POSITIONING
Which way the crowd is leaning
How many accounts are betting on a rise against a fall, and how much of the day's trading hit the buy side.
Long against short across 3 exchanges
Recorded on all 3 exchanges. This reading has been complete since Jul 6, 2026; the 4 earlier days are not published, because a total that climbs as exchanges arrive is not a market move.
Latest displayed reading: 1.16 on Sep 7, 2026. This view shows 64 dated readings from Jul 6, 2026 to Sep 7, 2026. The latest displayed reading is lower than the first. The displayed low is 1.06 on Sep 4, 2026. The displayed high is 1.69 on Aug 16, 2026. Every day in this range has a reading.
Pick two to four dates and read this chart's values side by side.
Where does it stand?
1.20accounts betting up for each betting downOn 6 of the 11 exchanges we cover, as of Sep 7, 2026.48.1%of trading was buying at marketOn 3 of the 11 exchanges we cover, as of Sep 7, 2026.$7.18 billionbought at marketOn 3 of the 11 exchanges we cover, as of Sep 7, 2026.$7.73 billionsold at marketOn 3 of the 11 exchanges we cover, as of Sep 7, 2026.Is the crowd betting up or down?
More accounts betting up, 1.20 to 1This counts accounts rather than money, so one very large trader weighs the same as one very small one. Read beside funding, which is weighted by money, the two often point different ways, and that disagreement is the interesting part. Neither is a forecast: a crowded side is not a side that is wrong.
Recorded on 6 of the 11 exchanges we cover; the rest publish no reading of this kind. This reading has been complete since Sep 6, 2026; the 66 earlier days are not published, because a total that climbs as exchanges arrive is not a market move.
Which side would not wait for its price?
51.9% of it was selling at marketTrading at market means taking whatever price is on offer rather than waiting at a price of your own. The side that will not wait is the side paying for speed, which is the closest thing here to a measure of urgency. On this day that was $7.18 billion bought against $7.73 billion sold.
Recorded on 3 of the 11 exchanges we cover; the rest publish no reading of this kind. This reading has been complete since Sep 3, 2026; the 63 earlier days are not published, because a total that climbs as exchanges arrive is not a market move.
What does this cover?
11 exchanges: Binance, Bitget, BitMEX, Bybit, Coinbase International, Gate, HTX, Hyperliquid, Kraken Futures, MEXC and OKX.
Recorded on 6 of the 11 exchanges we cover; the rest publish no reading of this kind. This reading has been complete since Sep 6, 2026; the 66 earlier days are not published, because a total that climbs as exchanges arrive is not a market move.
3 exchanges: Binance, Bybit and OKX.
Recorded on all 3 exchanges. This reading has been complete since Jul 6, 2026; the 4 earlier days are not published, because a total that climbs as exchanges arrive is not a market move.
Two views of the same number run side by side on this page and are never joined into one line. The wider one covers every exchange here and is the more complete picture; it is also the younger, because it only begins on the first day all of them reported. The narrower one is three exchanges with months of history behind it. A single line spliced from both would draw a step upward on the morning the other exchanges arrived and it would look exactly like a market move.
Understanding Bitcoin Futures Positioning
Where it stands today
There are 1.20 accounts betting on a rise for every one betting on a fall as of .
48.1% of futures traded at market was buying as of .
$7.18 billion was bought at market as of .
The long-short ratio counts accounts, not money. One account with a very large position weighs the same as one with a very small one, so the ratio describes the crowd rather than the capital. Read beside funding, which is weighted by money, the two often disagree, and the disagreement is the interesting part.
The share of trading that was buying at market is a different measure of the same question. It says which side was unwilling to wait for its price during the day, rather than which side is sitting in a position at the end of it.
Neither is a forecast. A crowded side is not a side that is wrong, and the cases where a lopsided reading preceded a sharp move in the other direction are the ones people remember rather than the ones that are typical.
Two views of the same number run side by side on this page and are never joined into one line. The wider one covers every exchange here and is the more complete picture; it is also the younger, because it only begins on the first day all of them reported. The narrower one is three exchanges with months of history behind it. A single line spliced from both would draw a step upward on the morning the other exchanges arrived and it would look exactly like a market move.