For the frozen window from 2026-07-19 13:01:50.270 UTC to 23:37:41.873 UTC, this on-chain identity did one narrow job: submit and cancel DeepBook orders at machine-like cadence. Every one of its 112,410 transactions was sponsored by the same separate gas owner, 0x59e8e02d…bdc610. The repetition is consistent with automated order maintenance, but the evidence does not establish a person, company, operator, bot, or profitability.
The rhythm is the character
These whole-history figures cover all 112,410 transactions, not just the detailed tail. The median gap was 0 milliseconds because many transactions shared timestamps; the mean gap was 339.4 milliseconds and the 90th-percentile gap was 673 milliseconds. The longest idle period was 115.975 seconds. Activity peaked at 888 transactions in the minute beginning 22:01 UTC, and the longest uninterrupted run without an execution error was 4,094 transactions.
The repertoire barely changed by hour. Of the 112,386 transactions where MotiveLayer asserted a behavior, 57,184 were place-limit-orders (50.88%), 54,866 were cancel-order (48.82%), and 336 were owner-authorized placement attempts (0.30%). The remaining 24 transactions had no asserted whole behavior. This is a paired order-book routine, not a broad-purpose wallet history.
What the orders show—and do not show
The detailed examples below come from the most recent 500 explained transactions. All 500 used the same DeepBook owner-proof pattern; 263 were placements and 237 were cancellations.
2Fn61ZxF describes a cancellation in the DEEP/USDC book at a resting price of 0.0173 USDC per DEEP and a resting size of 400 DEEP. 9VeR21xM describes a placement of 400 NS at 0.0124 USDC per NS in the NS/USDC book. Those prices and sizes describe resting order data, not execution prices or proof of a completed trade. The stored narrative also records that reaching the placement or cancellation call alone does not establish the final economic outcome.
The one failure in this 500-transaction tail was 92uPXtT1. It attempted to cancel an identified order but aborted at big_vector::leaf_remove. No resting price or removed size was retained for that attempt, and application effects were rolled back. Its structural explanation remains clear even though the transaction failed: “understood” and “successful” are separate facts here.
Across the complete history, 611 of 112,410 transactions failed (0.544%). Failures stopped most often at big_vector::leaf_remove—251 cases, 41.08% of failures—followed by balance_manager::withdraw_with_proof with 178 (29.13%) and order_info::assert_execution with 158 (25.86%).
The money boundary
MotiveLayer retained no sender-bound value-movement records for this address across the complete history: zero movement sources and zero omitted sources. That is the evidence boundary, not proof that no assets moved. The complete-history net gas total was 87,686,974,332 MIST; the most recent 500 transactions accounted for 409,721,040 MIST. Because every transaction was sponsored, those gas charges belonged to the separate gas owner rather than directly to this sender.
Nothing here supports a profit, loss, or cross-asset exchange claim. It supports a much narrower conclusion: during this frozen ten-hour window, the address repeatedly submitted and cancelled DeepBook orders, at sub-second burst cadence, through one sponsored account path, with a small but visible cancellation-failure tail.
Useful exits are 2Fn61ZxF, 92uPXtT1, sui/place-limit-orders/deepbook-owner-proof/v1, sui/cancel-order/deepbook-owner-proof/v1, and package revision 0xcaf6ba059d539a97646d47f0b9ddf843e138d215e2a12ca1f4585d386f7aec3a.