Thanks for pointing that out.
NSDIC explains why they do this in their website.
That's why I usually stick to absolute numbers and ratios, rather than anomalies pegged to arbitrary standards.
In my opinion, it is pretty clear from the data the arctic summer sea ice is diminishing.
I suspect the rate at which it is diminishing is increasing, but that is not as clear from the 30 year sample we have. That increase in downward slope my eye wants to fit into the points could be purely subjective.