Insights on a Unique Fork Propagation at Height 928484

Hey everyone, I've been diving into some research on Bitcoin block propagation and set up a probe network to check live block propagation on the mainnet. So far, I've looked at over 2000 blocks and noticed some pretty wild propagation trends. One specific fork at height 928484 between GDPool and AntPool has me scratching my head. Anyone else seen this?

Interesting stuff! Just a heads up though, what really matters for block selection is the hashpower, not just the nodes. You can monitor loads of nodes, but that can be misleading since some key nodes might not even accept incoming connections for security reasons. So you might miss the real action. It's a tricky game.

Yeah, I feel you. Running a node isn't a big hassle for most of us, but it's a whole different ball game for mining pools. There’s all this extra security involved like checking with various nodes and making sure everything’s locked down tight. Just throwing it out there, but maybe AntPool didn’t fully see the block before they orphaned it? Who knows?

Man, I kinda wish I posted this earlier, I need more time to think it through. About that fork... I did notice something similar in my data. GDPool seems to have a more concentrated distribution compared to AntPool. This suggests that if public nodes can only reach out to peers within a limited scope, it might affect how miners broadcast blocks... What do you all think?