Quote from chicagodon:
Can you elaborate on this further? I'm about 800 pages into this thread so if you have addressed this recently I just haven't seen it yet. I read that you start a new 30 day line each month, but you roll your 5 day line. When you say you match your number line to your A values, do you mean if you have a monthly A up level, you will then check your 30 day number line, and if you have a weekly A up level you will check with your 5 day line?
Sure. All my number lines serve a specific purpose. One of the things one needs to become good at when analyzing statistics is making sure the data you want to analyze matches up with another correlated data set. For example, if I wanted to measure obesity rates in the US I would not collect data or mix that data with all of North America. Statisticians often refer to this as "polluting" the data.
So the monthly number line I use is a static number line that attaches itself to a specific data set, in this case, the monthly A values. Again, we would be "polluting" the data if I used a 5 or 30 day to measure monthly A levels as the data would include specific data not from that monthly data set. So because of this, the monthly HAS to re-set. It can only contain data from "that" month. This data is NOT used to measure strength or weakness on a "cumulative" basis. Rather it is used only at a "specific" point in time. In this case, at A levels. Since both data sets should be telegraphing the same information, the conclusions from both should be consistent.
Now the 5 day and 30 day have completely different data sets. Since the 5 day is measuring momentum, the data you want in that set should be short term term data. It does not matter if that data crosses over from a previous week or month or quarter. Because what you are measuring is consistent in both sets.
Same with the 30 day. The 30 is measuring long term trend. Again, this data has to be consistent with that which you are seeking to measure. It does not matter if this data crosses over from week to month to quarter. You are measuring long term trend and therefore use a "rolling" number line as you do with the 5 day.
Most of this will be common sense to you as you let it sink in. You just need to be careful anytime you are making quantitative decisions that you are using the proper data to make that decision. For example, why would I use a 200 day moving average to daytrade shares of AAPL? Why would I use a 5 min breakout pattern to enter a long term trend?
Again, I want to stress, none of what I do is gospel. I created a way of collecting, organizing and analyzing data that fits MY style of trading. This is why I hate giving out specific values and formulas as "the way" to trade. You have to create your own values and formulas and they should make sense to YOU. I cannot stress this enough. And no, I'm not trying to throw you off my trail.
I once had an office full of traders who I taught this method to. I gave them my code, my levels, every detail of this. Not a single one of them made money. Again, a large part of this is because they didn't "understand" what it was, they simply accepted it as some kind of truth. That's not going to work. Trading is not about following someone else's system or ideas but finding your own and executing it flawlessly. I cannot stress that enough either. At the end of the day, you have to execute your strategy flawlessly. And more often then not, using someone else's work prohibits that.
I often get asked why I share this information on here. I share it because it's meaningless to others who just want to duplicate it. Just as I couldn't earn a single penny singing Billy Joel songs. This is going to requires years of absolute dedication and understanding. Upon hearing this, most traders move on to the next thing. I applaud your dedication so far to getting through the first 800 pages.