Hi Mav, I could really use your insight. Let me first describe what I am working on. I'm building a trading system that can predict intraday minimum range with 70% accuracy. The system does not pick direction. It predicts with 70% accuracy a range that a stock will travel through up OR down. If I were to tell you that the SPY will move $.01 up or down from the open you would hardly call that a prediction. The minimum range movement that the system can predict is more significant than that. Picture if you had a few years of backtested data to back up an A Up or A Down signal. How would you trade it? I'm not trying to build a red light green light system here. For example, if an experienced trader like you Mav knew that there is a 70% chance that the SPY will go up or down $.65 from the Open tomorrow then how would you trade based on that info? Would you try to start the day with a directional bias? Would you set minimum thresholds within that (Open+$.65) and (Open-$.65) range? Would you look at the entire market and say to yourself "hey the entire market is up strong today and the SPY is currently up $.40 off of the Open..I know there is a 70% chance that it will break through +$.65 so I'm going to get long right now". Thanks.
There is not much you can do with an ATR. ATR's are probably more then 70% accurate btw. But it still does not help. The only real value add that I have found with ATR's is when you hit them very early in the day. That's usually a good sign you are going to do twice the ATR in range that day. Depending on the product, you want to hit it maybe in the first hour to 90 minutes. Otherwise you are just dealing with random data. An ATR is really no different then a standard deviation. Talk to anyone who has tried to trade using bollinger bands. It looks all so easy to buy the bottom band and sell the top but there is a standard of error though that usually is about equal to what you are trying to make on the trade which essentially wipes out your profit.
If I may also ask a questions Mav. The thread is very long. And I know I will get lost in it. I am wondering if the ingredients of the system are readily available. Are we looking at indicators, trend lines, daily high low open and close with basic calculations? Or will I need level 2 quotes with a super computer, or will I need to program. Can it be back tested manually ? Any idea how consistent it is ? How is the learning of the system presented, Is there a single place with the needed information Can it be learnt relatively fast or do I have to go through the whole thread? Is it based on statistics and are there any records of this ? Naturally you don't have to answer any of my questions, I have not looked in to your system but I remember it from years back. In a sea of information and misinformation it would be appreciated, Thanks
I know the thread is long, just read it. There are no short cuts in this business. You can probably get 95% of it from the first 10% of the thread, say 200 pages. Also, ACD is not a "system", it's a methodology. There is a difference. If I can take the time to write the thread, you can take the time to read it. There are no cliff note versions of it. It's not red light/green light. It's not a grail. If it's important enough to you, you'll sit down and put in the work. If you are looking for a quick fix, move on to something else. There are plenty of guys on ET lazy as shit that would rather fade large moves instead of taking the time to learn how to trade properly. The good news is they have a short shelf life and usually move on and out pretty fast. Good luck with your journey.
I read the whole thread. First 200 pages focus is more on technical aspects, read up to about page 500 and you get a lot of discussion on applying ACD and number lines, which is really useful. After about page 550 it is mostly calls so it is easy to skim through.
mav, just to say thanks for the building block for my system. Goes like that: You have to find your own private space in market