What share of reported BSC DEX volume is wash or MEV, and what is the wash-adjusted band?

Is this venue's liquidity real enough to route through?

Status: OK
Confidence: low
As of: 2026-10-09T16:16:40.000Z
Method: Method
Feed: bsc-dex-volume-realism.json

67.33%

PancakeSwap WBNB–USDT · 1,200 blocks

Band 67.33%–67.33%. Unflagged 32.67%–32.67%. Previous window 67.98% as of 2026-10-09T07:26:50.000Z.

Combined 67.33%

PancakeSwap v2 WBNB–USDT 93.14%

PancakeSwap v3 WBNB–USDT 63.76%

The solid tick is this window. A dashed tick, when present, is the previous stored window. The mark is the band from low to high, on a 0–100% scale. Pools stay in published order.

PancakeSwap WBNB–USDT, blocks 126665064-126666263. v3 0x172fcd41e0913e95784454622d1c3724f546f849 topic 0x19b47279256b2a23a1665c810c8d55a1758940ee09377d4f8d26497a3577dc83

Pool	Suspected	Low	High
PancakeSwap v2 WBNB–USDT 0x16b9…0dae	93.14%	93.14%	93.14%
PancakeSwap v3 WBNB–USDT 0x172f…f849	63.76%	63.76%	63.76%

Method

Public BSC RPC eth_getLogs for Swap events on PancakeSwap v2 WBNB–USDT, and on the v3 pool the factory resolves, over a short pinned block window. The quote side is BSC USDT. Quote atoms weight the share. They are not published.

The low band is direct self-swaps: the pair caller and the recipient are the same address. The point estimate adds same-recipient, same-block, both directions — wash or MEV, not split. The high band also flags a recipient that repeats an identical quote size at least four times. The wash-adjusted band is what that high and low leave unflagged.

The router is not the trader. Shared funders and gas-price fingerprints are not in this reading. One pair window is not every BSC DEX.

This reading is pulled again from the chain head, three times a day. The feed keeps the previous window's share beside the new one. A pull that fails, or a window with too few swaps, does not replace the stored sample.

Method URL /onchain/bsc-dex-volume-realism/#method

Sources

Limitations

Sample is PancakeSwap WBNB–USDT (v2, and v3 when the factory resolves a pool) over a short pinned block window. It is not all BSC DEX flow and not a venue board.
Same-block round trips mix wash, atomic arbitrage, and sandwiches. This method does not split them. The point estimate is that mix.
The low band is direct self-swaps only (sender equals recipient). Routed swaps rarely look like that, so the low band is a floor, not a best guess.
Identical repeated sizes can be a TWAP or a DCA. They sit only in the high band.
Shared-funder clustering and gas-price fingerprints are not measured.
The router is not treated as the trader. Actor is the swap recipient.
A scheduled pull takes the latest window at the chain head. A pull that fails, or a window under the minimum, does not replace the stored sample.

Disclosure: Money2046 publishes receipts for agent payments: fee, time, fail, facilitator, chain — or INSUFFICIENT_EVIDENCE. Not financial advice. Evidence can be insufficient. We charge you in USD. Payment processing by Airwallex. Money2046 does not move customer funds and is not a wallet, an exchange, or a payment processor.