cryptoeducation

Reading a funding rate takes five steps: find it, annualize it, compare it to the year, pair it with price, count the streak. Today's $BTC print, -0.0002% per 8h at the 08:00 UTC settlement on September 5th, 2026 (Friday), looks dramatic on step one and turns into a footnote by step five, because it is the only negative print in the last 90 settlements.
## Before you start: 3 terms
Perpetual: a futures contract with no expiry, so nothing forces it back to spot. Funding: the fee longs and shorts pay each other every settlement to pull the perpetual back toward spot. Positive means longs pay shorts, negative means shorts pay longs. Open interest: the size of all open positions, a different number that funding does not tell you.
## Step 1: Find the number and its interval
On the Binance Futures screen for BTCUSDT the field is "Funding / Countdown". BTCUSDT settles at 00:00, 08:00 and 16:00 UTC. As of 15:37 UTC the estimate for the 16:00 settlement was 0.0019%, while $ETH showed 0.0021%. Common mistake: reading the estimate as the paid rate. The paid rate is the one at the countdown's end, and today it moved from 0.0010% at 00:00 to -0.0002% at 08:00.
## Step 2: Annualize it
Multiply the 8h rate by three settlements a day and 365 days. 0.0066%, the 7-day average, becomes 7.2% a year. The 30-day average, 0.0068%, becomes 7.5%. The default 0.01% per 8h is 11% a year. Common mistake: comparing 8h rates across pairs with different intervals. Many smaller perpetuals settle every 4 hours, so 0.01% there is 22% a year.
## Step 3: Compare it to the year, not to zero
Over the last 365 days the BTCUSDT average was 0.0031% per 8h, 3.4% annualized. A daily average above 0.0079% was in the top 10%, below -0.0028% in the bottom 10%, and 76% of all 1,095 settlements were positive. The 7-day average of 0.0066% is not extreme, but it is twice the yearly average. Common mistake: treating 0.01% as a sign of greed. That is the rate the formula returns when the perpetual sits exactly on spot, and 7.9% of the year's settlements printed exactly that.
## Step 4: Pair it with price and a level
Funding only means something next to what price is doing. Rising funding while price stalls under resistance is the crowd paying to lean into a wall. September 3rd was the textbook version: 0.0089% at 81,317, and no daily close above 82,300. July 4th, 2026 had a 0.0098% average with $BTC at 63,115, and a month later price was up 0.6%. Same rate, opposite context. Common mistake: calling a top from the rate alone. Of the 11 days in the past year with funding at or above 10% annualized, the six old enough to measure had a median 30-day change of -12.9%, and that is a sample of six.
## Step 5: Count the streak, not the print
One settlement is noise. Eleven straight negative days ending April 21st, 2026 were a message, and price was 4.1% higher 30 days after the deepest of them. Today's single negative print, after 89 positive ones, is a note that the premium left after the 82,300 rejection. It is not a regime change until it repeats.
## The 3 numbers I check every time
1. Latest paid rate versus the 7-day average: -0.0002% versus 0.0066% today.
2. The 7-day average versus the year's top 10% line: 0.0066% versus 0.0079%. Close, not there.
3. Negative settlements in the last 30 days: 1 of 90. A cluster of five or more would change my read.
If this week's average climbs back over 0.0079% with price still under 82,300, what would you do differently?
## FAQ
**What is funding rate in crypto?** The periodic payment between longs and shorts in a perpetual futures contract, designed to keep the contract's price near spot. On Binance BTCUSDT it settles every 8 hours.
**Perpetual futures funding rate positive: who pays?** Longs pay shorts when the rate is positive, shorts pay longs when it is negative. The payment is the rate times the position size, so a 10,000 USDT position pays 1 USDT at 0.01%.
**What does a negative funding rate mean?** The perpetual trades below spot and shorts pay to stay short. In the past year it happened on 76 of 366 days, mostly from February to April 2026.
**What is funding rate arbitrage in crypto?** Holding spot and shorting the same amount in the perpetual to collect positive funding. At 7.2% annualized the carry exists, but the rate can flip, as it did today.
#Bitcoin #CryptoEducation

Un hachage est une empreinte de longueur fixe calculée à partir de données. Entrez les mêmes données et vous obtenez toujours le même hachage ; modifiez un seul caractère et le hachage change entièrement, au point de devenir méconnaissable.
Les blockchains s’appuient sur cela partout. Chaque bloc contient le hachage du précédent, de sorte que modifier d’anciennes données change leur empreinte et casse instantanément tous les liens qui suivent. Votre propre transfert a aussi un hachage, et cette chaîne est ce qui permet de le retrouver sur un explorateur de blockchain. Ce n’est pas un numéro de référence qu’on vous a attribué - il est calculé à partir de la transaction elle-même, ce qui explique pourquoi il ne peut pas être falsifié.

Block height is simply how many blocks have been added since the very first one. It is the closest thing a blockchain has to a clock - not a time of day, but a count of how far the record has advanced.
It matters practically because confirmations are measured in blocks, not minutes. When a service says it needs a set number of confirmations, it is waiting for the height to grow by that much on top of your transfer. On a fast network that is minutes; on a slower one the same number of confirmations takes far longer. Same requirement, very different wait.
Binance Academy covers this in more depth: https://www.binance.com/en/academy/glossary/block-height
Trading crypto from Dubai since 2019.
$BTC $ETH
#Crypto #CryptoEducation #Binance

Les transactions ne sont pas écrites individuellement dans une blockchain. Elles attendent, sont regroupées dans un bloc, puis le bloc est ajouté à la chaîne comme une seule unité.
C’est pourquoi les transferts ne sont pas finalisés instantanément lorsque vous appuyez sur envoyer, et pourquoi ils ont tendance à être finalisés par grappes plutôt que de manière fluide. Cela explique aussi pourquoi un réseau très sollicité semble lent : un bloc a une limite de taille, donc lorsque davantage de transactions attendent que celles qui peuvent y tenir, certaines attendent le suivant. Rien n’est cassé quand cela se produit. Vous êtes dans une file d’attente, et la file d’attente fait partie de la conception.