Home › Market › Daily brief

Daily Brief — September 12, 2026: the contracts came off

By TradingPrimer · Published 12 Sep 2026

Yesterday's brief left an open question: contracts had been added into a 3.5% decline, and open interest can never say which side of them was right. Twenty-four hours later the question has an answer, and it is the least dramatic one available. The positions were closed. Nothing had to be liquidated to close them.

Where the market actually is

Read at 03:53 UTC on 12 September, from the Binance spot and perpetual APIs:

All three are green, and all three finished a long way below where they got to. Measured against each asset's own opening price on the 11 September session, Bitcoin travelled 5.02% between its high and its low, Ether 9.52% and Solana 7.91%. Bitcoin reached $79,890 and gave back about 69% of the climb from its low before the session closed; Ether gave back about 64%, Solana about 43%.

Hold those range figures against the recent record. Yesterday's brief reported 2.68% for Bitcoin, Sunday's reported 0.95%. The range has roughly doubled in each of the last two sessions, and it has now doubled in both directions rather than only down. Turnover over the 24 hours was about $1.45 billion on BTC/USDT, $1.58 billion on ETH/USDT and $325 million on SOL/USDT — the unusual detail there being that Ether out-traded Bitcoin in dollar terms.

Why it moved: a meeting nobody can price

The August CPI print landed on 11 September, days before the FOMC decision on 16 September, and it did not settle the argument it was supposed to settle. Reporting through the session described the data as neither confirming nor easing the case for a rate increase, leaving the decision resting more on the committee's own judgement than on the number.

What makes this week worth watching is not the odds but the disagreement between them. On 11 September, CME's FedWatch tool put the probability of a 25 basis point hike at roughly 69%. On the same question at roughly the same time, Kalshi traders were near 57% and Polymarket near 49%. Three venues, one binary outcome, twenty points apart. That gap is a useful thing to see with your own eyes: probabilities quoted to a decimal place can still be an expression of how little anybody knows, and a market coin-flip is a genuine coin-flip no matter how confidently any single dashboard states it.

Underneath sits the inflation input that has driven the quarter. Brent crude was up more than 16% over the previous month and about 8.4% over five sessions, with the war in Iran pushing diesel to record highs. Energy feeds into almost every other price with a lag, which is why an oil chart has been setting the tone for an asset class that has nothing to do with oil.

The number most people skipped

Binance Bitcoin perpetual open interest at the time of reading was about 103,450 BTC. Yesterday it was roughly 107,031 BTC. That is a fall of about 3,581 contracts-worth of Bitcoin, or 3.3%, over 24 hours in which the price went up 0.57%.

Now apply yesterday's rule. Priced in dollars, that open interest is worth about $7.99 billion against roughly $8.22 billion yesterday — a fall of 2.8%. Coins down, dollars down. Both units agree, which is exactly the case yesterday's brief said to treat as real: when the coin count and the dollar figure point the same way, positions were genuinely opened or closed rather than revalued by the price. Yesterday they pointed opposite ways and the dollar figure was lying to you. Today it is not.

The second question is how they came off, and funding answers it. The Bitcoin rate settled at +0.003006%, +0.006122%, +0.003593% and +0.006093% across the four eight-hour intervals spanning that move, and was accruing at +0.005237% when read. The top of that band annualises to roughly +6.7% a year for longs. Through a 5% range, a $3,843 rally and a two-thirds giveback, the cost of being long a perpetual did not visibly change.

Ether went the other way on every reading. Its open interest rose from about 2,314,811 ETH to 2,322,725 ETH — up just 0.34% in coins but 3.6% in dollars, the split that means price did most of the work again. Yet Ether funding settled at +0.000055% per interval, about +0.06% a year, and was slightly below zero when read. Ether outperformed Bitcoin fivefold on the day while leveraged longs paid essentially nothing to hold it, which places the buying somewhere other than the perpetual book. For context on the mechanics, open interest, funding rate and liquidation each explain the term from zero, and live readings sit on Market Pulse.

One reading cuts against the tidy story, and it belongs here for that reason. Binance's global long/short account ratio on BTC/USDT was 1.60 at the last hourly print, meaning about 62% of accounts with a position were long, and that share rose from 1.55 across the previous three hours. Fewer contracts, but the accounts that remain are more one-sided than they were at midnight.

So what

The lesson that survives this week is about volatility, not about the Fed. A market that moved 0.95% on a normal day six sessions ago moved 5.02% on Thursday, and 9.52% in Ether. Nothing about a position's size changes when that happens — but everything about its risk does. A stop placed 2% away was a considered decision in a 1% market and is noise-bait in a 5% one, and the position that was 1R of risk on Sunday is closer to 5R of exposure to the same daily wobble today. This is the least glamorous and most reliable use of a range figure: it is an instruction to recalculate, not a hint about direction. Position sizing works through the arithmetic, and leverage and margin covers why a multiplier applied to a widened range is what actually ends accounts.

The second point is what the open interest figure does and does not license you to think. It does not say the decline is over, and it does not say a rally is coming; every contract that was closed had a buyer and a seller, and the number counts both. What it changes is mechanical. Less open leveraged position means less of it sits within reach of a forced close, and forced closes are what turn an ordinary move into a violent one — the subject of anatomy of a liquidation cascade. Going into a decision that three separate venues price twenty points apart, a market carrying 3.3% less leverage is a marginally less fragile one. Marginally. It is not a reason to carry more.

The practical version has not changed since yesterday and will not change on Wednesday: the decision that matters is not what the Fed does, it is what your position does if you are wrong about it. When the professionals quoting probabilities cannot agree within twenty points, the honest confidence to assign your own view is low, and size should say so. The event calendar has the FOMC time in your local zone.

← Market analysis