Bitcoin and Ether Passed. Gold Didn't.

Last Friday I wrote down three things that would move me from "relief" to "recovery", and said I'd review them after the September 25 close, either way (September 19). Here they are.

The test: two ETF sessions in a row with net inflows, and a CoinGecko daily reading above $80,000 on both days.

Last Friday brought in $433 million, with bitcoin's daily reading at $81,124. Monday brought in $999 million, the biggest day of the year on Farside's numbers (Blockhead), and the reading was $86,421. Passed on day two.

That took bitcoin straight into the band I flagged on September 6, older supply at $83,000-86,000 (September 6). It went through the top, peaked near $87,200 early Wednesday, and slid back. Friday's reading was $84,069. Back inside.

The money kept coming, just less of it each day: $999 million on Monday, $191 million by Thursday. That's six straight inflow sessions starting September 17, about $2.8 billion through Thursday (Cointelegraph). Farside's Friday numbers aren't complete yet.

This is the band I said might supply coins into a rally. My guess is some older holders cashed out into this one, but price and ETF flows can't actually show me who did. A daily reading back above $86,000, with inflows still coming, would say demand is getting through. Below $83,000 as the flows fade would say it isn't.

The test: three inflow sessions in a row, with a daily reading above $2,650.

Monday brought in $270 million (KuCoin), Tuesday more, Wednesday $104.5 million (crypto.news). Ether's daily readings on those three days: $2,778, $2,753 and $2,689. Passed.

Same caveat as bitcoin, a little sharper. Ether peaked near $2,780 on Monday afternoon and has sat around $2,690 since, while the daily inflows got smaller. It passed the test I set. That isn't the same thing as a trend.

The test: a Friday 4pm New York reading above $4,440, with the ten-year yield between 4.90% and 5.10%. Neither half made it. Gold spent Friday around $4,280, about 2% lower on the week (Reuters). The ten-year touched about 5.22% during Friday's session, its highest since 2007 (Reuters).

I also wrote that gold back below $4,300 would weaken my read. It's below $4,300. Weakened.

The read was this: gold and bitcoin rising together while the ten-year sat near 5% might mean some money wants dollars without taking on long-dated Treasury risk. This week the two split. Bitcoin held up while ETF inflows had their biggest day of the year. Gold fell as real yields climbed: the ten-year TIPS yield went from 2.68% last Friday to 2.85% on Thursday (FRED). The boring rate story fits this week better than my more interesting one.

I hold GLD. That doesn't change the grade.

The frozen September 17-30 window asked for two things: no new supply shock, and yields and credit that stop getting worse.

Oil is the murky half. It jumped on Thursday, then gave most of it back on Friday. Brent settled at $104.32, close to last Friday's $103.87, as traders weighed reports of phased Hormuz talks and a possible US diesel-export ban (Reuters). A jump isn't proof of a new supply shock, and I'm not calling that half yet.

Yields and credit are the clear half, and they got worse. The ten-year went from 5.01% last Friday to 5.18% on Thursday (FRED), and high-yield spreads widened from 2.68% to 2.80% (FRED). Stocks rose anyway. The S&P 500 finished the week up about 1.2%.

Three sessions left. As it stands, that isn't stabilization. The window gets graded on October 1 as written, alongside the August windows, and the next dates go out with it.

Disclosure: I hold long positions in ether (since 2025) and in GLD. No bitcoin position.

Market data: crypto prices are CoinGecko UTC-day readings (the last hourly sample of each UTC day), not exchange closes. ETF flows are Farside Investors figures as reported by the linked outlets. Yields and spreads are FRED daily series (DGS10, DFII10, BAMLH0A0HYM2), whose latest point is Thursday.

Public forecast record and personal judgment, not personalized investment advice or an instruction to transact. Past samples and historical observations do not guarantee future performance.

This Free website mirror preserves the Perchery article dated 2026-09-25. Original content ID: perchery-2026-09-25-bitcoin-ether-passed-gold-didnt. Section headings label the original movement; the first-person interpretation remains Perchery's. This import does not convert the source into a Quant-native empirical audit.

Substack: https://perchery.substack.com/p/bitcoin-and-ether-passed-gold-didnt

X: https://x.com/readperchery/status/2103672167942721769

Bluesky: https://bsky.app/profile/perchery.bsky.social/post/3mwfadqwxws2m

Information cutoff: September 25, 2026, US close, with FRED series through September 24; crypto figures use CoinGecko UTC-day samples, with other market-data cutoffs retained in the source prose. The October 1 grading of the frozen September 17-30 window and the August windows remains open in this dated record. Inline link markup is adapted to plain text; original source URLs follow in order of appearance.

Related website mirror (the September 19 letter that set the three markers): https://quant.zylotechnology.com/research/perchery-2026-09-19-obvious-trade-didnt-pay

https://perchery.substack.com/p/the-obvious-trade-didnt-pay

https://www.blockhead.co/2026/09/23/bitcoin-etfs-pull-in-999-million-in-a-day-as-btc-holds-above-86-000/

https://perchery.substack.com/p/the-dollar-stays-central-capital

https://cointelegraph.com/markets/bitcoin-etf-six-day-streak-2-8-billion

https://www.kucoin.com/news/flash/ethereum-etfs-record-270m-inflows-on-september-21-led-by-blackrock-and-fidelity

https://crypto.news/will-ethereum-rally-as-etfs-post-5-days-of-inflows/

https://www.marketscreener.com/news/mounting-rate-hike-bets-keep-weekly-loss-in-sight-for-gold-ce785adfde8aff23

https://ca.marketscreener.com/news/bond-yields-hit-multi-decade-highs-despite-oil-pullback-ce785adfde81f023

https://fred.stlouisfed.org/series/DFII10

https://www.marketscreener.com/news/oil-prices-slide-about-2-as-us-iran-explore-path-out-of-war-ce785adfd180f425

https://fred.stlouisfed.org/series/DGS10

https://fred.stlouisfed.org/series/BAMLH0A0HYM2

Editorial research and personal judgment, not personalized investment advice or an instruction to transact.