Oct 28 FOMC: @CMEGroup FedWatch prices a 90% hold, with hike odds down from 68.6% to 10% in a month. → After the Sep 16 hike, $BTC rose 10.7% in a week vs Nasdaq 100 +5.3%, S&P 500 +2.0% and gold −1.6%. → Since Sep 24 the hike got mostly priced out, yet BTC and Nasdaq both fell ~5% and gold 6.7%. ★ A hold is priced in, so it may not move BTC much; a surprise hike is the real tail risk.

Bitcoin dominance is holding ~60.6%, a level last seen in late May. But $BTC is down 5% in 30 days. The share went up because everything else fell harder. ★ The chart (@tradingview, BTC.D) → 60.6% now, up from 59.4% on Sep 24. → Above its 50-day (59.8%) and 200-day (59.6%) averages. RSI is 67, close to the 70 line. → The 1-year ceiling is ~61%: a 61.4% wick on Nov 4, 2025 and a 61.1% close on May 5, 2026. ★ Where the share came from (last 30 days) → BTC −5.0%. Total crypto market cap −7.0%. Everything except BTC −9.8%. → ETH/BTC −8.9% to 0.0290, RSI 25. $ETH dominance is down to 10.7%. → USDT dominance +6.7% to 6.87%. Stablecoins don't fall in a selloff, so their share rises by default. ★ The last two times BTC.D reached ~61% → Nov 2025: BTC.D fell from 60.8% to 59.1% by Nov 21 as BTC dropped 16%. USDT + USDC dominance rose 1.2 pts. ETH/BTC was flat. → May 2026: BTC.D fell from 61.1% to 57.9% by Jun 3 as BTC dropped 21%. USDT + USDC rose 1.8 pts. ETH/BTC −3%. → Both times, stablecoins took more than half of the share BTC lost, and ETH's share fell too. → Caveat: TradingView counts stablecoins in the total. Without USDT and USDC, BTC's share is ~67%. @coingecko shows 60.1% because it tracks a different set of coins. And two tops aren't a pattern. BTC.D is back at the ceiling with RSI near 70. If it rolls over again, is that an alt season, or another move into stablecoins?

Banks found AI agents inside customer accounts. The interesting part isn't the agents. It's how they got in. > Per Axios, Morgan Stanley, BofA, JPMorgan and Wells Fargo spotted third-party agents in customer accounts, in some cases using logins the customers handed over themselves. > Some banks are blocking them. Others are restricting what they can do. > Jamie Dimon reportedly called it a "liability transfer" to the customer. That last line is the one I keep coming back to. An agent that needs your bank password is borrowing your identity. Crypto has been building a rail where it doesn't have to. x402, the @coinbase standard, lets an agent pay for an API call with stablecoins. No login involved. Ran the numbers on Agent Economy's tracker (public Dune data): • 194.8M payments settled • $42.2M in total volume • ~$0.22 per payment on average • 96.7K payments yesterday • 45% of all payments have settled on @base So the rail exists. The money on it doesn't, yet. ★ My read: if banks shut the password-sharing door, agent payments get pushed toward permissioned rails, meaning bank APIs or stablecoins. At ~$0.22 per payment, x402 is still a micropayment rail, not a money rail. Which one gets the agent volume first?

Memecoins are bleeding. The memecoin factory just had its best month since March 2025. ★ The coins (@coingecko) → The meme category is worth ~$31.0B, about 1.1% of the whole crypto market. → 7 of the 8 largest memes are down over 30 days: $DOGE −13.8%, $SHIB −11.9%, $PEPE −5.5%, $TRUMP −11.5%, $PENGU −4.5%, MemeCore −48.8%. → Most are far from their highs: DOGE −89%, SHIB −94%, PEPE −83%, TRUMP −97%. → The only one up is $PUMP, +15.0%. ★ The factory (@DefiLlama) → Pumpfun took in $62.6M in fees over the last 30 days, +37% on the 30 days before. That's its best 30-day stretch since March 2025. → Its launchpad earned $45.4M in that time. The next one, BONK(.)fun, earned $3.7M. That's ~89% of tracked launchpad revenue. → $30.8M of pump revenue went to PUMP buybacks in 30 days. Annualized, that's ~16% of PUMP's $2.34B market cap. → On DefiLlama's weekly comparison, pump(.)fun out-earned @HyperliquidX in 2 of the last 4 weeks. → Caveat: fees come from churn, and most new launches go to zero. Buybacks shrink when volume drops. And 54% of PUMP's supply isn't circulating yet (FDV $4.19B). The coins are down and the platform is printing. Is this a meme cycle, or just a busy casino?

Filecoin's genesis vesting just ended. The price did most of its move before that. ★ What changed on Oct 15 → Vesting for @protocollabs and @FilFoundation, set at the 2020 launch, is complete. Filecoin's blog says daily new filecoin:native supply drops ~75%. → @FilecoinTLDR's model puts current supply growth at +286K FIL/day: ~104M a year, or ~12.5% of the 835M circulating. → A ~75% cut takes that to ~72K/day, about 3% a year. Their base case reaches +49K/day in 2027. → The Solstice upgrade also routes more block rewards to providers with paying customers and burns the unearned share. ★ What the price did (@coingecko) → FIL is +44% in 30 days, vs +6.7% for $BTC. → Almost all of it came between Sep 17 and Sep 27: $0.81 → $1.13 (+41%). Since then it has ranged $0.99–$1.25 and sits at ~$1.14. → Perp open interest is $513M against a $953M market cap, about 54%. → Caveat: the supply figures are a model (base case, updated Sep 29), not realized data. And with OI at 54% of market cap, part of the move is likely leverage, which can unwind either way. The supply cut was scheduled six years ago. Was it priced in by Sep 27, or does it only start to matter once paying-customer demand shows up onchain?

In its first full month of mining, one company took ~14% of all ZEC miner rewards and added all of it to its treasury. ★ What @cypherpunk's update implies → Miners get 1.25 ZEC per block (80% of the 1.5625 subsidy). September had 34,405 blocks onchain (Blockchair) → ~43,000 ZEC to miners. → Cypherpunk mined 5,944 of them, about 14% of miner rewards. That roughly implies ~14% of network hashrate. → As of Sep 30 it held 332,362 $ZEC, 1.96% of supply. That's ~$408M at today's $1,228. ★ It's not the only one → @FortitudeCrypto, DCG's ZEC miner, said on Oct 15 it crossed 5.0 GSol/s. Against today's 32.55 GSol/s network, that's ~15%. → Together, these two companies could account for roughly 29% of ZEC hashrate. → At the pool level (MiningPoolStats, % of known hashrate): @ViaBTC 32.0%, @FoundryServices 16.6%, F2Pool 15.3%, Luxor 13.4%. The top 2 hold 48.6%, and the top 4 hold 77.3%. → Caveat: Cypherpunk's share is a September average implied from rewards, while Fortitude's is a live, self-reported number. Company hashrate sits inside these pools, not on top of them, and neither update says which pool it uses. Hashrate concentration is mostly a security and censorship question, not a privacy one. How few operators can ZEC rely on before shielded transactions depend on their goodwill?

Tokenized stocks just hit a record. Almost nobody is trading them. ★ Owning is up → Distributed value is $3.32B, up 13.5% in 30 days (@RWA_xyz). → Holders are at 4.39M, up 18.9%. ★ Trading is down → Monthly transfer volume is $10.06B, down 72% in 30 days. → Monthly active addresses fell 64% to 1.26M. → People are buying and holding. They aren't moving these tokens around. ★ Who is winning share → bStocks: $983M, +39.7%. It's now the largest platform. → @Ondo: $914M, +8.7%. → @xStocksFi: $595M, −1.6%. → @Securitize: $357M, +25.2%. It started onchain trading of US stocks on Oct 8, after the SEC's conditional exemption in September. → The biggest underlying names are crypto-adjacent. $CRCL leads with $330M, Securitize's own stock is next at $291M, and $MSTR adds ~$431M across two share classes. ★ The big money is still waiting → @MorganStanley pushed its own tokenized equity support back to Q1 2027. → Its head of digital asset strategy still says tokenized assets will reach $3–10T by 2030. → $3.3B against that range is still a rounding error. → Caveat: a 72% drop in volume after a busy month can partly be noise. Also, "distributed value" moves with the underlying stock prices, not only with new demand. Is a holder base without trading the start of a real market, or just a wrapper people buy once and forget?

This reset didn't stay in AI stocks. AI tokens sold off with them. ★ What the table shows (@coingecko AI category, red = down) → 11 of the 12 largest AI tokens are down over 7 days. → $VVV −21.3%, $VIRTUAL −18.8%, $FET −15.1%, $TAO −9.3%, $ICP −8.7%. → @NEARProtocol held up over the week (−2.9%), then dropped 9.7% in the last 24 hours. → The only green name is Pearl, up 46%. → The whole category is at ~$24.8B, down about 5% on the day. ★ Why the timing matters → The @OpenAI number came out on Oct 8–9, and AI-linked stocks sold off. → The tokens that fell the most are the ones that rallied in September on the AI story. None of them reported weaker usage of their own this week. → So the market priced AI tokens off the AI equity story, not off anything the networks themselves did. ★ What it means for the "real revenue" argument → Last week I wrote that Bittensor (@opentensor) subnet revenue covers ~7–9% of its emissions and Venice (@AskVenice) burns offset ~18%. → Revenue at that size doesn't stop a token from trading like a sector bet. It only matters once it's big enough to set the price. → Caveat: one week of moving together doesn't prove a lasting link. Crypto as a whole was weak too, with $BTC down 3.2% last week. Which AI token breaks away from AI stocks first, and what would the data need to show for that?

Old coins are beating Bitcoin this month. Over a year, almost all of them are still deep in the red. ★ The 30-day picture → $BTC is up 7.5% in 30 days. 11 of the 14 old large caps I track did better. → filecoin:native +40.0%, algorand:native +29.8%, hedera-hashgraph:native +23.3%, bitcoin-cash:native +22.4%, litecoin:native +19.9%, $ADA +19.2%. → The laggards are $XRP (+3.2%), $DOGE (+1.9%) and $TRX (−2.4%). ★ The 12-month picture → Every one of them except TRX is down 34% to 62% over the year. BTC is down 25%. → $ADA is −62% and $DOT is −61%. Most sit 85–98% below their all-time highs. → So this is a bounce from very low levels. It is not a new high for anyone. ★ What's behind the moves → @litecoin already has a spot ETF trading (@CanaryFunds' LTCC), listed under the generic standards. → @Cardano, @Polkadot and @hedera ETF proposals are still stuck in the SEC's staking review, which opened on June 30. There is no confirmed decision date. → @Filecoin has the clearest catalyst. Protocol Labs and Foundation vesting is reported to end on Oct 15, which would sharply cut new FIL issuance. → For most of the others I couldn't find coin-specific news. They moved together, which looks like rotation. ★ This week the bounce is being tested → Over 7 days: BCH −12.8%, HBAR −10.4%, XLM −9.9%, LTC −9.8%, ALGO −8.7%. → FIL (+3.5%), DOT (+4.1%) and ATOM (+10.4%) are still holding up. What I'm watching next: → Oct 15: does FIL hold its gains once the vesting change is in the price? → Any SEC signal on staking ETFs for ADA, DOT and HBAR. → Whether the old coins keep beating BTC if BTC dominance turns back up. → Caveat: a 30-day outperformance after a −40% to −60% year is what a relief rally looks like too. One month of data can't tell the two apart. Which of these has real usage behind it, and which is only moving with the group?

Update on this. Both levels got tested this week. ★ S1 at $1,272 gave way. $ZEC flushed to $1,112.77 on Oct 8, just above the S3 zone ($1,085–1,105). → It has since bounced ~10% to ~$1,225 and is sitting right on S2 at $1,222. → The bounce also pushed price above the downtrend line from the $1,698 top. That line now sits near ~$1,180. ★ The fundamentals calendar hasn't moved. → NU7 mainnet height decision: Oct 20. Mainnet target: Nov 5. → Shielded pools hold ~4.94M ZEC, about 30% of supply. → Caveat: after the Ironwood pool migration, shielded totals aren't directly comparable with the 5.1M "records" from earlier this year. The structure is simple now. If $1,222 holds and $1,272 is reclaimed, the breakdown gets repaired. If $1,222 fails, $1,100 comes back into play. Will price get ahead of the Oct 20 decision, or wait for it?