ClearTrace

DEX Weekly

October 6, 2026  ·  Ethereum, Base, Arbitrum, Optimism

1 · Lead finding

Base's router traffic doubled in the week to October 5, while Ethereum's and Arbitrum's fell.

Two of our tables show it, one counting dollars by the contract that carried the trade and one counting transactions sent to venue routers; they count overlapping sets of the same router activity in different units, dollars settled by carrying contract and transactions sent to venue routers, so they corroborate each other without being independent. Our attribution, which assigns each settled trade to the contract that carried it, has the chain at $5.77B in the window written 2026-09-29 and $7.60B in the window written 2026-10-05, up 32%. Our revert query, which counts transactions sent to each venue's routing contracts, has Base at 3,735,035 user transactions in the week to September 29 and 7,598,031 in the week to October 5. A user transaction there is one from a sender whose own transactions to the router succeeded at least 90% of the time over the window, a threshold a well-run bot clears as easily as a person. Dollars rose by a third and transactions doubled, so the average Base transaction fell by about a third. That divides one table by another; read it as a direction.

It was a Base week on the chains we can compare. In Dune's DEX-category table the seven days to October 5 were down on every other chain in that table against the seven days to September 28, and router transactions fell on Ethereum and Arbitrum. Optimism, which Dune's table does not carry, rose 2.4% on transactions and 54% on attribution volume, $148.4M to $228.4M. Four unknown contracts at $26.4M to $27.0M each, absent from the previous window, add $107.4M against that $80.0M net rise, so other Optimism contracts fell. Identical volumes in a first window suggest one operator; one window cannot establish it, so they are counted in section 2 and not named.

DEX-category volume from Dune's table, which does not carry Optimism; router user transactions from our revert table, which does not carry BNB or Solana. The two revert snapshots were written 6.1 days apart, so their seven-day windows share about a day; the DEX-category windows do not overlap.

Seven contracts account for the whole of the net rise: their increases sum to $1,854M against a chain rise of $1,828M, so the rest of Base, including its largest carrier, a Wintermute-linked market-maker contract that went from $1.79B to $1.69B, fell slightly on net. The seven are Uniswap's UniversalRouter, Aerodrome's router, 1inch's resolver-settled Fusion path, a vault swap executor we have named only by role, two contracts labelled earlier (one attributed to Binance Wallet by Arkham on September 14, one carrying Etherscan's verified name "Diamond" since September 21, which describes a proxy pattern and says nothing about an operator), and OKX's router. Not every venue took part: KyberSwap's Base transactions fell to 0.8x of the prior week, CoW's to 0.1x, and the venue with the most Base router transactions a week earlier, which is under the publication freeze and not named here, fell by about 6%, while Uniswap's rose 3.5x, OKX's 2.8x, 1inch's 11.9x and SushiSwap's 41.2x. The transactions are not the dollars: dividing each venue's attribution volume by its transaction count puts Uniswap's router near $350 a transaction, OKX's near $290, KyberSwap's near $250, 1inch's near $130 and SushiSwap's near $7, on $3.6M of volume. SushiSwap's 41x is a count of very small transactions, and the same two tables are being divided, so treat these as scales. The user-only revert rate moved where the small transactions arrived: SushiSwap's Base rows went from 0.18% to 2.91% and Aerodrome's from 0.48% to 0.94%, while Uniswap's held at 2.25%, OKX's eased from 2.30% to 2.11%, KyberSwap's from 1.49% to 1.13%, and 1inch's went 1.75% to 1.80%.

Base attribution volume by carrying contract, the seven largest increases between the windows written 2026-09-29 and 2026-10-05. A contract under the publication freeze is excluded from this list and included in the chain totals.

Three different rows can claim the biggest step, depending on the measure. By dollars it is Uniswap's UniversalRouter, +$718.0M. By ratio it is the Binance Wallet contract, 50x, then the Diamond proxy, 33x. 1inch's Fusion path is third by ratio, $28.9M, $13.4M, $224.8M across the windows of 2026-09-21, 2026-09-29 and 2026-10-05, 17x off the middle window and 7.8x off the first, with the same venue's direct AggregationRouter down to $4.2M from $5.7M. It gets the detail here because it is the one row we can check against two other tables. Fusion is the intent path: the user signs an order and a resolver submits the settling transaction. Its Ethereum figure moved $220.4M, $73.5M, $179.4M over the same windows, so week-to-week swings of that size are normal for it; the Base step is larger than any of them.

A Fusion settlement does not go to a 1inch router, so our trace vectors alone would file it as an unknown proxy; the row is named from Dune's decoding of 1inch's own contracts. The same decoding feeds Dune's aggregator table, which has 1inch at 6.0% of Base's aggregator-category volume in the seven days to September 28 and 42.0% in the seven days to October 5, in a category that grew from $395.2M to $601.4M. DefiLlama's seven-day Base figure for 1inch, fetched October 6 from its protocol endpoint, is $255.7M against our $229.0M; we checked how DefiLlama computes it, and its adapter is a Dune query over 1inch's own spellbook table, so that is the same decoding read a third time. Ours sits about 10% under both; our denominator collapses a multi-leg fill to one notional per transaction and excludes fee-recipient legs, which is the direction of that gap, and we have not reconciled it line by line. Direction and scale agree; none of the three says who sent the flow. The question a reader asks first, whether an incentive or points programme is paying for these transactions, cannot be answered from these tables, which hold no sender identity and no reward flow.

The data bounds what the flow is without naming it. Dune's market table splits each venue's volume by a sender-frequency rule, more than 3,500 trades in the week, and on Base almost all of the week's increase sits above that threshold while volume from senders below it was close to flat. We do not publish that split as a bot share, because the rule measures how often an address trades and not whether a person is behind it, and it has never been validated on Base; as a statement about frequency it stands. For 1inch's Fusion rows it says even less, since the sender of a Fusion settlement is the resolver. Dune's sandwich feed for Base has been one pair for two weeks: the 500 rows we hold on both the September 29 and October 5 seeds are almost all USDC-XDP, each sample covering about three minutes, with a median victim of a few hundred dollars. What changed this week is the detector's weekly Base aggregate, which jumped and is quarantined until the floored query lands in the seed, so this issue prints no sandwich figure. A three-minute sample cannot say whether that pair and this surge are one event. One of the labels that landed on Base this week is a contract Sourcify verifies as XdpUsdcVault, carrying $1.7M, the first named contract we hold that references that pair; it has not been traced. Two of the seven contracts above carry labels two and three weeks old, one a third-party attribution and one a verified name that only describes a proxy pattern; the forensic routine works that queue in volume order and what it finds will be in a later issue.

2 · Who carried the flow

The contracts behind this week's volume, named where we can name them

Attribution volume by the contract that carried the trade, per chain, for the attribution window written on the date shown (the refresh is weekly, usually a Monday, and a missed one is forced on the next sync). A name is printed only when our labeling has one; the rest is in the totals and in the count below each table.

Ethereum, window written 2026-10-05: $5.81B across 3,874 contracts.

12 unnamed contracts over $5M carried $222.4M (3.8% of the chain); 31 contracts carrying $320.9M are withheld under a publication freeze and included in the total.

Base, window written 2026-10-05: $7.60B across 3,492 contracts.

35 unnamed contracts over $5M carried $618.6M (8.1% of the chain); 3 contracts carrying $64.2M are withheld under a publication freeze and included in the total.

Arbitrum, window written 2026-10-05: $879.1M across 724 contracts.

2 unnamed contracts over $5M carried $79.0M (9.0% of the chain); 3 contracts carrying $33.0M are withheld under a publication freeze and included in the total.

Optimism, window written 2026-10-05: $228.4M across 209 contracts.

5 unnamed contracts over $5M carried $113.2M (49.5% of the chain); 3 contracts carrying $4.4M are withheld under a publication freeze and included in the total.

Percentages are of the chain's attribution total in the window; the five rows are the largest named carriers, not the whole.

Entrants that persisted. A contract over $5M this window that was also in the previous attribution window and not in the one before. A contract seen in one window only is churn and is not listed.

No named contract met the rule this week.

Unnamed entrants, by chain: Ethereum: 1 unnamed, $5.2M; Base: 13 unnamed, $207.6M; Arbitrum: 2 unnamed, $79.0M. Their addresses are in the issue manifest and the labeling queue.

Named this week. Labels dated inside the seven days to 2026-10-06, on contracts carrying at least $1M in the current window, that were not already in the previous sync's label table. A label the labeler merely revisited is not new and is not listed.

A further 2 contracts carrying $4.9M received a role label (proxy shape, delegated EOA, bot, bucket) rather than a name; they count as attributed in coverage and are not listed as names.

3 · Execution scoreboard

Realized fill against the Uniswap default, and who reverts

Each venue's realized output against the same trade routed through the Uniswap default, on Ethereum, matched on pair, size cohort and 30-minute window. This is the realized comparison, not the quote gap: a quote gap measures a venue against its own quote and carries no price level.

Median basis points of realized output against the Uniswap default route. Positive means the taker receives more than the default route delivers, negative less. A cell is printed only when rated: at least 30 matched windows spanning at least 7 days and not stale; n/a otherwise. 777 to 3,271 matched windows per printed cell. Sample captured 2026-06-23 to 2026-10-06, 142,964 successful captures. Venues under a publication freeze or wound down are not shown.

Router revert rates, Ethereum, 7 days. Every router we measure that we may name, best first. RFQ and intent venues are not routers (the user does not submit the swap), and a venue whose execution model is unclassified is not compared either. 1 router with fewer than 10 reverted user transactions in the window is not listed: that few events is not a rate.

User revert rate is the revert rate over senders whose own transactions to these routers succeeded at least 90% of the time over the window. It is a threshold on each sender's own outcomes, not a classification of who the sender is. Data written 2026-10-05. The highest rate across the full measured population on this chain, named here or not, is 1.63%.

4 · Fees, declared and undeclared

What we have reproduced, what DefiLlama reports, and what gas costs

Only a fee we have reproduced from our own quote sample is called a fee here. A flat shortfall we have not reproduced is a parameter we have not attributed, and section 3 labels it that way.

Explicit fees we have reproduced.

Protocol fees, 30 days, as DefiLlama reports them.

DefiLlama's figures and DefiLlama's categories, fetched 2026-10-05; take rate is the share of fees DefiLlama books as protocol revenue. Shown for context; nothing here is reproduced by us. Venues under a publication freeze or wound down are not shown.

Gas as a fraction of a $10,000 trade.

Median of realized gas valued in USD over the trade notional, for successful captures at the $10k cohort with a valued gas price. On the L2s the figure is a lower bound: it covers execution gas and not the L1 data fee.

5 · Market context

Share of volume, inside one category at a time

Seven days to 2026-10-05, with the change in points against the seven days to 2026-09-07. Each table is one category of one chain: the DEX category reads pool-level fills and the Aggregator category reads aggregator-routed fills, and because an aggregator's fills also appear at the pool, the two overlap and are never summed. Total volume; no organic-versus-bot split is published.

Share of DEX-category volume, Arbitrum, 7 days to 2026-10-05 ($898.5M)

Share of DEX-category volume, Base, 7 days to 2026-10-05 ($8.02B)

Share of DEX-category volume, BNB Chain, 7 days to 2026-10-05 ($9.00B)

Share of DEX-category volume, Ethereum, 7 days to 2026-10-05 ($6.47B)

Share of DEX-category volume, Solana, 7 days to 2026-10-05 ($17.32B)

Share of Aggregator-category volume, Base, 7 days to 2026-10-05 ($601.4M)

Share of Aggregator-category volume, Ethereum, 7 days to 2026-10-05 ($3.07B)

Source: dune_global_market, written 2026-10-05, which reads dex.trades for the DEX category and dex_aggregator.trades for the Aggregator category, both collapsed to one notional per transaction. Venue-days under $1M are not in the source. Optimism is not in this source; its attribution appears in section 2. A venue under a publication freeze is counted in Other and not named; removing it from the denominator would misstate every other row. Other absorbs rounding, so each table sums to 100.0.

The dashboard carries the full attribution tables and the methodology notes set out the difference between the quote gap and the realized comparison, and what the user-only revert rate does and does not say.

ClearTrace · Cross-chain DEX attribution. Every figure in sections 2 to 5, and the lead's, is listed with its source table, filter and as-of date in the issue manifest, committed beside this issue. Sections 2 to 5 are generated from the data; section 1 is written.