Recently, people keep using "coincidental transfers" as conclusions, and I find it a bit unacceptable... How can there be so many coincidences on the blockchain? Honestly, it's just that the paths haven't been broken down. For example, a transaction that looks like a random payment, I will first pull out the previous two or three hops: where the funds came from, whether they were just split/aggregated, whether they went through the same cross-chain router, and whether the gas/time stamps are stuck in the same window. Many "accidental addresses" are actually the same group of people using the same process to move funds.

Now some on-chain data tools and tagging systems are criticized for being laggy and possibly intentionally misleading. I really feel this: once a tag is applied, people become lazy to look at the original flow, and as a result, they follow a fake path fed to them. I no longer trust charts that jump to conclusions just by looking at tags; I prefer to spend ten more minutes checking the bridges in the middle, failed route retries, and delay queues... Anyway, I have zero tolerance for poor design.