The ACD Method

Quote from Maverick74:

Thanks cd. Yeah that spike we had a few weeks ago showed up on there and subsequently came in. I would probably be more inclined to look at the OTM call skew to see how steep it's getting. Those options have a small weight in the OVX calculation but they are very revealing about the "magnitude" of the expected move. You can look at those skews on TOS if you have it.
More...

yeah.. i do have it.... and i have looked at at skew on there before.. good idea.....
 
Quote from Maverick74:

OK, so a couple of things here. My 30 day confirmed on 8/27 at 18.00. I have no idea how you are doing your number line and that is not to say you are doing anything wrong. But I've modified my number lines quite a bit from Fisher. Again, not to say there is anything wrong with how he does it either. I've just put a lot of time in the process and I feel I have my process down pretty good. Now, having said that, let me address your question:

A 30 day confirm is NOT a buy signal. The 30 day number line is a reflection of internal price action. It's telling you there is a "potential" for a sizeable move due to the way the product is acting. I have created number line derivatives that help in that process. But there are many ways you can play a number line confirm. Yes, you could simply get long on the day of the confirm. I personally use monthly A downs as stops. Since your confirm happened above the monthly A up, I would probably take a smaller position to account for the wider stop. And I would look to add to my position on failed weekly A downs at better prices "provided" that the number line was still strong.

Another way to play it is once you get the confirm wait for a failed weekly A down to get long. This would give you a tighter stop. You could wait for a re-test of the monthly A up which it did do at 18.00. Remember I mentioned earlier that more then half the time you get a re-test of an A level.

There are countless ways to do this. You have to be the ball coach and make the call. Read the defense, call in the play, protect your quarterback. Just like trading. And another thing regarding number lines, markets rarely take off right after the confirm. I got lucky that X did that but that is not the norm. I'm getting in for the "potential". Not because it's going to take off to the moon. If you want to play pure momentum I would focus primarily on week one monthly A up confirms. Those have much higher octane. Those have their drawbacks as well so everything is a trade off.
More...

Thanks Mav, food for thought as always.

The early confirm was based on Fisher's textbook scoring system. Don't use it any more but still keep the table I made when I was learning. :)

Yes, I do remember your saying that you'd modified the number lines to suit your needs.
 
Quote from justrading:

Thanks Mav, food for thought as always.

The early confirm was based on Fisher's textbook scoring system. Don't use it any more but still keep the table I made when I was learning. :)

Yes, I do remember your saying that you'd modified the number lines to suit your needs.
More...

Let me add one more thing here. I've said this many times on here but it's important enough to keep re-mentioning. The ACD is a large eco-system. In this eco-system everything should align. Most technical systems don't have this and that is where many break down. You hear traders say this all the time, this indicator is telling me to buy but that indicator is telling me to sell....confusion sets in or a trader falls back on his bias (perma bull, perma bear, fader, etc) to "select" the indicator that "agrees" with his bias. BAD! Don't do this.

My rule is that an ACD signal is NOT valid IF it CONTRADICTS a current live signal. A simple example: XYZ confirms on the 30 day number line but it's below the monthly A down and still within the first 2 weeks of the month....invalid trade. So once you get that down, you are able to discern the complete picture much better. Again the problem is if you pick and choose "what" you want to use you will always revert to your bias which will almost always be wrong. You've seen me talk about this on oil or bonds or gold or something where I say Bonds are weak on the number line but made a weekly A up. Or AAPL looks good here but it's still below the QTR A down.

Let me briefly explain the importance of this concept. It is actually IMPERATIVE that you BELIEVE in what you are doing. I cannot stress this enough. In the military it's called the chain of command. They teach 18 year old's to follow orders to the T without exception and to obey their commanding officers. They do not want them to think. They want their to be an order in the process of how decisions are made. This is so when life and death is on the line and a soldier is given a command they they do NOT get emotional and make discretionary decisions. And it's important that since they are following orders without questions, they have to BELIEVE absolutely in their officers and trust they they are giving them the right orders or the entire chain of command breaks down.

What happens with traders is they don't really "believe" in their system. Usually this comes from taking signals that contradict their system and they work. So if they work, they begin to ask themselves does their system even have any value at all? All the ACD metrics have value. They are TELLING you something. You need to LISTEN to them. It does not mean you take every signal it generates, it simply means you do not fade it. If you start doing the opposite then that opens the door to you being the "discretionary guy who always knows best. This leads to a breakdown in discipline. Where you here guys say I know I should stop my self out here but I think it's a head fake, I'm doubling down. See they don't really "believe" their signals, they are constantly full of doubt and hence the emotional anxiety and bad discipline.

OK enough of that. So regardless of how you use your signals, what I'm trying to say is that they should be in agreement. If the number line is weak but something is making a monthly A up, how is that happening? If you want to get long ABC but it confirmed a weekly A down, wait. The weekly A down doesn't get pushed to the side just because you really really want to buy something. Stay within the ecosystem and look at the complete picture. The more you practice it, the better you will get. Almost every trader I know who has failed has done so because they break their rules, they don't believe in their system and they fall back on emotional decision making. One of the clues you'll hear is when guys talk about stop running, the Fed, the PPT, insiders, Goldman, market makers, HFT's, Algos, etc, the list goes on. The fact of the matter is, you have a collection of data. That data is not right or wrong, it has no opinion, it simply is stating something like the current temp in NY. It is YOUR job as a trader to take the raw meaningless data and organize it and create information out of it. Then with that information to seek value. And then to act on that value without prejudice or hesitation. All other variables are meaningless. Once you blame them, you have stepped outside your ecosystem and handed control to outside forces. You will be hard pressed to ever make money if you do that.

Anyway, that was a long explanation to a question you didn't ask. :)
 
Quote from Maverick74:



Anyway, that was a long explanation to a question you didn't ask. :)
More...

And a very welcome explanation at that, but let me ask you a follow up question.

When you tweaked your number line system, did you do anything to improve timeliness of the signal/confirmation? Is that even possible?

I don't mind putting in whatever effort necessary to tweak my system if it is doable, but I don't want to set off on a wild goose chase. Curve fitting something to a few symbols or instruments and having the Law of Unintended Consequences hitting me on the rest is something I'd rather avoid.

As my little trial with the 30 minute OR on X showed, making qualification for A up and down more difficult cuts both ways. Lose a few points on A ups, but also lose out the marginal A downs and the nett result was an even earlier confirmation.
 
Mav, Just to extend justrading's interesting follow up question.

Though I realise that your 'tweaking' did improve the number line system. Also I dont think ACD is like a technical indicator where you end up 'curve fitting'. It seems to me it is 'logical' and about 'price', so its more about getting the 'logic' right, than just trying to find what seems to work best on average.

As justrading shows though, you have to do the right tweaks and not the wrong ones!

Did you feel your tweaks in general ended up extending the time before you would enter?

Could you give a heads up on the main issues you felt were perhaps improvable over Fish's method? Even if not the solutions, but what you felt was not quite right.
 
Quote from Sedulous:

Mav, Just to extend justrading's interesting follow up question.

Though I realise that your 'tweaking' did improve the number line system. Also I dont think ACD is like a technical indicator where you end up 'curve fitting'. It seems to me it is 'logical' and about 'price', so its more about getting the 'logic' right, than just trying to find what seems to work best on average.

As justrading shows though, you have to do the right tweaks and not the wrong ones!

Did you feel your tweaks in general ended up extending the time before you would enter?

Could you give a heads up on the main issues you felt were perhaps improvable over Fish's method? Even if not the solutions, but what you felt was not quite right.
More...

I really can't go into details as I have too much intellectual property invested in it. I will say that anyone has has traded long enough will understand the exogenous variables (inputs) needed to go into the number line to increase it's accuracy. The output (effect) results are earlier signals are generated. I should note that not only did I change the "scoring" of the daily values but
I added a 2nd derivative to the number line itself.

As I've mentioned many times, with enough screen time you will understand the limitations of what's given and how to optimize those limitations. Some points I can add here are the importance of collecting data. Every one of you guys should be storing data in every imaginable form. Even if you are not going to use it now, perhaps in the future you will want to reference it. It makes data analysis 100 times easier when you have the raw data saved and you only have to go back to manipulate it vs going back in time and going through the laborious process of getting it day by day and product by product.

I put that part in bold to truly emphasize the importance of this. For those of you concerned about labor costs (your investment in time), you need to make a point to maximize your efficiency in your research. This means become more efficient at collecting data, storing it, accessing it, and try to create functions that reference previous functions to make calculations easier. In simple parlance, don't use a hammer to cut down a tree.

I know to many of you reading my posts you will be discouraged by the amount of work I have laid out before you here. Trust me when I say this, there is NOBODY that is going to make consistent long term profits in the market WITHOUT doing this kind of work. And anyone who tells you different is probably trying to sell you something.
 
You must log in or register to reply here.
Back
Top