I think the label of 'pattern' is a misnomer, on both your parts, especially where it concerns probability.
A pattern, according to the
Oxford Dictionary, is a decoration (more accurately an adornment) of a repeated and recurring design, and the adjective form gives a thing that recurring design.
In speculation, a true pattern does not beget itself, but is rather a creation of the art form of statistics, at what point do individuals in a data set become predictable that we may allocate an unwavering pattern to them? I for one think the samples far too few, which in turn make the predictable correlation extremely weak, not to mention the intense complications of the other factors creating hidden variables and confusing our ability to link causation between the two. What we label as pattern springs from our estimation of this very thing, where Golden is concerned, he does not see a pattern from this association, and thus rejects the notion that it is a pattern, outside of that he then sees the outcome of his judgement and makes the judgement that there is no proper 'pattern' which can be assigned, which he then articulates through the other forms of 'unpredictability' and the like. That itself is not a pattern, but rather the natural product of what he perceives as no pattern. Not only that, but when we say patterns we generally assume a predictability from that statistical association, it is that assumption which governs whether there is a pattern or not, and as we have not seen the future, we cannot say if the pattern will apply, most especially with such low relation.
It's really not a pattern, it's what we think of as a pattern that matters. Get to the root, I say, and when I'm at the root I see a low r^2 value and no causal relationship, therefore I wouldn't call it a pattern.