The thing is, the people on the other side say identical things about HN, just with one bit flipped (which political side they identify with). The comments are otherwise so similar that I don't think it's plausible that one set is accurately analyzing HN while the other set is just wrong. Rather, some underlying phenomenon is giving rise to both sets.

You can not only predict people's politics from what they claim about HN bias, you can predict the level of intensity they feel about it. At scale, it's a mechanistic phenomenon. The question is what's the mechanism. What I call the notice-dislike bias is my attempt to explain the mechanism.

I'd be very interested in any other explanation, but it's not plausible to think that the answer is "one side is good and has good views and sees HN accurately, while the other side is bad and has bad views and sees HN completely the wrong way". Even though the commenters in question all seem to think that.

It's common for people to interpret the above argument as a defense of political centrism, but that's a non sequitur. You can't derive a centrist position or any other political position from what I'm saying here, which is an empirical observation about social psychology on the internet.