Steam machine sales numbers are nowhere near high enough for that

So instead of it being crazy in itself that Linux use should triple within a few months, you suggest that a tiny fragment of the Linux ecosphere alone is responsible?
I think the more likely explanation here is that the collection of data or use of the data behind the stat is flawed.

I'm sure they've worked to raise awareness that Linux is a great alternative to Windows, but they aren't directly making this large of a difference.