The same coins arriving at exchanges as on our flow page, split by how big each single deposit was. Four groups, each a share of that day's arrivals; the four add up to 100%.
53.2% how much arrived in deposits of 100 BTC or more · 30 days on Oct-22 2026, the last day in this build, measured and published by askthetape.com.
A day with 20,000 coins arriving reads one way if it came in a handful of 500-coin deposits and another if it came in thousands of small ones. This page splits the same arrivals our flow page counts by the size of the single deposit that carried each coin. Over the last thirty closed days 53.2% of everything that arrived came in deposits of 100 BTC or more; over the last year the middle day of that thirty-day number was 47.8%, and across the whole history it has been as low as 26.8% (2023-10-01) and as high as 61.7% (2024-10-09). We report each group’s median daily share separately; these medians do not describe one typical day: 45.2% for deposits of 10 to 100 BTC, 38.0% for 100 to 1,000, 10.4% under 10 BTC and 0.0% above 1,000 — that group fired on 155 of the last 364 days and is spiky by nature. 🩸 Read it as deposits by size, not depositors by size. One event here is the net coin a single transaction adds to one venue, so several customers paid into one batched transaction merge into one larger deposit: 5,290 events a day on the median of the last year, 4.15 BTC apiece across the history. 🩸 Nothing under 0.01 BTC is here at all: the tape keeps an event only above that, so the dust group is a structural zero (0 BTC over 1,756 days), and we name it instead of drawing an empty line. The group above 10,000 BTC is rare, not empty — 28 days in the whole history, 649,321 BTC in total, most recently 2025-08-01. A flat top on a one-year chart is correct. 🩸 This page closes on the previous day. The tape's own tip day is still half-written when our node bridge hands it over (about 06:10Z), so it is held back until the next slice closes it, and a day that reaches the end of the tape is never drawn here and can never be the headline number. This is a measure of composition, not a price signal, and nobody here has measured it against returns.
| leg | our exchange tape (a public copy of the chain to Sep-12 2026, our own node since) of exchange deposits, bucketed by the size of the single event. An event is one (transaction, venue) pair and its value is the net BTC that transaction added to that venue's address set. Eight half-open groups in BTC — [0, 0.01) [0.01, 0.1) [0.1, 1) [1, 10) [10, 100) [100, 1,000) [1,000, 10,000) [10,000, inf) — folded into the four you see. 1,756 rows in the file, 2022-01-01 → 2026-10-22, of which 1,756 drawn; the market-wide row only. Rebuilt from the tape, never carried forward: a day that is not on the tape has no row at all. |
|---|---|
| cadence | daily. The series CLOSES ON THE PREVIOUS DAY, after our node bridge runs (about 06:10Z): the tape's own tip day is still half-written when it arrives, so it is held back until the next slice closes it, and this page never draws it. A rolling tape is rebuilt on every run and only the tail days are recomputed, so a recent day can still be REVISED. On this build the tape reaches into 2026-10-23, the last complete day is 2026-10-22 and 0 rows were held back. A light day is not a fault of the count: the last drawn day is 22,447 BTC against a full-day median of 23,110 — rank 833 of 1,756 from the bottom. We state the included basket beside its total and compare it with the exchange-only flow series as a separate number: that total is every entity in our map, three of them not exchanges, and the exchange-only flow series reads 22,447 BTC on the same day. |
| since | 2022-01-01 · 1,756 closed days drawn |
| 🩸 events of 0.01 BTC and larger only | the tape keeps an event only when it moves more than 0.01 BTC, so the [0, 0.01) group is empty BY CONSTRUCTION — a structural zero, not «no dust today». Measured on this build: 0 BTC across all 1,756 rows. We do not draw that line; we name it. |
| 🩸 deposits by size, not depositors by size | one event = the net BTC a single transaction adds to one venue, not one customer's transfer. Deposits batched into one transaction merge into one larger event, which moves weight out of the small groups and into 10–1,000 BTC. Our own count on this build: median 5,290 events a day over the last year, 5,893 over the whole history, 4.15 BTC apiece. |
| 🩸 the 10,000 BTC group is rare, not empty | it fired on 28 days in the whole history, 649,321 BTC in total, most recently 2025-08-01. Its biggest day is 2022-11-18, 68,200 BTC — and the line that builds this series says that day was ONE event of 68,200 BTC, the largest on record (that comes from the tape itself, which is not on this page). The 1,000–10,000 group was empty on 209 of the last 364 days. A flat band is the truth about a rare event, not a rendering bug. |
| 🩸 whose deposits: our own basket | our own address map, 133 entities in it, and the row drawn here is every one of them (our map v2.1b). 57 of those entities produced at least one deposit in 2022-01-01..2026-10-22. The market-wide row drawn here includes 3 entities that are not exchanges (a mining pool, an ETF holdings address, a reserve wallet set) worth about 2.92% of the coins. Drop those three and the series reproduces our production inflow series exactly — 0 days of disagreement out of 1,715, largest difference 0.000000000 BTC. Measured against that same production series on THIS build, the row drawn here runs +2.92% above it over 1,756 shared days, which is those three entities and nothing else. |
| the four shares add up to the day | the eight groups are a decomposition of a number already on this site: on every one of the 1,756 rows the eight add back to that day's arrivals to within 0.0000000000 BTC, and the four shares drawn here add to 100% to within 0.0000000200 of a point. That is checked on every build, not once. |
| 🩸 what this page is not, and what is not on it | it is not a count of depositors, not a balance, and not a signal: nobody has measured this composition against returns, so no rule is drawn on it. The per-venue table exists — the line that builds this series keeps one row per venue per day for every venue that took a deposit, 57 of the 133 we map — but it is 7.0 MB and it is NOT on this site: the showcase carries the market-wide row only. Ask for the venue cut if you need «which exchange took the whale deposit». |
| policy | no backfill · read it here · export: whole history |
| hourly window | our own exchange tape, transaction level (rebuilt daily 06:07Z) · since 2026-07-13 00:00Z by data, 2,444 hours with data inside a 2,453-hour span → 2026-10-23 04:00Z (the Tape and Desk windows draw up to 400 days of that, never more than we hold) · captured 2026-10-23T21:21:56Z |
| this build | data to Oct-22 2026 |
| policy | no backfill · read it here · export: whole history |
The chart itself is drawn in the browser: turn JavaScript on to hover a day, pick a window and draw your line. The numbers above come from the last build, Oct-23 2026 21:24Z.
What the tape says today · what's new · all 111 metrics