88 mid-size books ($10–100M each, all coins together) hold BTC: long 52 %, short 48 % of their BTC dollars — a description of who leans where, not a forecast.
+4.8 % net long net lean of mid-size books ($10–100M), in their BTC dollars · last closed day Oct-22 (positions polled Oct-22 22:01Z) · daily free, hourly on Tape, daily history through Oct-22 2026, measured and published by askthetape.com.
What this means: the line is how one size class of Hyperliquid books splits its dollars in a coin between long and short, one snapshot a day since Jul-06 2026; the newest point is the last closed day Oct-22 — +100 % would be every dollar long, 0 as much long as short. What it does not tell you is where price goes next: we checked the size classes from $1M up, and all books together, on our own record since April 2026 and nothing survived the check, and a forward test registered in September 2026 is judged around mid-2027. So read it as a picture of who holds which side, not as a forecast.
| leg | our own hourly record of Hyperliquid positions: about 7,000 accounts we poll every ~18 minutes on Hyperliquid's main perpetuals exchange (HIP-3 markets are not in it). The hour's snapshot is the fullest poll cycle between 2 hours and 20 minutes before the hour; an hour whose cycle carries fewer than half the day's median rows is thin and is not used. Read-only from our production database; only class totals leave it. |
|---|---|
| daily free, hourly on Tape | this page draws one snapshot a day: the last closed day Oct-22, whose positions were polled at 22:01Z Oct-22. That daily layer is free, since Jul-06 2026. The same classes hour by hour for the last 14 days, the last closed hour and the coin-by-coin view at that hour are on the Tape plan, under the same rule, except that there the $10–100M class prints from 10 books (here from 20). |
| books and classes | a book here is one account; its size is the dollars of all its positions added up, every coin, long and short alike, at that hour. Classes: $100M or more · $10–100M · $1–10M · under $1M. Sub-accounts of one owner count as separate books — we do not link accounts. |
| net lean | (long − short) ÷ (long + short) over the class's dollars in the coin: +100 % = every dollar long, −100 % = every dollar short, 0 = as much long as short. All tracked books together sit near 0 by construction — every long on the exchange has a short on the other side — so the split between classes is the picture, not the total. |
| 🩸 the privacy rule | a cell (coin × class × day, or hour on Tape) is printed only when enough books of the class hold the coin — on this free daily page at least 20 for the $10–100M class and 10 for the smaller classes, on the hourly Tape layer 10 for every class — and the largest of them carries no more than a third of the class's dollars in it; a day that fails the rule is a gap in the line. Why 20 here: the free history grows a day at a time, and anyone can set two neighbouring days side by side; when one book enters or leaves a coin, the difference can point at it. On our own check of the day-to-day steps since Jul-06 (measured Sep-26 2026), with 10 books about 10 % of the $10–100M books could be picked out on at least one day, from 20 books about 5 %; the smaller classes hold more books per coin. Books of $100M or more are never shown coin by coin. No addresses, names, ranks or sizes of single books anywhere on the page or in its files; dollar amounts are rounded as they are shown. The free page and its files carry one snapshot a day; the hourly layer is at least an hour old. Hyperliquid positions are public on-chain; the rule means this page is not the source of an address and does not make one easier to find. |
| the largest books | accounts above $100M are not shown: too few to stay anonymous — not coin by coin, not as a count, not as a share of the total. |
| coverage | coins printed on the last closed day Oct-22: mid-size books ($10–100M) 25 · $1–10M books 86 · books under $1M 151. Coin buttons: the 23 coins where the $10–100M class passed this page's rule (20 books) on at least 13 of the last 14 closed days; each coin's class buttons follow the same test (10 books for the smaller classes). All tracked books together are not printed coin by coin: that total minus the three classes would give away the $100M+ books coin by coin. How much of the exchange this sees, measured Sep-26 2026: the tracked books held about 85 % of the open positions on Hyperliquid's main exchange (open interest counted on both sides, times the mark price); BTC 82 %, ETH 93 %. |
| price line | the coin's mark price read from the same snapshot (dollars ÷ coins over every tracked position), drawn as % change from Jul-06 2026 on the net-lean axis; the dollar price is in the hover. |
| days | each day is its last closed hour (00:00Z of the next day; if that hour is thin, 23:00Z or 22:00Z). History starts Jul-06 2026: the tracked pool grew about 4.5× on Jul-05 2026, so earlier days describe a different set of accounts and are not drawn. |
| not a forecast | a description of who holds which side. On our own record since April 2026 the lean of the classes from $1M up, and of all books together, told nothing about the next day or week (the under-$1M class on its own was not tested; the check's own positive control failed, so small effects are not ruled out either); a forward test registered in September 2026 is judged around mid-2027. |
| cadence | the site rebuilds hourly; the free layer moves once a day, when a day closes — the newest point is the last closed day Oct-22. The hourly layer is on the Tape plan. |
| since | 2026-07-06 |
| judges | every build recomputes each printed cell from the carrier by the same formula and rule, checks that no $100M+ cell, no address-like string and no hour-by-hour number reaches the free page or its files, and an independent script re-reads the last day straight from the database. |
| this build | data to Oct-22 2026 |
| policy | no addresses, names or single-book sizes · no cell under 10 books (20 for $10–100M books on the free page) or with one book over a third · accounts above $100M are not shown: too few to stay anonymous · daily free, hourly on Tape · read it here · export: whole history |
| coin | net lean | long / short | books | class dollars in the coin | the coin's share of the class's dollars |
|---|---|---|---|---|---|
| ETH | +23.5 % net long | 62 % / 38 % | 96 | $1.11B | 27 % |
| BTC | +4.8 % net long | 52 % / 48 % | 88 | $1.11B | 27 % |
| HYPE | +1.9 % net long | 51 % / 49 % | 82 | $493M | 12 % |
| SOL | +28.6 % net long | 64 % / 36 % | 55 | $265M | 6 % |
| ZEC | +20.5 % net long | 60 % / 40 % | 47 | $164M | 4 % |
| NEAR | +18.4 % net long | 59 % / 41 % | 47 | $153M | 4 % |
| LIT | +21.8 % net long | 61 % / 39 % | 32 | $39M | 1 % |
| DOGE | +32.1 % net long | 66 % / 34 % | 28 | $37M | 1 % |
| PUMP | +0.5 % net long | 50 % / 50 % | 33 | $36M | 1 % |
| VVV | +18.6 % net long | 59 % / 41 % | 25 | $31M | 1 % |
| WLD | +10.1 % net long | 55 % / 45 % | 30 | $27M | 1 % |
| AAVE | +8.3 % net long | 54 % / 46 % | 27 | $27M | 1 % |
| ENA | −13.5 % net short | 43 % / 57 % | 31 | $22M | 1 % |
| BNB | +16.6 % net long | 58 % / 42 % | 24 | $20M | 0 % |
| LINK | −3.2 % net short | 48 % / 52 % | 28 | $19M | 0 % |
| UNI | −56.5 % net short | 22 % / 78 % | 26 | $18M | 0 % |
| TAO | −54.7 % net short | 23 % / 77 % | 23 | $14M | 0 % |
| GRASS | +9.4 % net long | 55 % / 45 % | 23 | $13M | 0 % |
| XPL | +29.9 % net long | 65 % / 35 % | 22 | $12M | 0 % |
| ZRO | +18.7 % net long | 59 % / 41 % | 26 | $12M | 0 % |
| BTC | +9.7 % net long | 55 % / 45 % | 411 | $701M | 30 % |
| HYPE | +3.0 % net long | 52 % / 48 % | 336 | $458M | 20 % |
| ETH | −0.7 % net short | 50 % / 50 % | 297 | $337M | 14 % |
| ZEC | +36.7 % net long | 68 % / 32 % | 194 | $136M | 6 % |
| SOL | +12.3 % net long | 56 % / 44 % | 163 | $101M | 4 % |
| NEAR | −21.1 % net short | 40 % / 60 % | 151 | $60M | 3 % |
| LIT | −48.0 % net short | 26 % / 74 % | 95 | $44M | 2 % |
| XRP | +6.0 % net long | 53 % / 47 % | 105 | $34M | 2 % |
| PUMP | +34.8 % net long | 67 % / 33 % | 96 | $27M | 1 % |
| UNI | +19.9 % net long | 60 % / 40 % | 83 | $26M | 1 % |
| XMR | −36.0 % net short | 32 % / 68 % | 62 | $25M | 1 % |
| ZRO | +18.7 % net long | 59 % / 41 % | 82 | $23M | 1 % |
| WLD | −4.9 % net short | 48 % / 52 % | 99 | $22M | 1 % |
| VVV | +22.7 % net long | 61 % / 39 % | 87 | $19M | 1 % |
| AAVE | −9.9 % net short | 45 % / 55 % | 70 | $19M | 1 % |
| ENA | −1.7 % net short | 49 % / 51 % | 103 | $18M | 1 % |
| DOGE | +14.1 % net long | 57 % / 43 % | 76 | $16M | 1 % |
| XPL | +13.3 % net long | 57 % / 43 % | 78 | $15M | 1 % |
| TAO | +5.1 % net long | 53 % / 47 % | 75 | $13M | 0 % |
| MON | −16.9 % net short | 42 % / 58 % | 62 | $13M | 0 % |
| BTC | +22.4 % net long | 61 % / 39 % | 2,553 | $277M | 31 % |
| HYPE | +23.7 % net long | 62 % / 38 % | 1,925 | $161M | 18 % |
| ETH | +33.3 % net long | 67 % / 33 % | 1,490 | $101M | 11 % |
| ZEC | +56.9 % net long | 78 % / 22 % | 863 | $43M | 5 % |
| SOL | +25.4 % net long | 63 % / 37 % | 829 | $30M | 3 % |
| NEAR | +43.1 % net long | 72 % / 28 % | 718 | $27M | 3 % |
| PUMP | +45.3 % net long | 73 % / 27 % | 503 | $19M | 2 % |
| LIT | +59.9 % net long | 80 % / 20 % | 446 | $16M | 2 % |
| ENA | +79.7 % net long | 90 % / 10 % | 475 | $12M | 1 % |
| XRP | +43.2 % net long | 72 % / 28 % | 454 | $10M | 1 % |
| XMR | +8.9 % net long | 54 % / 46 % | 318 | $9.1M | 1 % |
| WLD | +60.8 % net long | 80 % / 20 % | 351 | $9.0M | 1 % |
| ZRO | +47.6 % net long | 74 % / 26 % | 279 | $7.5M | 1 % |
| VVV | +36.6 % net long | 68 % / 32 % | 360 | $7.5M | 1 % |
| UNI | +67.2 % net long | 84 % / 16 % | 372 | $6.6M | 1 % |
| TAO | +58.8 % net long | 79 % / 21 % | 354 | $6.3M | 1 % |
| DOGE | +60.6 % net long | 80 % / 20 % | 367 | $6.0M | 1 % |
| LINK | +57.0 % net long | 78 % / 22 % | 328 | $5.9M | 1 % |
| AERO | +52.8 % net long | 76 % / 24 % | 232 | $5.7M | 1 % |
| kPEPE | +1.3 % net long | 51 % / 49 % | 220 | $5.7M | 1 % |
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