Thanks for the chart. I refer to this as CCC standing for congestion , convergence and centering. Lets say a person wanted to back test this. congestion would be defined as a lateral flat channel. Convergence would be tested to show that the downward dfirft in volume would be influencing one or more sides of the former congestion. A signal on the first test would initiate the second test. Centering is a noise dominated VDU backtest period. So all the above is a way to get a failure to show statistical significance in backtesting. What is missoing so far? Well several things. Make a list and add them to the backtesting scenario. The idea of back testing is to get a signal that lets you make money. you ask in the back test for a specific signal. what is the signal for BO? Look at sunnyskies comments. Do you see where he operates? For some reason he believes that doing the equities trading and the commodities trading have the same trading basis. They are as different as can be it turns out. Equities are position straded by using entries and exits. Both "anticipatory" where the fulll range of the half cycle is used to extract at least 50% of the run. Commodities use a strategy of hold and reverse, being in the market all the time. What are the signals for "entry" in back testing? What are the signals for exit in back testing? What are the signals for holding in commodities back testing? what are the signals for reversing in commodities backtesting? Of course they are all leading signals of price. Do you think most people do back testing using price instead of the leading indicator of price? It looks like they may. What does lead price for anticipating a signal that will let you make money? Why does price BO? What changes that prevents price from not breaking out? Di you think anyone is designing backtesting software to deal with anything remotely related to making money by using signals that lead price? What does a person who can backtest well and get the picture write? How does that person address what he looks for as a signal to then judge if a backtest indicates that any money is being made? Do you know what JohnnyK is saying? Do you know how to measure what a person did when he saw the combination that he needed to take and action? Monitor, analysis, decide, act. Repeat at the market dictated pace and volatility. no one ever even gets a glimmer of what backtesting is. It certainly is not an oversimplification of not using the pertinent data in the first place. Making money has nothing to do with looking for anomolies. Anomolies do not create a decision to either hold nor reverse. what creates any decision is quite clear: you have to have the knowledge, skills and experience to analyze what you have monitored which is that which is important to monitor. As the market has alternative after alternative cut off as a possibility, you have to know that the alternatives are being cut off and then know which one remains and what gates (triggers) that opportunity that remains. What is it that not longer prevents a BO in price? when your backtest shows this particualr signal, then you have the first half on the money making back tested... Then you can go to work on the important second part..... LOL.
Mad, This is a question. This is the question. I believe this was your original question, slightly rephrased. Do you know what I'm saying? Is this the answer you were looking for? Original question #2: "Also, in order to enter a position what other things are essential to be looked before entering the position on BOs." JohnnyK
Wolfie Since you seem to understand this little nuance, the difference between direction and continuation, would you please explain it to us? Icarus also mentions this but some of us intellecually challenged still dont get it.
So you know the answers? I guess teaching with questions is one approach but it can be frustrating as hell for us students.
Johnny That is correct. I wanted to know what people in ET think about BO forecast. Jack literally said : go figure out and I do respect that . the other people whether did not say anything or they did not believe in trading BOs. Like I said , I am a newbie and have not traded for long, But I look at BOs as a profitable trade under below condition : 1- need to see a considerable consolidation time ( ie: on ES > 0.5 hr ) 2- I do not like any major resistant line parallel or next to my BO direction 3- BO should have a relatively high volume 4- Usually wait for the end of 1st 1 min tick before I enter. Because I think if it pulled up / down a lot in 1 min , it has a potential failure Beside jack advices, I did not any hear thing else, from anyone whether or not I am an idiot or they seem to be accurate.
Mad, I'm curious. Have you been testing this setup? I was a student of marketing, and learned that no matter how good a headline we thought we came up with, we must test, and then test, and then test again...and then look at the data. Change a word in the headline and then test again. Before I step in, I'm going to give JH some more time to clarify the *leading* signals which might augment this list, or replace them. So far, we've been told at length, and with many questions, that a leading indicator, can't be put in a pipe and smoked. Perhaps. Certainly, so far, such a leading indicator, and how to analyze it, cannot be revealed. Nor, I expect, the *cutoff* points, which eliminate other alternatives...leaving only one way for the market to go. JohnnyK
Johnny That is absolutely true that everything needs to be tested, However to me testing is the next step. First I need to make sure that I am not wasting my time if there is some thing else I need to consider in my factors. ( for example : some people think if the BO takes the RSI or MFI to the red zone, it will not continue or etc⦠) Again that is why I initially ask the question, regarding BO definition / detection.
Wolfie. My comment was in the context of Jack's earlier postings, where I recall that he recommended a bracket entry and "washing" on a reversal. That was what I tested. Mike.
No other B-Teamer has responded to Jack's post, so I'll hazard one. Converting what he said about CCC to a word algorithm would produce rules something like this: 1. Test for a "congestion", "lateral flat channel": 1.a. channel test time duration 1.b. absolute value of channel slope less than a threshold value 1.c. channel width less than a threshold 2. Test for "convergence" 2.a. convergence test time duration 2.b. channel width change negative 2.c. channel width change less than a threshold 2.d. absolute value of slope of extrema of one channel edge greater than slope of extrema of the other channel edge by some threshold 3. Test for "centering" 3.a centering test time duration 3.b. absolute value of channel slope less than a threshold 3.c. channel width less than a threshold 3.d. average bar volume less than a threshold This is codable, but untestable with any significance because it has eleven "strong" rules. And in sketching it out I probably forgot some.