The Long/Short Ratio Comes in Two Versions — 35% by People, 11% by Money
Columns · · Satoshi Note Editorial
Read in: 한국어
There is no single long/short ratio. Count accounts and BTC is 34.6% long; count dollars and it is 10.7% long. Same moment, same accounts.
As of 07:14 KST on November 4, the Satoshi Index BTC page tracked the BTC positions of 243 top accounts.
84 were long and 159 were short.
By headcount, longs were roughly a third.
But adding up position sizes, longs came to about $109M and shorts to about $910M. By money, longs were 10.7%.
So "what is the long/short ratio?" has two answers.
A long/short ratio has to be read as a people version and a money version, and the money version can be swung by one or two large accounts.
1. What the long/short ratio is
The long/short ratio expresses how much of the market is positioned long (betting on a rise) versus short (betting on a fall).
The catch is what "how much" counts:
- By people (account count): how many accounts hold a long, and how many hold a short
- By money (position size): the total dollar size of longs against the total dollar size of shorts
Exchanges and sites differ on which one they publish, and on which accounts they include. Comparing ratios without knowing that means comparing different things.
2. BTC, ETH and SOL side by side
We took the Satoshi Index top 100 accounts by 7-day return on each of Hyperliquid, Lighter, dYdX and Pacifica at one moment and computed both versions.
| Asset | Longs | Shorts | Long share, by people | Long share, by money |
|---|---|---|---|---|
| BTC | 84 accounts · $109M | 159 accounts · $910M | 34.6% | 10.7% |
| ETH | 58 accounts · $73M | 138 accounts · $667M | 29.6% | 9.9% |
| SOL | 45 accounts · $5.7M | 101 accounts · $184M | 30.8% | 3.0% |
For all three, the people version sits around 30% and the money version between 3% and 11%.
SOL shows the widest gap. Nearly one account in three is long, but longs hold only 3% of the money.
The longs are not few, just small. The median BTC long is about $48K; the median BTC short is about $157K.
The median is the position size of the account in the middle when accounts are lined up by size.
3. The money version moves with one big account
The money version feels closer to "real" exposure, but it has a weakness.
The largest BTC position is a single Hyperliquid short of about $184M — 20% of all BTC short exposure in this set.
Remove that one account and recompute the long share by money:
| Calculation | BTC long share by money |
|---|---|
| All accounts | 10.7% |
| Excluding the largest account | 13.1% |
| Excluding the three largest | 19.0% |
One account moves it by 2.4 points; three move it by 8.3.
ETH has the same shape: 9.9% becomes 13.9% without one account and 18.8% without three.
SOL goes from 3.0% to 4.5% to 7.0%.
So the money version is less "the mood of the market" and more "how a few large accounts happen to be positioned right now."
4. The people version has its own flaw
None of this makes the people version better.
It counts a $1,000 position and a $10M position as one account each.
A wave of small accounts can swing the ratio while representing very little money.
The two numbers cannot stand in for each other:
- The people version shows how many accounts are on the same side.
- The money version shows where the dollars are stacked.
A wide gap between them is information in itself: shorts are concentrated in a few large accounts, while longs are spread across many small ones.
5. What these numbers cannot tell you
Some limits are worth stating plainly.
First, the scope is top accounts, not the whole market.
Second, it is a snapshot. Positions refresh every 30 minutes (about 3 hours for dYdX), and a ratio can flip within a day.
Third, hedges are invisible. An account may hold the opposite exposure in spot or on another exchange.
Fourth, none of this says where price goes next.
Wrap-up: ask "counted by what?" before reading the ratio
Top BTC accounts were 34.6% long by people and 10.7% long by money.
By money, the figure moved to 13.1% without the largest short and 19.0% without the largest three.
When you meet a long/short ratio, the first thing to check is whether it counts people or dollars, and which accounts it collects — that is where reading it correctly starts.
Frequently Asked Questions
If the ratio is 70:30 short, which side has the edge?
You can't tell from the ratio alone. More shorts means more accounts or dollars are positioned for a decline; it is not evidence that price will go that way. The figures here describe current positioning, not a forecast.
Which is more accurate, the people version or the money version?
Both are correct calculations answering different questions. The people version shows how many accounts sit on the same side; the money version shows where the money is stacked. A wide gap between them means a few large accounts are driving the ratio.
Why remove the largest accounts from the money version?
To test how much one or two large accounts move the ratio. In BTC, removing one took 10.7% to 13.1%, and removing three took it to 19.0%. If removing them barely changes the ratio, the figure is broadly distributed.
Check it yourself on the Satoshi Index
Each coin page on the Satoshi Index (for example BTC) shows long and short account counts and sizes, with per-account positions below. It opens without login and refreshes every 30 minutes (every 3 hours for dYdX).
Data basis: position snapshot at 07:14 KST on November 4, 2026; accounts among the top 100 by 7-day return on each of Hyperliquid, Lighter, dYdX and Pacifica that hold a position in the asset. This content is market-data analysis, not investment advice.