Yeah, pull-to-refresh doesn't make sense on a web browser at all IMO. Its obvious, specific application is "refreshing a vertical list of entries, typically populated via a network request". This action doesn't map well to "loading a page of content".
One fundamental point of the interaction is that you're already scrolling to the top of the list because you're viewing a reverse-chronologically-sorted list of network-driven content. If, after hitting the top of the list, you keep trying to scroll up, you're already suggesting to the software that you want to see more. On a website, this implication doesn't apply. I'm just trying to get to the top of the page.
One fundamental point of the interaction is that you're already scrolling to the top of the list because you're viewing a reverse-chronologically-sorted list of network-driven content. If, after hitting the top of the list, you keep trying to scroll up, you're already suggesting to the software that you want to see more. On a website, this implication doesn't apply. I'm just trying to get to the top of the page.