Karel --
I have looked at the 6 month version of the chart.
Yes, what you see is correct.
Yes, what I see is correct.
We are seeing different things because we are looking at and for different things.
Your are focusing on up and down. (magnitude)
I am focusing on left to right. (time)
A critical part of the reasons for XDEV's performance is due to the inherent time lag of the 12 day SMA. (At least until yesterday). Except for the one day MA (and then what is the point), all moving averages, by their nature, are lagging indicators of the time series data they are applied to. The SMA is the slowest (lags the most) of the kinds of MAs I am aware of. The current value of the SMA is really more reflective of the data point at the midpoint of the MA time span used. The base assumption of the SMA is that all of the data points are of equal value. They all contribute equally to the current MA value. In the case of the 12 day SMA, the equal contribution is 8.33...%. The data point 12 days ago is just as important as today's data point. Move ahead one day. The oldest data point in yesterday's MA time span drops off. What was yesterday worth 8.33...% of the value of the MA, today is worth zero.This is logically inconsistent, (at least that's what Spock tells me). It seems very clear, that either yesterday it wasn't worth 8.33...% or today it is not worth zero. The SMA tries to have it both ways. And we buy it. In any case, I cannot grasp the concept intellectually, that the data point 12 days ago is just as important as today's data point, in helping us to deal with tomorrow. If the data point today is higher than the oldest data point dropped off the average goes up, if the data point today is lower than the oldest data point dropped off the average goes down. WOW, the significance of that higher-lower range, of today's data point, seems to me, to be worth more than 8.33...%+...hence EMA, and it's additional benefits. (Please excuse the lecture, I'm just clarifying my thinking through the issue, and writing it out helps.)
Myst has very cleverly found a way to turn THE major "weakness" of the SMA time lag into a "strength" in XDEV. For this (and other things), he has my admiration. I do not like logical inconsistencies. I believe that systems built on or around illogical, flawed concepts have shaky foundations, to a degree. I want to see XDEV succeed. I would like to be able to contribute to it's success, if I can, with the talents I have. (I also wouldn't mind having a more comfortable retirement, when the time comes, and it appears, so far, that XDEV may be able to significantly help). So, I have a vested interest in XDEV being as good as it can be.
Maybe the current target of my attention, the SMA time lag, is just irrelevant crap in the bigger XDEV scheme of things, but then again, maybe not. I'm checking it out and looking for better alternatives.
Does all of this give you a better understanding of where I'm coming from?
Thanks,
Robo
Market Data 
Markets 
AI
