You mean if the market is trending or braketing ? 
"How" isn't as important as "WHY"........ My original question remains....... What is the purpose of defining the long term condition of the market? 
1)a specific set of overaly days (3,5,10,15, etc) 2) the behave of the ref. points we could get a better picture of price behaviour, aka if the 3D s trending into his braket range[assuming is braketing], or there are some early sign of a potential breack (first toward the limit of his range then maybe to the next) etc etc. 
3 Attachment(s)
You're trying to give examples of "how" overlays are use.....my question is "why". What is the purpose of using overlays? #1 Is this overlay bracketing or trending? If it is..... were are the bracket limits? #2 Is this overlay bracketing or trending? If it is..... were are the bracket limits? #3 Is this overlay bracketing or trending? If it is..... were are the bracket limits? 
try again........see if you understand why #2 & #3 are not correct 
1 Attachment(s) 2 distribution 
1 Attachment(s) 1 dist but close outside the limite, warning for range expansion or new trend 
I forgot about the "Close Rule". The Close should be Lower than UL and Higher than LL in a bracketing market. As Stain also said. 
There is not enough info by looking at the chart to determine which one is true. Only by learning how to organize and read your "data" (references) can you differentiate whether the current overlay is "testing for new trend" or "testing for end of trend". But the proper description for #3 is "testing". "testing for new trend" or "testing for end of trend" requires the ability to read the "data" (analysis) 
1 Attachment(s) Hey its me again. Can someone please help me with this question. And please please dont tell me im missing the whole picture and that this is just the **** of the tree or what someone told me. The overlay quant analysis in here is just one way to trade, its not the only. Im not looking to copy Mzvega 100% like most people seem to do. I have my own way of trading and try to pick the raisin of the cake which i would recomend others to do too, you can never understand the strategi or spreadsheet as well as the person who made it. Just my 2 cents. Okey anyway. In this example we dont have any attemped direction but we can see that value is lower. What would you asume out of this if volume was higher lower or unchanged? clearly only 1 distribution (more or less) which significe balance but please tell me if you know this i would be glad for all inputs, thanks Dave. 
https://www.forexfactory.com/showthread.php?t=482744 
May I suggest you direct your MP related questions, to a MP thread.. 
Cheers. 
1 Attachment(s) 

My EOD program still "automatically" retrieves data from fxcm But I can manually tell it to retrieve from either forex.com and/or oanda. There is little or no significant difference in the end result. 


6 Attachment(s) Serial Correlation........ The lack of daytoday serial correlation in auction markets is often the starting point for random market theory. Studies as early as Labys and Granger (1970) confirm the daytoday randomness. Even in trending markets, the expectation of up price or down price for the next day is still about 5050. Even odds in nontrending markets is easy to understand; in trending ones it is not so obvious. A market that is trending up or down does have an enhanced probability in the trend direction; it is just pretty well masked by the random component that is always present. For example, a trend of one cent per day (on average) is virtually lost in the average daily trading range of, say, ten cents. By grouping, however, the random component can be averaged out and the underlying price movement can be seen. We will illustrate this point with a trend in soybeans. Comparing one day's data to the previous days data (or comparing one candlestick to the next candlestick, or comparing one price to the next price) following the market in this way, are all examples of "linear measures". Now if you follow the data, linearly, "day to day" in sequential/chronological order, the market appears to move randomly. There is no "day to day" evidence of a underlying trend. This is where the random theory conspiracist's see this as some type of evidence that the markets are a random walk. When you look at the data in a linear, chronological , sequential, series (time series) it only appears to be random What this proves is that even in a trend there is no "day to day" serial correlation. Even in trending markets, the expectation of up price or down price for the next day is still about 5050 Rather than "linearly" follow the data price to price. Lets start measuring the same exact data, using "non linear" measures used in auction market analysis. Here we will use a 2 day sample of data. Rather than following the data "day to day" in sequential order. The results look just as random as following the market day to day. There is still no evidence of the underlying downtrend Even with a 2 day sample of data. Let's look at a 3 day sample of data......... Now things start to get interesting..Same exact data.. still using "non linear" measures.Now you start seeing some "evidence" of the underlying downtrend. By using a 3 day sample of data you began to filter out the "day to day" random fluctuations ("random component") It also becomes more apparent with the 4 & 5 day sample of data. So what does this prove?.......... do the math Testing a Trend for Serial Correlation There is one long downward trend from March 6, 1991, through July 10, 1991. This run covered 88 trading days (127 calendar days) or just over four months. The total price drop is 121 cents or 1.375 cents per trading day. The daily tradingrange average over the 88 trading days is 8.17 cents, ranging between a high of 29.25 and a low of 3.25 cents. Table 15 has the prices and a list of higher/lower prices on the basis of 1 day, 2 days,. . . through 5 days. Since a downtrend was Preselected, we have inserted a bias to the downside. Therefore ties those cases where the compared prices were the samewill be awarded to the H, or higher count. That results in Table 16 One of the basic fundamentals in AMVT is an understanding that the markets are not linear. Using linear measures to describe & analyze a non linear market is like using a tape measure to measure how many gallons it takes to fill a bucket If you group your data, group prices over the correct sample of time (price over time) rather than look at the data "linearly", day to day through a series of candlesticks, you see the markets are in no way a random walk. "A market that is trending up or down does have an enhanced probability in the trend direction" To test that statement in a downtrend using the data starting with the 1 day sample of data. If you traded in the direction of the previous day's trend (following price) with the probability of trading with the underlying trend, you would have correctly traded in the direction of the underlying trend 52% of the time and you would have been wrong 48% of the time. Correctly trading in the direction of the underlying trend following price "day to day" is still pretty much a coin toss. Let's look at the 2 day sample.had you traded in the direction of the underlying trend following the 2 day samples of data rather than following price, you would have correctly traded in the direction of the underlying trend 62% of the time and you would have been wrong 38% of the time. Now things get interesting.. Let's look at the 3 day sample.had you traded in the direction of the underlying trend using the 3 day samples of data rather than following price, You would have correctly traded in the direction of the underlying trend, 70% of the time and wrong 30% of the time. 4 day sample you would have been correct 73% of the time and wrong 27% of the time. 5 day sample you would have been correct 76% of the time and wrong 24% of the time.etc. This is what you call "proof of principle"... 
© Forex Factory