ClearTrace · Execution Quality · Interactive

What execution quality actually costs you

A DEX aggregator shows you one number: the quote. ClearTrace's fork-simulation sampler requests the same trade from every aggregator and measures what happens next. Two things erode a quote before the tokens reach your wallet. The route can revert, or it can fill worse than quoted. The cheapest-looking quote is often neither.

Explore the sampler

Each line is one aggregator across four order sizes on Ethereum. Switch the metric, rebase to an indexed view to compare trajectories, click a legend chip to isolate a series, or go fullscreen. Everything hovers.

Metric
View

How to read it

Quote fails to realize (%) is the share of fork-verifiable quotes whose execution reverted when ClearTrace replayed it against a mainnet fork: the route the aggregator quoted failed to settle. Off-chain RFQ quotes are excluded here: a fork can't replay a market maker's off-chain commitment, so counting those as failures would confuse routing with reliability. Realized cost (bps) is the median shortfall of what actually filled versus what was quoted (fee + MEV + slippage + gas), for the quotes that landed. Off-chain RFQ routing (%) is the share of quotes filled through off-chain RFQ desks instead of on-chain AMMs.

Three characters emerge. Odos posts among the cheapest fills of any aggregator here, at a near-zero median cost, yet roughly one in seven of its quotes fails to realize in simulation overall (14.5%), and nearly one in five at $1k (19.8%). Its routes are all on-chain, so this is genuine: a quote you can't rely on settling isn't cheap. KyberSwap tells a routing story instead. By $1M it sends 86% of flow through off-chain RFQ makers, where essentially no other aggregator routes. That is why it's read on the RFQ axis, not by simulated failures; its actual on-chain revert rate stays low. OpenOcean almost never fails (0.4%) but charges for it, at an 11.5 bps median fill, an order of magnitude above the near-zero-cost routers. Meanwhile Uniswap's direct route and ParaSwap quietly clear nearly every quote at near-zero cost.

No aggregator wins outright, and that is the finding: no single number, least of all the quote, ranks an aggregator. Reliability, fill quality, routing topology, and how each moves with size are different questions, and the best answer changes with the trade. This is the layer ClearTrace measures that a quote screen can't show you.

What this is and isn't. These figures come from ClearTrace's controlled quote sampler: the identical trade requested from each aggregator API and replayed against a mainnet fork from a single fixed taker, so no address-rotating solver bots inflate the counts. "Fails to realize" is a fork-simulation reliability signal for on-chain routes, not a live on-chain user-revert rate. The two answer different questions. This sampler asks whether the quoted route settles exactly as offered; a genuine user's front-end retries and re-quotes, so the rate they actually experience is far lower. Odos is the clearest case: its ~14.5% here is the highest in the set, yet ClearTrace's separate on-chain measurement puts its genuine-user revert rate near 0.2–1.1% over the same period (the v6 user-only rate, which strips the address-rotating solver-bot spam that inflated older headline figures). Read the sampler as a test of the quoted route, not as the rate a trader lives with. Off-chain RFQ routes can't be fork-verified and are reported on their own axis, never as failures.

Methodology & sources

Sampler window , Ethereum, token pairs, quote samples across four size cohorts ($1k / $10k / $100k / $1M). Realized cost is the median all-in basis-point shortfall versus the quoted output for successfully simulated fills. "Fails to realize" counts reverted simulations over fork-verifiable quotes only; off-chain RFQ and non-simulatable routes are excluded from that axis and reported separately as RFQ routing share, because a fork cannot adjudicate an off-chain commitment. SushiSwap (overlaps the reliable-cheap cluster exactly), OKX (began returning fillable quotes only mid-window, from 2026-07-10, so it lacks full-window coverage) and LI.FI (began mid-window, too few full-window samples) are excluded from the chart. Every figure traces to the pinned snapshot research_drafts/snapshots/quote_samples_2026-07-20.csv: a frozen copy of the ClearTrace repo's live sampler output, pinned 2026-07-20 so every number stays reproducible even as the live seed/quote_samples.csv keeps growing.

Reviewer verification: every plotted number, traced to source

Regenerate: groupby(source, size_cohort) on research_drafts/snapshots/quote_samples_2026-07-20.csv. Fails % = mean(reverted==1) over fork-verifiable rows (rfq_routed!=True & sim_unsupported!=True); realized cost = median(all_in_cost_bps) over ok==1; RFQ % = mean(rfq_routed==True) over returned quotes.

← All ClearTrace research · Methodology · Live execution-quality dashboard