I’m not surprised that you are confused by my previous post. My wording was confusing, to say the least. Let me try again.
There are countless ways to calculate an ADR, and they all purport to indicate the current “range” for a particular pair, on a particular time-frame, taking into account a particular amount of past history (the moving-average look-back period).
Some of these ADR’s are more useful (for ICT-style analysis) than others. What makes one ADR better than another? Only the degree to which it tells us something useful about the next day’s range. And there’s only one useful thing that it can tell us: What would be a “normal” range for the next day, and what would be an “extreme” range for the next day?
If “size didn’t matter” in the calculated ADR, then we could simply use a nice, flat, easy-to-read 200-day moving average of the true range — and be done with it. For the GBP/USD, the ATR(200) has been stuck around 138 - 140 for weeks. That’s an easy number to remember. No daily checking or calculating required — just use 140 pips as the magic number. If GBP/USD gets too far outside it’s 140-pip “normal” range, then trade accordingly.
Obviously, the ATR(200) would make a pretty crude ADR for our purposes here. So, obviously, fine-tuning the ADR to be more flexible and responsive, would be a good thing. How to define “flexible” and “responsive” is another matter, and that’s where I ran amuck in my previous post.
In an attempt to make my calculated ADR more flexible and responsive to the changing daily range of the GBP/USD, I have scrapped my previous algorithm, and am now trying something new. I’ll put off until later explaining the math, and my rationale for using it.
For now, I’ll post calculated ADR’s for the 12-day period that we have been discussing here. See whether you agree with me, that these figures are more useful for estimating a “normal” next-day price range.
![](http://i612.photobucket.com/albums/tt204/Day-Trader/ADRcomparison.png)