Staking Your Crypto Once Locked It Up. Liquid Restaking Lets You Stake the Same Coins Twice. Here's the New Risk That Creates
5 min read
Loading...
5 min read
Ordinary staking has always had one tradeoff: you lock up your coins to earn a yield, and locked coins can't do anything else. Liquid restaking breaks that tradeoff. It lets the same staked ETH secure a second protocol on top of Ethereum itself, and hands you a tradable token so your money isn't actually stuck. That sounds like a free upgrade. It isn't. Every extra protocol your ETH secures is an extra place something can go wrong, and in April 2026 one of those extra places lost about $292 million in a single attack.
Staking is locking up ETH to help run the Ethereum network, in exchange for a yield paid in more ETH. Liquid staking (Lido is the largest provider) solves the "locked up" part: you deposit ETH and get back a token, like stETH, that represents your staked position and can be traded, sold or used elsewhere while your original ETH keeps earning. Lido's own published rate has stETH paying a net staking APR of roughly 2.2% to 2.6%, after Lido's fee, according to StakingRewards and Lido's own rate disclosures.
Restaking is a separate idea layered on top. A protocol called EigenLayer lets that same staked ETH be pledged a second time to help secure other services (oracles, bridges, data-availability networks) that aren't part of Ethereum's own core protocol. In exchange, you can earn additional rewards from those services. A liquid restaking token, or LRT, such as ether.fi's eETH, Renzo's ezETH or Kelp DAO's rsETH, is the receipt token for that restaked position, letting it stay liquid and usable in other DeFi apps the same way a liquid staking token does.
The mechanics are straightforward: you deposit ETH (or an existing liquid staking token) with an LRT protocol. That protocol restakes it through EigenLayer, delegating it to an operator who runs infrastructure for one or more external services. You get an LRT back, which you can hold, trade or deposit into other DeFi protocols for further yield. Your original ETH is now doing three jobs at once: securing Ethereum, securing whatever external service the operator opted into, and backing the LRT you're holding or using elsewhere.
Plain ETH staking already carries slashing risk: a validator that misbehaves or goes offline can lose a portion of its stake. Restaking adds a second slashing surface on top: opting into an external service means accepting that service's own penalty conditions, enforced by whichever operator you delegated to. A careless or dishonest operator can cost you money through no fault of your own.
On top of that sits smart contract risk, and it compounds rather than replaces the layer below it. Your ETH now passes through Ethereum's own staking contracts, then EigenLayer's contracts, then the LRT protocol's own contracts, and often a cross-chain bridge so the LRT can move between blockchains. A flaw anywhere in that chain, not just in the newest layer, can put your funds at risk. That is not a hypothetical.
On April 18, 2026, an attacker drained about $292 million in rsETH, Kelp DAO's liquid restaking token, according to Coindesk and blockchain-forensics firm Chainalysis, which attributed the attack to North Korea's Lazarus Group. The point of failure wasn't a staking or slashing bug at all: it was the cross-chain bridge, built on the LayerZero messaging protocol, that let rsETH move between blockchains. That bridge was configured to trust a single verifying node before releasing funds, a "1-of-1" setup with no backup check. The attacker compromised that one node and used it to approve a fake, oversized withdrawal.
The damage didn't stay inside Kelp DAO. Lending platforms that accepted rsETH as collateral froze those markets to limit further losses, and total value locked across all of DeFi fell from $99.497 billion to $86.286 billion in the following 48 hours, a roughly $13 billion drop, according to DefiLlama data reported by Coindesk. One bridge, in one restaking protocol, was enough to shake the entire sector.
A protocol advertising extra yield for restaking on top of ordinary staking is rarely paying all of it in ETH. A meaningful share is commonly distributed as the protocol's own token, or as "points" convertible into a token later, rather than as a guaranteed return. That means part of the "yield" you're quoted depends on a token holding its price after you receive it, which is a separate bet from the staking yield itself. Treat any advertised restaking APR as two different things bundled together: a smaller, real portion paid in ETH, and a larger, speculative portion paid in something else entirely.
Restaking doesn't just add a new yield source. It adds new operators, new smart contracts and, as Kelp DAO showed, new bridges, each one a separate place something can fail. Before adding that stack on top of an ETH position, it helps to know what the base asset itself already does without any of these extra layers. NetWort's Explore screener lets you check ETH's own volatility ("bumpiness") and worst-case historical loss side by side with other assets, a starting point for judging how much extra risk you actually want to add on top. If you're still weighing whether ordinary staking is worth it in the first place, the risk-adjusted math on plain crypto staking is the layer underneath all of this, worth understanding before adding restaking on top of it.