars out of SPOT forex, I will still not be motivated to say anything about trading. I'll leave it to the folks making billions to do that. I hope this has made any sense.

Good luck, dude. And good god, I should've been working on my thesis! F*@^!

I have a question for you computer gurus...Is it possible to program a diagonal line that when hit by price would open and or close an order?

Thanks a bundle,

Scotty B

WOW - groundbreaking discovery here:

Fundamentals and Technicals such as Patterns/Indicators/Fibs/Pivots work only some of the time.

Darn, now if only we can get them to work correctly 100% of the time then we wouldn't have to worrying about minor details like trade management.

Scotty: Yes, it's possible.

If you don't like my threads, don't visit them.

I am sure it already exists...I just need to find it.

Tesla's trendline alert indicator....

An indi can't open/close trades... as far as I know, at least. But you could use the same concept w/in an EA to do the same thing.

U cud probably argue for eternity that the markets are either random or there not. Some times TA work som times it dont.
Wud u say that the weather is random? U cud say no because there are many factors that can be calculated to predict what the weather will be like. Some times tho the weather people get it wrong, not to often tho. They get it right most the time. But because they get it wrong som times is it random?
The markets just have millions of different factors, which are the traders. Can u predict this like the weather?
Micro minime said somthing about dimensions, got me thinkin about what are the dimensions of the market?
You got time, price and volume. Can these be used to predict the market? Who knows, u just gotta do whatever it is that works.

Forecasting a hurricane involves feeding in streams of data such as precipitation, temperature, air streams, and geographic landscapes. Those are physical dependencies that have nothing to do with beliefs... they are a hard science.

Now prediction of price would need similar inputs that offer the same dependency.... Perhaps one could argue interest rates and economic reports as the driving factors of price change. But when you factor in fibonacci retracements and TA, you are implying that the psychology of people are dependent on those chart metrics.

What if everybody is trading TA on a different timeframe? They wouldn't match, and you'd get lots of random noise. And factor in that the majority of forex volume is driven by large corporations converting paychecks for employees, you'd say those operations have nothing to do with charting at all.

Finally, what you are really trying to do is predict what a few central bankers are thinking at any moment in time, as they are the handful of people that actually make any difference.

From what I've read from multiple credible sources, the vast majority of FX transactions are speculative in nature. I wish I had the sources handy, but alas, it was awhile back.

Also, this thread is a bit redundant, wouldn't you agree? This has to be at least the third time we're having the same exact discussion (let alone on ET where there's a new thread each week on this topic). Just curious, do your clients read these posts? How did your EAs perform in '08 (live trading, not backtests)?

Howdy KGB.

You're right, it is redundant. I suppose I find it to be the most relevant topic in trading.... it shows whether we actually know what we are doing, or if we are simply trading in the dark.

Now I am not here to make enemies, and think its about time for me to stop posting and let this thread fall out of circulation. As for your questions... My outside investors don't really care about my opinions one way or the other. They simply know that it is an extreme risk, extreme return opportunity that would have made lots of money in previous years. 2008 did very well until Lehman brothers went under.... I have tweaked the method to avoid high volatility times, and hope that 2009 will make a mint.

Off topic, I'd like to let you know that you are one of the few people on FF I actually pay attention to.... your posts are usually at the center of the bullseye. Best.

No reason to. These threads usually attract some interesting comments and thought provoking points/counter-points. At least here at FF, the level of discourse stays relatively courteous and intelligent (unlike a different trading forum I occasionally frequent).

Indeed, my post-Lehman strategy was FX aversion, switched focus to ES/NQ futures and certain ETFs, didn't have much choice as the liquidity on several ECNs really dried up. Anyone who stuck around for those concurrent +/- 10% daily moves in the yen crosses is worthy of my respect.

Thanks, same to you, keep up the great posts. Good luck in your trading in '09!

As relates to the weather forecasting analogy:

Ask 10 forecasters to predict a week from now and you will easily get 5 different if not 10 different forecasts. No matter the "hard scientific" evidence they are just as variable as a simple moving average.

As for forecasting the weather,... Is it predictable?? Sometimes.
If you want know the exact weather forecast for Saturday, 8 months from now. No, not possible.
But we don't need to be able to "read the minds" of the top Central Bankers either. That is also not possible.
At what price will Euro close at the end of the month?? Dunno.

So, yes it's random, but it's not all random. Sometimes we CAN predict the weather tomorrow with such certainty that we are willing to bet a lot of money that we're right. And when the current conditions allow reasonable certainty, we win 80% or 90% of the time. You just have to be patient when the barometer is mixed or changing. And if you see potential rain clouds forming when you have predicted sunshine, it's probably best to take some money off the table and save it for when the barometer is reading a bit higher.

You should dive deeper into the rabbithole.
The only reality is that most people lose out because they do not master their emotions.They start to float because they made a few good fills and are too focussed on materialism.When TA fails , and you control your emotions, all you do is you cut the loss.When TA works, you let the profits run.You can check the threads, they might give you perspective.. among validation seekers, exaggerated cliches and bitter cynicisms, buried gems of insight..