Recently, while browsing logs on the chain, I kept seeing those "coincidental transfers"—A just received money and immediately sends it to B, B then swaps it in a pool and returns to C. To put it plainly, many of these aren't coincidences; it's just that the paths aren't broken down: first, trace the source of funds by time window, then see if the same routing contract / same set of relayer addresses are moving the funds, and finally check if it involves approval, swapping, cross-chain steps. Once broken down, the mystery immediately drops by half, and only the remaining half is worth watching.

By the way, I see that on the L2 side, there's again chatter about TPS, fees, subsidies... I'm now more concerned about "where these users are coming from and where they're going," no matter how cheap it looks, the flow on-chain can't be fooled.

Someone also complained about me: you're just looking at a bunch of 0x-starting mempool records, can't you just clear out the trash first... Fine, I'll go clear the trash first, and the script will keep running.