Bitcoin 2026-09-29 daily brief — AI stance: bearish

Open-interest-weighted funding across venues annualizes to about 8.1%/yr, under the roughly 11%/yr exchange-default neutral rate, so longs are paying a below-baseline carry: neutral, not crowded. The single major venue reads 4.9%/yr, about 3.2 percentage points under the aggregate, so what mild long premium exists sits on other exchanges, not there. That venue's 14-day range ran from -0.0012% to +0.0095% per 8h, flipping briefly negative on 2026-09-18 and 09-25 and never hot. This matches the news read that perpetual positioning is close to neutral after the September flush.

Market-wide futures open interest is 52.3B, down about 15% from 61.6B on 2026-09-21 and the lowest of the 14-day window, while price fell only 3.4% from the high: deleveraging, not liquidation-driven distribution, and the news read of coin-denominated OI at its lowest since March agrees. The single venue's 6.08B is down about 9% from 6.70B on 09-22, the same direction. Options OI of 35.8B reflects the 2026-09-26 quarterly expiry (52.2B to 36.0B). The 30-day skew of -0.57% leaves calls marginally bid at the 33rd percentile while the 7-day skew is slightly put-bid at the 70th percentile, so hedging is concentrated in the event window. Clean leverage lowers crash odds; it does not create an upside impulse absent a buyer.

Two-sided and small. 2026-09-29 saw 24.5M longs and 14.7M shorts liquidated, ratio 0.6, total at the 26th 1y percentile. The big prints were 391.6M shorts on 2026-09-21 (the 6.5% up day to 86,597) and 178.3M shorts on 09-18 (the 5.7% up day). Since then longs have led every session except 09-26 and 09-27, with 88.3M on 09-23 and 65.2M on 09-28. The squeeze fuel above was spent on 09-21; what remains is late longs being trimmed on the way down.

Coinbase premium is -5.18 bps, inside the +/-10 bps practical extreme, so labelled neutral, but the trend matters. It printed -7.97, -8.01 and -9.46 bps on 2026-09-16 to 09-18 during the pre-rally chop, then held a -2.8 to +1.2 bps band from 09-19 to 09-28 while ETFs absorbed roughly 28,000 BTC, and now shows the widest negative print since 09-18 as flows decayed. Offshore is leading again and US spot is not stepping in at 83,600.

CONFLICT. Trend, cleaned leverage (all-venue OI 52.3B, down 15% from 61.6B), funding under baseline and a NEUTRAL cycle read are constructive locally. Rising yields, a rising dollar, a hiking Fed and fading ETF absorption push the other way. I side with the macro tape because it is the factor that actually stopped the rally at 87,000, and it has not reversed.

ETF flow is the dominant micro story and it is decaying: 11,536 BTC on 2026-09-21, 8,293 on 09-22, 4,111 on 09-23, 1,600 on 09-25, 371 on 09-28, 181 on 09-29. The 7-day average of 4,050 BTC still sits at the 86th historical percentile, but the 30-day z-score is -0.52 and today's absorption was 316 BTC below miner issuance. Long-term-holder SOPR of 1.13 and the reported 84,000-85,000 long-term-holder cost cluster say older coins are being sold into this level. Strategy added 1,665 BTC at an average 85,681 between 09-21 and 09-27. Miner economics are neutral: Puell 1.16, hash ribbons 1.009 and rising. Transaction count 740,527 at the 99th percentile with fees at the 11th percentile is activity without fee pressure. The Bitget breach of 2026-09-24 (387.5M) did not touch BTC withdrawals materially. No supply unlock or regulatory catalyst falls inside the window; the CFTC crypto-margin FAQ of 2026-09-24 is a slow-burn positive. Cycle position is mid-range: MVRV-Z 0.98 at the 41st percentile.

hawkish. The supplied effective fed funds reading is 3.63%, while the news digest reports the FOMC raised the target range to 3.75-4.00% on 2026-09-16, the first hike since 2023 on a 12-0 vote, so the supplied figure lags the announced range. The 10y at 5.17% is up 23 bps from 4.94% on 2026-09-21. M2 is still growing 5.66% YoY, a partial liquidity offset that has not stopped front-end repricing: Kalshi and Polymarket price a second 25 bp hike at the 2026-10-27/28 FOMC near 69% and CME FedWatch near 77%. PCE, the preferred inflation gauge, prints 2026-09-30 with Michigan one-year inflation expectations already at 4.6%. Fear and Greed at 73 (Greed) says positioning sentiment has not yet adjusted to a tightening Fed.

10y at 5.17%, having climbed from 4.94% over 2026-09-19 to 09-21 to 5.18% over 2026-09-26 to 09-28 and holding there. Direction is up and the level is the window high. No credit-spread feed is supplied, so no spread read is offered.

DXY 101.34, up from 100.21 on 2026-09-18 to a 14-day high, a 1.1% rise over two weeks. A firming dollar alongside rising nominal yields is the classic headwind for BTC after a 43% quarter; the slide from above 87,000 on 2026-09-28 coincided with this DXY and yield push, per the CoinDesk live coverage.

S&P 500 at 7,670.8, 1.2% below its 7,764.7 close of 2026-09-21 and lower in four of the last five sessions, with VIX at 16.0. Not risk-off, but risk-on is fading: Conference Board confidence fell 6.7 points to 81.9 and Michigan sentiment finished September at 48.1, a four-month low. Stagflation-flavoured tape, and BTC has traded with equities rather than with gold at 4,213.8.

First shelf 83,000, the ATM strike and the 2026-09-28 close of 83,479; under one daily sigma away, so a test this week is more likely than not. Second shelf 80,900-81,200, the 2026-09-18 to 09-20 closes and the one-week straddle low; roughly a one-sigma week, so call it 40-50% touch odds under the straddle pricing. Below that, 79,600 (nine-day straddle low) and 78,400 (sixteen-day straddle low). The 50-day at 76,870 is 8.1% below spot and beyond the sixteen-day one-sigma range, so a gate test inside this horizon is a tail case that would need a macro shock such as a hot PCE plus a weak payrolls print. Base case: 81,000-83,000 tested this week, new highs above 86,597 less likely than a retest of 81,000.

low. DVOL 35.1 sits at the 5th percentile of the past year; 30d ATM IV 34.2% is at the 36th percentile of the past 90 days and fell 0.5 vol points over five days. Realized vol is 17.2% over 7 days against 41.4% over 30 days and 38.5% over 90, a compression that usually gives way to expansion. The options market prices +/-1.4% (81,835-84,165) by 2026-10-01 08:00 UTC and +/-2.5% (80,920-85,080) by 2026-10-03 08:00 UTC, the first expiry after the 2026-10-02 payrolls release; +/-4.1% (79,588-86,412) by 2026-10-09 and +/-5.6% (78,361-87,639) by 2026-10-16. These say how far, not which way.