this is a note about solstrategy, one of the projects i have been working on, and more generally about what you do when the financial instrument you want to build would be trivial on tradfi infrastructure and impossibly awkward on the crypto stack available today.
the original idea is, structurally, a port of michael saylor's microstrategy trade to solana. saylor's insight was that a public company trading above nav functions as a leveraged vehicle on its underlying asset: the company raises capital at a premium, buys the asset, and the per-share asset exposure compounds. holders get implicit leverage without touching a margin account. the mechanism's elegance is that there is no forced deleveraging. there are no liquidations because there is no loan. the leverage lives in the capital structure.
the crypto-native version wants the same property: amplified exposure to an underlying (sol) with no liquidation risk, low volatility decay, and on-chain transparency. the way most defi protocols approach this is wrong, and understanding why is the first fifty percent of the design work.
why perpetual futures are the wrong answer. perps give you leverage by posting a margin account and borrowing against it. when price moves against you, either you top up or you get liquidated. the liquidation cascade is not a bug; it is the mechanism by which the system stays solvent without loans being called back. for an individual trader this is fine. as a primitive for "leveraged exposure to sol that a buy-and-hold user actually wants to hold for a year," it is terrible. the user's exposure is continuously conditional on a price path they cannot control, and historical base rates say something like a third of perp longs get liquidated at some point during a normal crypto drawdown.
why leveraged tokens are the wrong answer. the standard 2x/3x-leveraged token rebalances daily to maintain its leverage ratio. this works in trending markets and is catastrophic in choppy ones, because rebalancing in a mean-reverting regime systematically sells low and buys high. the literature on constant-rebalanced portfolios has known this since 1985, and retail holders of leveraged etfs discover it every few years when a sideways market eats fifteen percent of their notional for no apparent reason. the fundamental issue is volatility decay: the more you rebalance, the more you pay for the privilege of maintaining leverage in non-trending tape.
what solstrategy does instead. the protocol raises usdc, sol, and liquid staking tokens and uses them to build a long-duration treasury of sol lsts that pay staking yield. the protocol issues two tokens against this treasury: $sstr, which represents a claim on treasury nav plus a premium that reflects market faith in the bullish thesis, and $cnt, which is a pure nav-tracking stable-ish instrument that also serves as a looping mechanism.
the leverage, crucially, comes from the premium, not from borrowed money. when the market believes sol will outperform, $sstr trades above the nav of the treasury it is backed by, and that premium grows non-linearly with the underlying thesis. when the market is bearish, the premium compresses toward nav, which functions as a floor. in the worst case — protocol bankruptcy, treasury decomposed at par — a user who bonded one sol worth of assets gets back at least one sol's worth of lsts, plus accrued staking interest. the downside is bounded by the treasury; the upside is unbounded by the premium. this is the shape of a leveraged bet without a loan.
the $cnt leg is how you turn this from a buy-and-hold vehicle into an actively-looped one. a user who wants more leverage sells their $cnt for additional usdc or sol and re-enters the $sstr trade. each loop adds exposure to the premium without touching a lending market or a liquidation engine. it is exactly the mechanism saylor used when microstrategy repeatedly issued convertible debt to buy more btc, translated into an on-chain primitive where the "debt issuance" is token re-emission and the "bondholder" is a voluntary $cnt holder.
the parts that are genuinely hard. three, in order of nastiness.
first, the nav accounting has to be clean. lst yields accrue continuously, treasury composition drifts, and the premium depends on users being able to trust that the nav they are paying a premium to is the nav they think it is. this requires public, verifiable, on-chain accounting with real-time tracking of treasury positions, which is a solvable engineering problem but not a glamorous one.
second, the redemption mechanism has to handle a run. if everyone wants their sol back at once, and the treasury is in lsts with unstaking cooldowns, you have a duration mismatch that looks exactly like a bank run. the protocol handles this by making redemption a two-step claim — submit, wait for the cooldown, collect — rather than instant, which rations redemption in a crisis without requiring a lender of last resort. this is unpopular with users who want instant liquidity. it is also the only design that does not blow up during the inevitable stress event.
third, the premium is both the product and the risk. when it is high, users pay more per unit of underlying exposure. when it is low, the vehicle offers less leverage. the protocol cannot directly set the premium; it emerges from $sstr's secondary-market price. what the protocol can do is emit or absorb tokens in response to premium levels — issuing more $sstr when the premium is extreme, buying back when it compresses — which is the microstrategy playbook in reverse. this is a pure mechanism-design problem and one of the places where the difference between a working protocol and a vaporware one lives.
the one-line summary: leverage-without-liquidation is not a contradiction; it is a capital-structure pattern that tradfi has known about for a century. the hard part is not the leverage. it is designing the treasury, the redemption, and the premium-stabilization mechanism in a way that holds up through the full crypto cycle. everything above is what solstrategy is trying to get right. some of it we will get wrong. that is why it is interesting.
for the docs and a longer technical walkthrough, see solstrategy.so.