Everyone in my feed called it a starting gun. It isn't. It's a receipt.
Glassnode's Altcoin Cycle Signal printed a fresh altseason trigger this week, and the timeline did what the timeline always does — screenshot, quote-tweet, "here we go." Then you read the actual condition set, and the part that matters isn't the signal. It's the lag. The trigger requires a month of combined BTC-and-alt market-cap growth before it fires. A month. That is not a forecast. That is an after-action report on money that has already moved.
Underneath it sits one more number, and it's the one nobody screenshotted: Bitcoin dominance has stalled below 60%. Not collapsed. Stalled. That distinction is the entire trade.
I've spent enough cycles on the wrong side of momentum indicators to know the difference between a signal that describes and a signal that predicts. This one describes. So the only useful questions are two. What exactly is it describing? And is the thing it's describing even real?
The Indicator, and the Ledger Underneath It
Glassnode doesn't issue a token. It doesn't run a chain. It sells a lens. That matters more than people admit, because when a data vendor publishes a cycle signal, they're not making a market call — they're making a measurement call. The market then treats the measurement as a call. Those two things get conflated constantly, and the conflation is where retail money goes to die.
The Altcoin Cycle Signal sits in a crowded category. Blockchain Center's Altcoin Season Index, the "top 100 outperforming BTC over 90 days" dashboards, the dominance widgets on every aggregator — they're all proxies for one underlying question: is capital rotating out of Bitcoin's relative safety into the long tail?
Here's the mechanical part most people get wrong. Bitcoin dominance is a ratio, not a price. BTC market cap divided by total market cap. It means BTC can rip to a new all-time high in dollar terms and dominance can still fall, if the alts rip harder. It also means the reverse: BTC chops sideways, alts get liquidated harder, dominance climbs, and everyone screams "altseason is dead" when nothing structural changed at all. Dominance is an index of relative marginal flow. Read it as a level and you'll be wrong half the time. Read it as a velocity and you'll be wrong less.
"Stalled below 60%" is therefore an interesting phrase and an imprecise one. It tells us BTC's relative advantage stopped widening. It does not tell us alts are being bought. It tells us alts are not being sold harder than BTC. Those are different states of the world, and only one of them pays.
The history is instructive here, if only because it repeats with better branding. In 2017, dominance fell because ICO capital genuinely left BTC — new money, raised in ETH, deployed into tokens that mostly didn't survive. In 2021, dominance fell because DeFi and NFT flows created a parallel economy with its own collateral and its own leverage. In 2024, dominance barely fell at all; it stayed high because the marginal buyer was an ETF allocator with no on-chain wallet and no interest in the tail. Three "altseasons," three completely different funding mechanisms. If your indicator can't distinguish them, it's not measuring a cycle. It's measuring an average.
What Actually Moved in That Month
The signal fired after a month of market-cap growth across BTC and the alt complex. So the honest first question is where that growth came from.
Three candidates. Rotation — existing capital leaving BTC for alts. New money — stablecoins minted and deployed. Or accounting — the number went up because the denominator got fatter, not because anyone bought anything.
I ran this exact diagnostic in mid-2020, when I was running a delta-neutral book on Compound and Uniswap. Three hundred grand of capital, stablecoins borrowed against ETH collateral, farmed into the high-APY pools, price exposure hedged with futures. The trade worked for one reason: I could separate "yield is high" from "yield is real." When the COMP inflation model broke that summer, half the market was still staring at the APY number. The other half had already left. I was in the second group — out in 48 hours, 22% up while the tape bled. The lesson was never about Compound. The lesson was that displayed returns and realized returns are different objects, and the gap between them is where the money lives.
Now apply that to market cap. Displayed alt market cap is a genuinely bad measurement this cycle, and this is the part I'd underline twice. Aggregate alt market cap double-counts. Liquid staking tokens — stETH, wstETH, the whole LRT complex — appear both as the derivative token and as the underlying staked ETH sitting in a contract. Wrapped BTC appears as an ERC-20 on Ethereum while also counting inside BTC's cap, depending on which aggregator you ask. LP tokens, receipt tokens, vault shares, restaked positions — the same dollar wearing three hats and voting three times.
So when a composite indicator reports that alt market cap grew for a month, a meaningful slice of that growth can be re-wrapping rather than capital entering. I'm not claiming it's the whole story. I'm claiming the methodology isn't public, the double-count adjustment isn't disclosed, and therefore the magnitude of the move has an error bar nobody has published. A signal built on a denominator you can't audit is a signal you can't size.
That's not a knock on Glassnode specifically. It's a knock on how the market consumes composite indicators — as if a vendor's proprietary weighting were a law of nature instead of a product decision.
The Flows That Actually Tell You Something
If you want to know whether an altseason is real, stop reading dominance and start reading the plumbing. Four series, in order of how much I trust them.
Stablecoin net issuance. The least gameable number in crypto. When USDT and USDC supply expands, dollars arrived. When it contracts, dollars left. During the month preceding any altseason trigger, this series decides whether the market-cap growth was funded or fabricated. If supply was flat and market cap rose, you're watching rotation and leverage — a repricing of existing chips, not an influx. If supply expanded meaningfully, there's fuel. Nobody screenshots the stablecoin chart because it's boring. It's boring because it's honest.
Perpetual funding across the alt complex. Funding is the price of leverage. When funding across mid-caps goes persistently positive, longs are paying to hold. That can be the first innings of a trend or the last — you need open interest to tell which. Rising OI plus rising funding plus flat price is the signature of a trap. Rising OI plus rising funding plus rising price is a trend with a crowd already in it.
CME basis and the futures term structure. This is where institutional flow shows up first, and it's where I made my money after the ETF approvals. Once the spot Bitcoin ETFs launched, the basis and the options surface started moving in ways that had nothing to do with retail sentiment. I ran a vol-arb book — CME futures against Coinbase Prime options — through the first month of ETF trading and pulled roughly $800,000 of premium decay out of mispriced implied volatility. The edge wasn't directional. It was that institutional desks price vol off a different model than the crypto-native market, and for a few weeks the two models disagreed loudly enough to trade.
The options surface itself. BTC options have a real surface — deep strikes, tight spreads, a term structure you can trust out to a year. ETH's is thinner but functional. Alt options are a different animal. Most liquid names barely clear enough volume for a genuine two-way market, which means the implied vol you see quoted is closer to a dealer's opinion than a market-cleared price. Greeks don't price narrative risk. Delta, gamma, vega — none of them have a term for "the influencer tweeted." If you're expressing an altseason thesis through alt options, understand that you're not trading volatility. You're trading a liquidity premium with optionality stapled to it.
One more, and it's a check on all of the above: exchange net position change. Coins moving onto exchanges ahead of a rally is historically sell-side; coins moving off is accumulation. If the month before the signal showed sustained outflows from venues into cold storage, the move has a base under it. If it showed coins piling up on exchanges while price rose, you're watching someone distribute into strength — and the signal's arrival is the last thing that seller needed.
The structural read on any altseason signal in 2026 is this: the marginal buyer isn't a Discord degen anymore. It's a desk with a risk limit and a vol model. When the desk's model says alt vol is cheap, flow shows up. When the model says alt vol is rich, flow evaporates — regardless of what the on-chain signal prints. That's the difference between 2021 and now, and most cycle indicators were never rebuilt for it.
The Regime Shift Nobody Recalibrated For
Let me put a sharper point on this, because it's the part I think is being missed.
The Altcoin Cycle Signal, like most of its peers, was designed and largely back-tested in an era when crypto's marginal buyer was a retail participant with an exchange account and a leveraged long habit. In that regime, dominance behaved like a sentiment dial — reflexive, fast, mean-reverting around narratives. Signals worked because the crowd the signal described was also the crowd that acted on it. Self-reinforcing.
That regime is gone at the margin. Not dead — the tail still trades like 2021. But the money that moves the aggregate now includes ETF creation baskets, basis desks, and market makers whose alt exposure is hedged, delta-neutral, and managed by a risk system that doesn't read Substack. When that cohort is the marginal buyer, dominance doesn't behave like a sentiment dial. It behaves like the spread between two funding markets. Different math. Different signal.
Which is why I keep coming back to the lag. If the signal fires a month into the move, and the marginal buyer is a desk that priced the move in week one, then the signal is announcing something the informed cohort already monetized. It's not that Glassnode is wrong. It's that the audience receiving the signal is structurally late. That's not a bug in their product. It's a property of publishing to a public feed.
And notice what gets launched into this window. Every altseason narrative arrives with a product attached — a new basket, a new index, a new L2 whose entire go-to-market is "the chains are fragmenting, we fix it." I've watched this cycle three times now. The complaint is always that liquidity is scattered across too many venues, too many chains, too many rollups, and the remedy always requires buying something new. The scatter is real. The implied fix is marketing. Fragmentation doesn't get solved by a token; it gets solved by latency, settlement finality, and whoever convinces enough projects to deploy first. That's a distribution question wearing an engineering costume.
I learned this one the hard way in 2017. I audited the contracts of an ERC-20 called CryptoGem that had raised $2.4 million — and found an integer overflow in the transfer logic that let an attacker mint arbitrary balances. I published the technical writeup rather than dumping my allocation, then shorted the token through Bitfinex's lending market. The rug came. I made about $150,000. Early adopters lost everything. What I took away wasn't that I was clever. It was that code is law, but trust is expensive — and that when you publish a finding, you hand the trade to whoever reads fastest while the majority reads last. An altseason signal works the same way at a bigger scale.
The Contrarian Read: Aimed, Not Accidental
Here's the part I'll be blunt about, because I've watched this movie from the short side.
The Altcoin Cycle Signal is a fine piece of analytics. It is also, functionally, a distribution instrument. Not by intent — by structure. Any signal that (a) fires only after a month of gains and (b) gets amplified by media and influencers will reach its largest audience at precisely the moment the earliest participants are hunting for exit liquidity.
I shorted this pattern once with conviction. Mid-2021, mapping BAYC-adjacent wallet clusters, I found wallets wash-trading floors specifically to trigger liquidations in lending markets. The floor wasn't a price. It was a puppet show. I put half a million dollars short on AAVE and ENS off that on-chain read, published the analysis on Substack, and got called a conspiracy theorist for three weeks — until regulators fined exchanges for wash trading and the thesis printed. The lesson was never about NFTs. It was that an NFT floor is a feeling, not a number — and so is every aggregate metric derived from one. When the input is a feeling, the output is a feeling with a dashboard.
Same shape here. The alt market cap that "grew for a month" includes assets whose quoted value is a thin-book mark, not a settlement price. When the signal fires, the crowd reads "altseason confirmed" and buys the tail. The tail's liquidity is exactly the thing that vanishes first. And the unlock calendar doesn't care about any of it. Vesting is linear, blind to sentiment, and merciless. In a strong tape, unlock-driven sell pressure gets absorbed and hidden. In a flat tape, it doesn't. The signal has no view on the supply schedule. Supply is the one variable that never got bullish.
Levels, Stops, and the Question That Matters
Watch the ratio, not the level. BTC dominance reclaiming and holding above 60% invalidates the altseason read outright — treat that as your stop. A clean break and hold below 56–57% is the confirmation that costs you something, which is exactly why it's worth more than a signal that fired a month ago.
Watch ETH/BTC before you watch anything in the tail. If the largest alt can't take share from Bitcoin, the small caps aren't going anywhere except down relative to the majors.
Watch stablecoin supply like it's the only honest witness in the room. Watch the funding curve for crowding. Watch the basis for institutional temperature. And keep position sizes small enough that being early on a correct thesis doesn't kill you before it pays — because with a right-side signal, that is the failure mode.
The question I'd leave you with isn't whether altseason is here. It's what the signal would need to remain true — and whether the market cap it measured was ever a number in the first place. Code is law, but bugs are justice. An unpublished methodology is neither.