Well that's your opinion. Question: Does the 83% correlation figure still currently stand when comparing one days uptick or downtick with following day's action? Gil's summary is "the markets ARE mostly random" "trading is more of an art than what most people believe" No big disagreement here. I am currently "beating" the market so that would support the non-random camp.
taleb convinced me of this when he stated very little is random. random is a word we use to describe outcomes whos cause is unknown or not understood.
Frankly, another poster suggested that using the word "random" is misleading. How could it truly be random when some of us can beat is routinely and systematically. My model makes 99.99% winning trades with very little draw down. How could that be possible if it were truly random? I like the term unpredictable better. It's because you can NEVER reliably predict from one second or hour to the next what it will do next or when it will do it--up/down/sideways. However, you can predict it's randomness with amazing accuracy. In other words, you always know that the current direction will never last forever. (That may not apply to stocks which could actually go bankrupt and drop to zero value.) Eventually the price will get so cheap or so expensive that other buyers or sellers will step in and mop up. That's how J. P. Getty become a billionaire. He used his oil profits as seed money to mop up stocks of companies with hard assets during the great stock market crash. So you never worry about picking bottoms or tops, just knowing that a top or bottom will come is guarantee enough to profit on all the intermediate rallies/selloffs. Wayne
I've gotta question for you about your 99% model: 1. Has this gone LIVE? 2. Have you used Limit Orders for the tests? 2a. If so, how did you determine whether order will get filled or not? 3. Who did you purchase the Forex Data from? 3a. Are you using Bid/Ask data or Tick TOS data? 3b. Is your broker a Market Making-style broker (Charges 0 pip commish but has spreads) or ECN-type broker (Charges commish but players determine the spreads) 4. You're making about 3-4 trades a day, what percent of the time are you in the market? Let's start there... I'll just say that what you provided in acrary's thread is very skeptical... and I have a few quotes you provided that set my alarm up for a possible testing fallacies....
I addressed all that already. Thanks. Going live soon with this particular model. I have auto traded others with mixed results for all the reasons you mentioned. I feel one thing that put me on the right track was that experience with failed auto trading. It taught me to examine EVERY SINGLE trade in a back test to ask myself, could I stand that much drawdown, MAE or loss and keep the system turned on and running. People make the frequent mistake of seeing a profitable system but they can't stomach the draw down periods. So it's critical to work those out beforehand and look at each trade as if it were live. That's impossible to comprehend until you really have done a live auto strategy and/or discretionary traded. Well, I been there and done that. Sincerely, Wayne
You and everyone else enters a trade on the expectation, anticipation or prediction that you will take it out profitably. Professional winners in trading do so all the time - reliably. It is a wonder how all the fools (coming on to ET) get together with some money in the first place.
Thanks Champion. People disbelieve. It seems to fail all reason that you can go short in the middle of a big rally and still make money even if the market continues going against you. It's all in how and when you exit that counts. I might even try some other strategies that I threw away in the past now that I know how to always exit profitably. Why not even use a dumb trend strategy? Then you make money on the whip saws and big trend moves also? It doesn't seem to matter any more how you enter. Right now, I just use a very dumb entry rule that would fail miserably except that it exits properly. Sincerely, Wayne
Mandelbrot, Very interesting point. That's why "random" doesn't really describe it. Still, when you say it isn't random, that implies it's predictable--that you can predict the next price much like driving down a winding road where you can see which way to turn next. But the markets offer zero road ahead, I'm sure you agree, are completely unpredictable as to when and how much they will wind and turn. However, amazingly it's completely simple, as you point out, to profit very consistently in spite of that. I feel like you do now that the randomness was only in my imagination--only in my mind--because I see price action so clearly now. But we can still never predict the next price or even the direction. Instead, we can only say definitively what trade to make so it profits regardless which direction the market goes next. Amazing, amazing, amazing. Maybe a getter analogy would be trying move through a room with obstacles while completely blindfolded. So in the markets, we can't "see" ahead of us but we quickly adapt as we bump into obstacles. Sincerely, Wayne
Yes, I like that analogy of walking while blindfolded. It's because to a blindfolded person who never saw the room before, it would appear that the path to the opposite side was somewhat random, unpredictable. But in reality if you could take off the blind fold you would see that the room was laid out in an orderly fashion. It's very hard to figure out the markets because you can never, ever remove the blindfold. The future remains as much a mystery today as it has been since the beginning of time. Wayne
Actually, maybe I'm mistaken that we can take off the blind fold so to speak by looking at charts of the past. However, it's like moving through a house blindfolded with many different rooms and you're only allowed to remove the blindfold and "see" the rooms after you leave them. If you then build a plan to make it through the next room based on the contents and obstacles of the previous room, you'll quickly fall and bang your head because the furniture and items are laid out differently in each room. Instead you need a system that helps you simply adapt to obstacles as you encounter them rather than attempt to predict their size and location. Wayne