Hook
Ansem rotated into LUNA. He also rotated out. The public version of that trade ends there — a clean entry, a clean exit, a compounded profit. The ledger version does not. LUNA's terminal value is zero. Anyone who followed the same rotation and missed the same exit holds the identical asset Ansem once treated as a rotation target, now as a dead position. That asymmetry is not a footnote. It is the story. The narrative describes a skill; the ledger describes a distribution. One trader compounded. The tail of that distribution lost everything, and the tail is invisible in the retelling. A trade described only by its winning sample is not a methodology. It is a marketing document.
Context
The source material is not a protocol. It is a trader's post-mortem. Ansem, a widely followed voice in the Solana ecosystem, summarized his largest lesson from roughly eight years of crypto trading: rotation between sectors, followed by disciplined profit-taking. The 2017 cycle, he admits, ended with his gains returned to zero. The 2020–2021 cycle ended better — rotating capital across SOL, LUNA, and AVAX, taking profit on the way up, and compounding the realized gains.
Two claims anchor the piece. First, that catching a rotation early is only the first step. Second, that knowing when to sell is an equal skill. Read plainly, that is behavioral finance, not alpha. Read critically, it is a survivorship-biased account of a negative-sum game dressed as a repeatable edge.
The assets named are not equivalent. SOL runs an inflationary schedule with partial fee burn. AVAX has a hard cap with a burn mechanism. LUNA, in the 2020–2021 configuration, ran an infinite-mint model tied to the UST algorithmic stablecoin. Treating three structurally incompatible supply models as interchangeable rotation targets is a statement about price momentum, not about value. The strategy's unit of analysis is the chart. The risk it ignores is the mechanism underneath the chart.
Core
I reconstruct death spirals for a living, so let me be precise about what the third asset actually was. The ledger does not lie, only the narrative does.
In 2022 I traced 50,000 transactions on the Terra blockchain across the UST de-peg. The finding was not that the market panicked. The finding was that the mint/burn arbitrage was deterministic. When UST traded below $1, the protocol's incentive was to mint LUNA and burn UST — which expanded LUNA's supply precisely when its price was collapsing. Arbitrageurs extracted roughly $4 billion in under 72 hours. The system did not fail because sentiment turned. It failed because the incentive structure had only one terminal state. Panic is just poor data processing in real-time; the arbitrageurs were not panicking. They were executing. The mechanism handed them the exit and the retail holder the bag.
The reserves behind UST were not reserves in any auditable sense. They were a treasury of volatile assets plus a reflexivity loop — confidence propping the peg, the peg propping confidence. Collateral was a mirage; solvency was a myth. The rotation framework never asks what backs the asset. It only asks whether the price is rising. That question is sufficient right up to the moment it is fatal.
Now place that asset inside a "rotation plus compounding" framework. The framework has no fundamental screen. It has a momentum signal and an exit rule. If the exit rule is "take profit when the chart says so," then the strategy's survival depends entirely on execution speed — and execution speed is exactly what a follower of a KOL lacks. The KOL sells, publishes, and the follower is already behind the tape. The information asymmetry is not a side effect of the format. It is the format.
There is a second accounting problem. Rotation is presented as compounding. Compounding assumes realized gains are redeployed into productive capital. In practice, each rotation incurs taker fees, slippage across thin order books, and — depending on jurisdiction — taxable events. A trader rotating quarterly across three volatile L1s can bleed a meaningful percentage of notional to friction alone. The published curve never shows this. The compound curve in the narrative is gross. The compound curve in the account is net, and the two diverge in proportion to how often you rotate. The more active the rotation, the wider the gap.
Third, the profit-taking advice is correct and unactionable. "Know when to sell" has no threshold. Is it a 2x? A 50% drawdown from the high? A moving-average cross? Without a quantified rule, the instruction collapses into a slogan. Worse, it contains a genuine paradox: take profit too early and you forfeit the main advance; take it too late and you surrender the gain. The author supplies the second failure mode from his own 2017 record — gains returned to zero — but not the metric that would have prevented it. A discipline without a number is not a discipline. It is a sentiment.
Fourth, the sample. The 2017 loss is disclosed. The 2020–2021 rotation into SOL, LUNA, and AVAX is disclosed. What is not disclosed is the failure sample: how many rotations lost, how many exits were late, how many positions were held through reversals. A methodology is only as good as its denominator. We have the numerator — the wins — and a story about the denominator. That is the definition of survivorship bias, and it is the single largest risk embedded in this content.
I have made this mistake in my own ledger. In 2021 I deployed a monitor across 1,000 low-cap NFT collections, tracking mint rate and holder concentration. The data showed that eight of ten trending collections had zero active developers — the price was driven by bots, not community. I had the signal and I still held one position three days past the point my own rule said exit. The rule was right. I was the variable that failed. That is the honest version of "know when to sell": the rule is easy to write and hard to obey, and no KOL post fixes that.
This pattern is not unique to Terra. In 2026 I audited NeuroPay, an AI-agent microtransaction protocol, and found a reentrancy vulnerability in its oracle integration that let an attacker drain $2 million from the liquidity pool in a single transaction. The novelty of the AI-crypto convergence did not matter. The logic gap did. Novelty is not security. Speed without formal verification is a liability, not a feature. The same principle governs rotation targets: an asset's momentum tells you nothing about whether its mechanism can survive contact with a stress event.
There is a compliance edge to this that rarely gets stated. In several jurisdictions, publishing a directional trade thesis while holding the referenced asset can trigger disclosure obligations — or, at minimum, the appearance of a pump. The source piece does not disclose the author's current positioning at the time of writing. That omission is not necessarily bad faith. It is simply the standard gap between a KOL's public signal and their private book, and the gap is where followers get hurt.
One more mechanic. As more participants adopt rotation, the window compresses. A cycle that once offered months between sector peaks offers weeks, then days. The strategy that worked in 2021, executed at 2025 speeds, produces more trades, more friction, and thinner margins. Rotation strategies decay with adoption, which is why the traders who profited from them are usually the ones who stop describing them. The act of teaching the strategy is the act of eroding it.
There is a structural point underneath all of this. Rotation is not creation. It is redistribution. When capital moves from SOL to LUNA to AVAX, no new value is minted — it is transferred between holders. The winners' realized gains are, by construction, the losers' unrealized losses at the moment of the cross. In a rotation market, the profit of the disciplined is the loss of the undisciplined, and the ledger balances to zero before friction. This is why "rotation alpha" cannot scale. The more followers execute the same signal, the thinner the spread between entry and exit, until the edge is arbitraged to nothing.
Contrarian
Here is what the bulls — and Ansem — get right, and it is worth conceding without decoration.
The most valuable sentence in the entire source is the 2017 admission. "Profits returned to zero" is a confession most traders never make. It signals iteration capacity, which is rarer than a winning year. Most accounts publish the curve and bury the drawdown. This one surfaces the zero. That single act of disclosure is more informative than the rotation thesis itself.
The second correct claim is the ranking of skills. Early recognition of a rotation is table stakes; it is the entry. Exit discipline is the edge, because it is anti-human. Holding a winner triggers the same neural loop as holding a loss — the reluctance to act against an open position. Structure outlives sentiment; code outlives hype. The only skill that survives a full cycle is the mechanical one: a pre-committed exit, executed without negotiation. Ansem is right that this is the lesson. He is simply wrong to imply that it is transferable by reading. Emotion is a variable I exclude from the equation, and so should anyone reading a strategy that depends on obeying a rule under pressure.
Takeaway
Watch the messenger, not just the message. When a KOL broadly distributes a "rotation made me rich" narrative, the distribution itself is a data point. Rotation alpha depends on a window; the moment the window is public, it is closing. The rotation described here ran in a specific cycle, on a specific asset set, one of which is now worth zero. The question for the next bull run is not which sector rotates next. It is who is still holding the tail when the music stops — and whether they had a number that told them to leave.
