I'm aware. However, Hanlon's razor suggests it's not intentional. Moreover, I think Occam's razor suggests the same: it's a lot simpler if we assume they just primarily test in their own browser (just like I'm sure Mozilla primarily uses Firefox) and just don't care enough to fix any performance deficiencies they do identify based on browser market share / amount they think users will care.

(Keep in mind: I'm not saying Google is doing the right thing and trying to ensure an equal experience across browsers; I'm just saying I don't think they're intentionally trying to make their pages break other browsers)