Smart Money This Week: The 24-Hour Group Swung From 6% to 80% Long, While 7-Day and 30-Day Groups Stayed About 90% Short (Nov 11–17)
Columns · · Satoshi Note Editorial
Read in: 한국어
Smart Money This Week: The 24-Hour Group Swung From 6% to 80% Long, While 7-Day and 30-Day Groups Stayed About 90% Short (Nov 11–17)
A weekly recap built from the Satoshi Index's twice-daily snapshots (09:00 and 18:00 KST), issue 3
The Satoshi Index shows what top-ranked perp traders on Hyperliquid, dYdX, Lighter and Pacifica hold right now. After issue 2 (Nov 3–10), this issue collects 12 snapshots from November 11 to 17. The shape repeated: the short-horizon group flipped hard, and the long-horizon groups barely moved.
At a glance
- 24-hour group: long share started at 10.1% on Nov 11, hit 70.8% on the 12th, 5.8% on the 13th and 79.6% on the 16th. The last snapshot (Nov 17, evening) was 63.0%.
- 7-day group: stayed between 7.3% and 15.6%. The start (8.9%) and end (8.3%) are nearly identical.
- 30-day group: 7.2% to 17.8%. The end (13.9%) is 4.2 points above the start (9.7%).
- By coin (7-day group): shorts outnumbered longs in all six coins all week. BTC ended at 57 long and 212 short accounts; HYPE shorts fell from 151 to 118.
1. The 24-hour group: three flips
The 24-hour group is the top 100 accounts by 24-hour return per exchange, up to 400 accounts. "Long share" is long notional divided by long plus short notional.
| Snapshot (KST) | Accounts with positions / ranked | Long | Short | Long share |
|---|---|---|---|---|
| Nov 11 Wed 07:35 | 385 / 479 | $236M | $2.092B | 10.1% |
| Nov 11 Wed 14:29 | 397 / 497 | $254M | $2.014B | 11.2% |
| Nov 12 Thu 08:09 | 373 / 474 | $674M | $279M | 70.8% |
| Nov 13 Fri 05:19 | 404 / 478 | $172M | $1.999B | 7.9% |
| Nov 13 Fri 15:22 | 413 / 490 | $123M | $1.984B | 5.8% |
| Nov 14 Sat 07:37 | 387 / 475 | $228M | $1.885B | 10.8% |
| Nov 14 Sat 14:54 | 383 / 469 | $266M | $1.588B | 14.4% |
| Nov 15 Sun 08:08 | 393 / 458 | $158M | $1.733B | 8.3% |
| Nov 16 Mon 07:53 | 395 / 488 | $819M | $210M | 79.6% |
| Nov 16 Mon 16:24 | 349 / 459 | $409M | $349M | 54.0% |
| Nov 17 Tue 07:06 | 344 / 474 | $465M | $472M | 49.6% |
| Nov 17 Tue 14:37 | 374 / 479 | $618M | $363M | 63.0% |
Amounts in USD. Each row is the snapshot used in that day's Satoshi Index article. The Nov 12 and Nov 15 evening snapshots have no stored values and are missing.
Same as issue 2: seven of the 12 snapshots had a long share under 15%, four were above 50%, and only one (49.6%) fell in between.
One difference: this time the group stayed at or above roughly 50% long for two straight days (four snapshots, Nov 16–17). Short notional fell from $2.092B on Nov 11 to $363M on Nov 17.
The number of accounts holding positions barely changed (344–413). A group of similar size traded places between long and short notional.
2. The 7-day and 30-day groups: about 90% short
| Measure | Start (Nov 11 morning) | End (Nov 17 evening) | Weekly low | Weekly high |
|---|---|---|---|---|
| 7-day long share | 8.9% | 8.3% | 7.3% (Nov 14) | 15.6% (Nov 17 morning) |
| 30-day long share | 9.7% | 13.9% | 7.2% (Nov 13) | 17.8% (Nov 17 morning) |
| 7-day short notional | $2.473B | $1.720B | ||
| 30-day short notional | $2.445B | $1.733B |
The long share stayed put while both long and short notional shrank. Short notional in both groups fell from the $2.4B range to the $1.7B range, a bigger drop than on the long side, so the share barely moved.
On Nov 17 morning both groups rose together (7-day 15.6%, 30-day 17.8%) and came back down by evening (8.3%, 13.9%).
3. By coin: shorts ahead in all six
Long accounts · short accounts in the 7-day group:
| Coin | Nov 11 morning | Nov 17 evening | Change in shorts |
|---|---|---|---|
| BTC | 60 · 172 | 57 · 212 | +40 |
| ETH | 58 · 138 | 59 · 170 | +32 |
| HYPE | 28 · 151 | 34 · 118 | −33 |
| SOL | 43 · 89 | 36 · 102 | +13 |
| NEAR | 19 · 69 | 15 · 70 | +1 |
| ZEC | 24 · 83 | 25 · 81 | −2 |
BTC and ETH added 40 and 32 short accounts. HYPE, which tilted furthest toward shorts in issue 2, lost 33 short accounts and gained 6 long ones, though it is still 118 short to 34 long.
The 24-hour group looks different. On Nov 16 morning BTC was 120 long · 94 short and HYPE 105 · 50. By Nov 17 evening ETH (84 · 79), HYPE (94 · 57) and ZEC (52 · 45) were long-heavy, while BTC (77 · 87) was still short-heavy.
4. What these numbers do and don't say
Short-window swings follow that day's price. "Top 100 by 24-hour return" is re-drawn at every snapshot. If price moved one way, accounts on that side rise into the list. A move from 5.8% to 79.6% long can reflect a change in who is on the list, not traders changing their minds.
The long-horizon groups are closer to a baseline. The 7-day and 30-day groups changed membership slowly and stayed at 7–18% long for a second week.
A short isn't always a bearish bet. Accounts holding spot can short perps as a hedge. The index only sees the perp side.
This is a snapshot. Twice a day, at slightly different times. Positions opened and closed between snapshots are invisible. This is not investment advice.
In one line: the short-horizon top group sat short for days and then flipped long within a day, repeatedly, while the long-horizon top groups stayed about 90% short with shrinking short notional.
Method
- Source: Satoshi Index daily snapshots. Top 100 accounts per exchange by 24-hour, 7-day and 30-day return on Hyperliquid, dYdX, Lighter and Pacifica, positions summed across the four.
- Snapshots: 12 snapshots from Nov 11 morning to Nov 17 evening as printed in the daily articles, taken between 05:19 and 16:24 KST. The Nov 12 and 15 evening snapshots are missing.
- Numbers are as printed in those articles. The live Satoshi Index refreshes every 30 minutes (every 3 hours for dYdX) and will differ.
FAQ
What's different from issue 2?
Issue 2 (Nov 3–10) used one evening snapshot a day; this issue uses 12 snapshots at 09:00 and 18:00. The 24-hour group also stayed long-leaning for two straight days.
Does a high long share mean price will rise?
These numbers can't tell you that. Short-window rankings pull in the accounts that were right during the window, so a high long share may be the result of price having already risen.
Where can I check the numbers?
The Satoshi Index shows the current values by group and coin.