crypto-execution-costs: 2026-08-29

README.md

@@ -19,52 +19,53 @@ size_categories:
 
 # Crypto execution costs
 
-What it actually costs to move money on-chain, recorded as it was quoted. Three angles: swap
-quotes from competing DEX aggregators, retail fiat on-ramp quotes, and how long an L2 takes to
-turn a sequencer promise into something L1 confirms.
+What it costs to trade, quoted at the moment someone would have traded.
 
-Quotes are computed per request against live state and stored by nobody. The live comparison is
-free to anyone at the moment they ask; a record of what was quoted over time is not, as far as we
-could find.
+Executed trades are on chain forever. Quotes for trades nobody placed are computed on demand and
+kept by nobody, so they only exist if they were recorded when they were served.
 
-## What is in here
+## Contents
 
 | name | one row is |
 |---|---|
-| `e10_quote_benchmark` | one swap quote from LI.FI, KyberSwap or CoW, at a fixed size, fee split out |
-| `e17_perp_depth` | one order-book snapshot from a perp DEX: spread, level counts, and resting notional within 5/10/25/50/100 bps of mid |
-| `e16_dex_routes` | one leg of the route a router chose: venue, pool, amount, and how many venues the trade was split across |
-| `e12_onramp_quotes` | one retail fiat on-ramp quote, buy or sell |
-| `e14_l2_preconf` | an L2 heartbeat (unsafe vs safe head) or a sequencer promise the chain replaced |
-| `e0_run_manifest` | one collection window: polls, failures, coverage counters |
+| `e10_quote_benchmark` | one swap quote at a fixed size, with the provider fee separated out |
+| `e16_dex_routes` | one leg of a route a router chose: venue, pool, amount, and how many venues the trade was split across |
+| `e17_perp_depth` | one order book snapshot: spread, level counts, resting notional within fixed distances of mid |
+| `e12_onramp_quotes` | one fiat on-ramp quote |
+| `e14_l2_preconf` | an L2 sequencer promise, and whether it held |
+
+## Reading it
+
+Quotes are what a provider served, which is not what you would necessarily have filled. Provider
+fees are separated from price so that a fee policy is not mistaken for routing quality.
+
+`e16_dex_routes` maps where routable liquidity sits rather than where volume went, and it
+surfaces venues too small to appear in volume rankings. The split widens sharply with size: one
+observation put the same pair through a single venue at small size and ten venues at large size.
+
+`e17_perp_depth` stores book shape rather than raw ladders, as cumulative resting notional within
+a band of mid. Where a spread is wider than a band, that band is correctly zero, which happens
+often on thinner markets.
 
 ## Before you build on this
 
-- `e16_dex_routes` records the route a router picked for a trade nobody placed. Executed swaps
-  are on-chain forever; a quoted route for a hypothetical size is computed on demand and kept by
-  nobody, so this maps where routable liquidity sits rather than where volume went. It surfaces
-  venues too small for volume rankings, and the split widens sharply with size: one round put
-  WBTC/USDC through a single venue at 0.01 BTC and ten venues at 10 BTC. Only KyberSwap exposes
-  a leg breakdown; the others contribute the venue name alone.
-- `e17_perp_depth` covers Aevo and Paradex, second-tier perpetual DEXs that no depth archive
-  carries -- Tardis lists 64 exchanges and includes dydx, dydx-v4 and hyperliquid but neither of
-  these. Neither venue serves its own history either. Rows are SNAPSHOTS at the poll interval,
-  not a tick-level book, so a move that reverses between polls is invisible. Depth is cumulative
-  resting notional within a band of mid; where the spread is wider than the band the value is
-  correctly zero, which happens often on the thinner markets.
-- Quote rows are what each aggregator served, which is not what you could have filled. One round
-  has LI.FI 5% above the other two; read outliers as provider behaviour rather than free money.
-- LI.FI applies a fixed fee of about 25 bps at every size. It is split into `fee_usd` precisely
-  so its pricing policy is not mistaken for worse routing. Read price and fee together.
-- The on-ramp panel is thin: Mercuryo answers full quotes without a key, Ramp gives a reference
-  price and fee bounds, and MoonPay, Transak, Guardarian and Mt Pelerin all require keys. It is
-  1.5 providers, not a market survey.
-- In `e14`, only the `violation` rows are hard to get elsewhere. The heartbeat lag can be rebuilt
-  after the fact from L2 block timestamps and L1 batch times, both of which are permanent, so
-  treat heartbeats as coverage attestation rather than as a scarce series.
-- Optimism's public endpoint serves a stale `safe` tag (tens of millions of blocks behind, where
-  Base is a dozen). Rows carry `safe_tag_plausible` so that shows up instead of being published
-  as a property of the rollup.
+- Not every provider answers every request. Refusals and rate-limit skips are written as explicit
+  error rows rather than omitted, so a provider that was unavailable is distinguishable from one
+  that had nothing to offer. Check the `error` column before treating absence as meaning.
+- One provider is queried on a subset of pairs. Its coverage is deliberately narrower than the
+  others and should not be read as a market-wide view.
+- Only some routers expose a leg breakdown. Rows from the others carry the venue name alone,
+  which is why `pool` and `swap_amount_raw` are frequently null.
+- Depth rows are snapshots at the observation interval, not a tick-level book. A move that
+  reverses between observations is invisible.
+- Depth covers a single venue, so it sizes that one book rather than the market. Treat it as
+  one participant's view; a second venue would be a control and there is not one here.
+- `e14_l2_preconf` is mostly an attested absence. Violations are rare and the value is in the
+  zeros being credible, which requires the coverage rows alongside them.
+- `lag_blocks` is only meaningful where `safe_tag_plausible` is true. One chain's endpoint
+  served a stale `safe` tag through much of the period, which yields a lag of tens of millions
+  of blocks. The raw numbers are published uncorrected so the endpoint's own inconsistency
+  stays visible; filter on the flag before using the column.
 
 Partitions are parquet, one file per collection window, under `dataset/YYYY/MM/`. This repo carries a FIXED 7-day sample (2026-08-25 to 2026-08-31) so you can check schema, coverage and quality before asking for more. It does not advance, so there is nothing to gain by re-downloading it. The full history is held privately, available on request.
 

e17_perp_depth/2026/08/2026-08-28T0242Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90942bdee0fd965534191af69d053091c4a0f60e6af3fcf0502c62eb0cb9afd1
-size 33405
+oid sha256:f31bb0430e5dc90af6750dbfc9284fd899040a7d2d957b4dbbf224dbd66f1b96
+size 25268

e17_perp_depth/2026/08/2026-08-28T0838Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8f30344134aba7819068b83b6421f9cdc54fc9ae2569008b081a6ac5ea1a03e
-size 16090
+oid sha256:d32696e0ca0c85f62544fccafd6aa739baaf5bfdb6e91e061c4c10f74b1d9752
+size 15433

e17_perp_depth/2026/08/2026-08-28T1019Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:139bdb3463a92e68bd49984ee707d369b0bb65831a7dc6bd50e23a2f5a712aaa
-size 16065
+oid sha256:b8b5d63acd003a443218e23666fb8ffd9dcc692a65d9bae8abc87f916850bd71
+size 15432

e17_perp_depth/2026/08/2026-08-28T1108Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:070141c9794f788d3aea4c26a42276e645787587a4f028be92d9c11cc1bbac44
-size 32992
+oid sha256:6dc8c38e1911dbcb2fd08bb1cb4f3a8db5a397fbafa7a35089247eb1e5839601
+size 24783

e17_perp_depth/2026/08/2026-08-28T1830Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93c5ee09f3be44c02a7a3399c31d5ee0abcb7f15c3b41f93d6e55eaaf235d76e
-size 33415
+oid sha256:e34f2aba9fe2da9a45da594b5a035e737f3205eb78a9a9ca3661495665be53a4
+size 25359

e17_perp_depth/2026/08/2026-08-28T2302Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:21dfd13a4159ec47902bd3620fad3499abffc14d278bec24392aae7d23b1b39e
-size 33584
+oid sha256:3b706c6f4f50c93084cc031050ed0d2aa6d9ed0a36cb787566c53f5cd1b46890
+size 25492

e17_perp_depth/2026/08/2026-08-29T0334Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:767147d1a3775bec2e4622e5e78305677f93fa19eec24dcc4cad17841fd790db
-size 33572
+oid sha256:67af96d2285146ab50a75c93d694ef31185e9e9982133ca08d20f6f03a13e0b1
+size 25437

e17_perp_depth/2026/08/2026-08-29T0805Z.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2661b4ae61356916aee4f726b328e8bfde59d25c2170f18d8db811c4fae1aeb
-size 33095
+oid sha256:f220917ee8d5b5408fd31357a9cfc987f0a056927e323144e90d80b7d132010c
+size 25602