[sweetjazz]

That's cool. The next stage, it seems to me, is to force yourself to make the same observations with stats present (playing one table). Look for ways in which you can integrate stats into your observational reads, and in particular whether it increases the speed at which you develop reads and/or gives you more reliable reads.

I think Evan's point is that you eventually need to reach a point where you are consciously doing the right thing without the aid of tricks or devices to "make you do it." I don't think that means your exercise can't be valuable, just that there is a follow-up exercise to it.