Any post that says "liquid staking is safe" is only half telling the truth. People have lost real money in liquid staking. Not because the concept is broken, but because a handful of specific failure modes catch out people who didn't know they existed. This piece names each one, points at historic examples, and covers what actually reduces the risk. Numbers below are illustrative.
Failure mode 1: selling into a depeg
A liquid staking token's market price can dip below the value of the ETH it stands for. That's a depeg. The classic example is stETH trading around 0.94 ETH for weeks in June 2022. Anyone who held through it saw the peg recover once Ethereum shipped withdrawals in 2023. Anyone who panic-sold at the low locked in a permanent loss.
The fix: understand that depegs are almost always temporary, driven by liquidity shortages rather than protocol failures. If the redemption path back to ETH is working, arbitrageurs will close the gap. See the liquid staking explainer for how the peg mechanism actually works.
Failure mode 2: exchange failure
Users who held stETH on Celsius in 2022 didn't lose the stETH to any protocol bug — Celsius went under and took customer assets with it. FTX did the same thing months later. The token was fine; the venue collapsed.
The fix: self-custody. If your stETH sits in your own wallet, no exchange failure can touch it. The extra work is setting up a wallet and writing down a seed phrase on paper. That's it.
Failure mode 3: liquidation from leverage
People use liquid staking tokens as collateral on Aave to borrow ETH, buy more stETH, and repeat. When ETH drops sharply, the loan-to-value ratio breaches the threshold and the platform sells the stETH to repay the debt — usually at the worst possible moment. 2022 produced a lot of this. So did the March 2023 stETH volatility episode.
The fix: if you loop at all, keep leverage low (1.3x, maybe 1.5x) and monitor daily. Or, more sensibly, don't loop.
Failure mode 4: smart-contract bug
Top liquid staking protocols carry multiple audits and long track records. That's evidence, not proof. A subtle bug in a widely trusted contract is exactly what past crypto disasters looked like. If it hits, losses could reach large percentages of the pool.
The fix: split funds across two audited protocols instead of one, so a single contract failure doesn't take everything. Ignore small unaudited protocols promising high yields.
Failure modes 5 and 6: slashing and picking the wrong protocol
Slashing hits validators that break Ethereum's rules — they lose part of their stake, and in pooled setups that loss spreads across all token holders. Historic slashing has been small in aggregate — under 1% of total staked ETH — but a badly configured operator can lose ETH quickly. The fix: pick protocols with diversified operator sets and multiple client teams.
Picking the wrong protocol is the flip side. Small protocols with unclear operators, no audits, or governance tokens marketed as "the future of staking" have failed. Some of that failure was outright exit scams. Some was contract bugs. Some was slow bleed as users left and yields collapsed. All of it looked like easy yield at the start. The fix: stick to protocols with real audits, real track records, and real operator diversity. Lido, Rocket Pool, Coinbase cbETH, Jito and Marinade on Solana — the boring names because boring is what worked.
Comparing the six failure modes side by side:
| Failure | Historic frequency | Reduces to |
|---|---|---|
| Selling into depeg | Common in stress weeks | Rare, if you don't panic |
| Exchange failure | Multiple times per cycle | Zero with self-custody |
| Liquidation | Regular for loopers | Rare with low leverage |
| Smart-contract bug | Not yet at top protocols | Halved with diversification |
| Slashing | Ongoing, small | Small anyway |
| Wrong protocol | Common with new tokens | Zero with major protocols |
Read that table twice. Most historic losses came from failures that individual users could have avoided with basic hygiene. The staking risks guide covers each in more depth.
Is liquid staking safe and when you can lose your crypto
The honest summary: liquid staking is reasonably safe in normal conditions, using major protocols, with self-custody, at position sizes you can afford to lose. It becomes dangerous fast if any of those four conditions break — leverage on top, custody on a shaky exchange, an obscure protocol promising 30% APY, or a position size that would ruin you if lost. Every failure mode above has victims who ignored one of those conditions. The rewards calculator on this page helps compare realistic yields against realistic risks. Just don't confuse "has held up so far" with "cannot break". Neither is quite true.