The entire article uses some very odd grammatical conventions. The author is from the Netherlands, which I think might account for some of the oddness, but I can't tell if the capitalization is because of that or just because of the desire to be quirky.
I didn't enter with a bias, I was just thinking "LOL, how are they going to justify this, this is gonna be good."
And then they had to be something completely different and yet completely justified. Darn.
One reading which might be valid is that testing on hardware that lives in the happy path will lead you to think your code has less issues than it actually has on real user systems.