Lido vs Rocket Pool vs Frax: ETH Liquid Staking in 2026

Ethereum’s staking market has turned into a story about one protocol and everyone else. By August 2026 roughly 41.4 million ETH, about 34 percent of the circulating supply, sits locked as stake, an all-time high, according to Coinpedia’s staking research. A large share of that ETH never touched a validator client directly. It was routed through a liquid staking protocol that handed the depositor a token in return: a receipt that keeps earning rewards while staying tradeable. Three names dominate that business, and they are the subject of this comparison: Lido, Rocket Pool and Frax.

They are nowhere near the same size. Lido’s stETH alone carries a market value near 17.9 billion dollars, per CoinGecko; Rocket Pool’s rETH is around 706 million dollars; Frax’s entire staked-Ether operation sits under 300 million dollars. Scale, though, is only one axis. The three make genuinely different bets on decentralization, yield engineering and how much control a staker keeps. This guide lines them up on the metrics that actually decide the question in 2026: size, yield, fees, node operator design, token mechanics, smart contract risk and the regulatory picture after the SEC’s 2025 shift. Every price and figure here is a mid-August 2026 snapshot; crypto markets move by the hour, so treat the numbers as a reference point, not a promise.

The 30-second version

If you read only one part of this article, read this table. Lido wins on liquidity and reach, Rocket Pool wins on decentralization, and Frax competes on engineered yield for people who live inside DeFi. The rest of the piece explains why, and where each of those advantages quietly turns into a risk.

What liquid staking actually solves

Native Ethereum staking asks a lot. You need 32 ETH per validator, hardware or a cloud box running an execution and a consensus client, near-constant uptime, and the willingness to lock that capital while it earns. Miss too many attestations and you leak rewards; sign two conflicting blocks and you get slashed. On top of that, deposits and withdrawals pass through queues that stretched past two months at points in 2026 as exchange-traded funds and corporate treasuries piled in. For most holders that is too much friction for a yield that now sits in the low single digits.

Liquid staking removes the friction. You deposit ETH into a protocol, the protocol runs the validators, and you receive a token backed by that staked ETH plus its accruing rewards. The token trades freely, works as collateral in other applications, and lets you exit by selling on the open market rather than waiting in the withdrawal queue. Where Bitcoin pays miners for proof-of-work, Ethereum pays validators for proof-of-stake, and the economics of that yield can get every bit as thin as the ones in our field guide to reading a Bitcoin miner’s margins. The three protocols below are the biggest ways to earn that yield without touching a validator client yourself.

The 2026 backdrop: more stakers, thinner yield

The context for any staking decision in 2026 is compression. Ethereum’s issuance is roughly fixed, so as more ETH gets staked, each validator earns a smaller slice. Total stake hit an all-time high near 41.4 million ETH, and the seven-day staking rate slid to about 2.66 percent, down from a 5.06 percent peak in June 2023, per Coinpedia. ETH itself trades around 1,887 dollars, according to CoinGecko.

The stake kept climbing even as the reward fell because the buyers were not chasing yield alone. Yield-bearing spot ETFs now pass staking rewards through to shareholders, corporate treasuries lock ETH at scale, and a months-long entry queue replaced the exit rush that defined late 2025. That wall of demand is precisely why the choice of liquid staking token got sharper this year. When the base yield was above 5 percent, a fee of one or two percentage points barely registered. At 2.66 percent, every basis point of fees and every scrap of extra yield changes the ranking. The three protocols keep different amounts of your reward, and they generate different amounts of it, so a tenth of a percent that looked trivial in 2023 is now a meaningful share of a sub-3-percent return.

Lido: the incumbent that ate the market

Lido is the default. Deposit ETH and you get stETH, a rebasing token whose balance in your wallet grows a little each day as rewards land. If you would rather hold a token with a fixed balance and a rising price, you wrap it into wstETH, which is the form most DeFi protocols prefer. As of mid-August 2026 stETH carries a market capitalization around 17.9 billion dollars across roughly 9.47 million tokens, per CoinGecko. That is somewhere near a quarter of all staked ETH, and roughly 60 percent of the liquid staking segment specifically, according to DefiLlama figures, which makes Lido larger than every competitor combined.

Lido charges a 10 percent fee on staking rewards, not on your principal, split evenly between the node operators running the validators and the Lido DAO treasury, per the protocol’s own documentation. Historically Lido ran a curated set of around 30 professional node operators, which is the crux of the centralization complaint against it. The protocol has been widening that base: a Community Staking Module now lets permissionless operators join with a modest bond, and a Simple DVT module splits validator duties across clusters using distributed validator technology, so no single machine holds the keys.

Two 2026 developments matter for anyone weighing Lido. First, Lido V3 went live on 30 January 2026 with stVaults, modular and isolated staking vaults aimed at institutions that want to keep custody and pick their own operators while still minting stETH liquidity on demand. Second, Lido activated dual governance, a mechanism that lets stETH holders veto or delay DAO decisions they dislike and, in the extreme, exit before a contested change takes effect, a design meant to blunt the risk of a hostile takeover of the DAO, as Cryptopolitan reported. Governance itself runs through the LDO token. In short, Lido is trying to answer the decentralization critique with process and modularity rather than by shrinking.

Rocket Pool: the decentralization maximalist

Rocket Pool exists to be the opposite of a curated operator set. Its liquid token, rETH, uses an exchange-rate model: one rETH is worth more than one ETH and the gap widens over time as rewards accrue, so about 1 rETH equals 1.17 ETH as of mid-August 2026, per CoinGecko. rETH’s market capitalization sits around 706 million dollars, a fraction of stETH, and the protocol’s total value locked is near 993 million dollars, per CoinGecko’s Rocket Pool page.

The real difference is who runs the validators. Anyone can be a Rocket Pool node operator, and several thousand independent operators do, a far more distributed set than Lido’s. Operators put up their own ETH bond plus, optionally, the RPL token as extra collateral, and the rest of each validator’s 32 ETH comes from the rETH pool. The Saturn 1 upgrade, live on 18 February 2026, cut the minimum node bond from 8 ETH to 4 ETH and introduced megapools that let one operator group many validators under a single contract for gas and capital efficiency. Saturn also reworked the fee model, moving toward a lower base commission for operators plus a revenue share paid to stakers of RPL, and made RPL collateral optional rather than mandatory.

The trade-offs are the mirror image of Lido’s. rETH is the most credibly decentralized major liquid staking token, which is exactly what Ethereum researchers say the network needs for censorship resistance. But rETH liquidity is thinner, its DeFi integrations are fewer, and running a node, while cheaper after Saturn, still asks more of you than clicking a stake button. The RPL token, which trades near 1.39 dollars with a market cap around 31 million dollars per CoinGecko, has also had a rough few years, because its value tracks node operator demand rather than the size of the staking pool itself.

Frax: the yield engineer

Frax approaches the same problem from a different angle: split the token in two. frxETH is pegged one-to-one with ETH but pays no staking yield on its own; it is designed to sit in liquidity pools and earn trading fees. sfrxETH is the vault you deposit frxETH into to capture the staking rewards. The trick is that not everyone stakes their frxETH, so the full validator yield gets concentrated among the smaller set of sfrxETH holders, which pushes sfrxETH’s effective rate above the raw network average, as the Frax documentation describes. sfrxETH trades at a growing premium to ETH, reflecting years of compounding rather than any peg to the dollar.

frxETH v2 pushed the design further by turning validation into a lending market. Node operators borrow ETH from a Frax lending pool at a variable interest rate to run their validators, posting collateral much as a borrower would on a money market, and sfrxETH captures both the staking yield and the interest those operators pay, alongside income from Frax’s automated market operations on Curve. It is the most financially engineered of the three approaches, and also the smallest: Frax’s entire Ether operation holds roughly 283 million dollars, per CoinGecko, less than two percent of Lido’s footprint.

Frax Ether does not live in isolation, and that is both its selling point and its risk. It is one primitive in a stack that includes the frxUSD stablecoin, launched in early 2025 with reserves tied to BlackRock’s tokenized BUIDL fund, and Fraxtal, Frax’s own layer-2 network. In a hard fork the project calls North Star, Frax renamed its long-standing FXS governance token to FRAX at the end of December 2025 and turned it into the gas token for Fraxtal, per CoinGecko. Founder Sam Kazemian has built Frax as an interlocking system rather than a single product, which means sfrxETH’s fortunes are bound up with the health of the wider Frax ecosystem in a way stETH’s are not.

Head to head: the numbers that decide it

The table below lines up the three protocols on the criteria most people actually weigh. Read it alongside the caveats in the sections that follow, because a single number rarely captures the whole trade-off.

How the three tokens accrue value, and why it matters

There are two ways a liquid staking token can grow. A rebasing token like stETH keeps a price near one ETH and increases the number of tokens in your wallet as rewards arrive. A reward-bearing token like wstETH, rETH or sfrxETH keeps the token count fixed and lets each token become worth more ETH over time. The distinction sounds academic, but it drives real differences. Rebasing is intuitive for a holder watching a balance tick up, yet it is awkward for smart contracts, which is why wstETH exists and why Rocket Pool and Frax skipped rebasing entirely.

The accrual model also shapes how the token behaves as collateral. stETH and wstETH are among the most widely accepted assets in DeFi lending, deep enough that traders loop them: deposit wstETH, borrow ETH, buy more wstETH, and repeat to lever up the base yield. That works until the token slips below its expected value and cascading liquidations hit, which is precisely the failure mode our on-chain credit risk map walks through. rETH and sfrxETH are accepted in fewer places and trade in thinner pools, so they carry less systemic weight but are harder to deploy and quicker to gap under stress. If your plan is to use the token across DeFi, Lido’s liquidity is a genuine, measurable edge; if you just want to hold and earn, it matters much less.

Decentralization and the one-third question

The strongest argument against Lido has nothing to do with its product and everything to do with its size. Ethereum Foundation researcher Danny Ryan laid out the concern in a widely cited note: a single staking entity crossing one-third of all staked ETH could stall the network’s finality, half could enable censorship, and two-thirds could in principle finalize invalid chains. Lido’s stETH share of all staked ETH hovers in the mid-20s to low-30s percent depending on how you count, close enough to that first threshold to keep the debate alive. The DAO put a self-limiting cap to a vote back in 2022, and the option to stay uncapped passed with more than 99 percent support, which tells you how the incentives inside a dominant protocol point.

Vitalik Buterin has repeatedly named staking centralization as one of the biggest risks facing Ethereum, framing it as a core target of his research phase called the Scourge and noting that a mere handful of actors already choose the contents of the overwhelming majority of Ethereum blocks, as The Block reported. His proposed remedies include capping how much any one entity can stake and splitting staking into risk-bearing and risk-free tiers. This is the backdrop against which Rocket Pool markets itself and against which Lido built dual governance. The table below compares the operator design that sits underneath each token.

Yield: why sfrxETH can pay more, and the catch

Start from the base rate. With the network paying roughly 2.66 percent, a stETH holder receives that minus Lido’s 10 percent cut, landing a little under the network average. An rETH holder receives the pooled validator rewards minus the node operator commission, which after Saturn sits at a lower base than the roughly 14 percent operators historically took, so rETH tends to track close to the average too. sfrxETH is the one that can print a visibly higher number, because the frxETH holders who provide liquidity but do not stake effectively hand their share of the rewards to the sfrxETH vault, and frxETH v2 layers lending interest on top.

The catch is that a higher yield on a much smaller and more complex system is not free money; it is compensation for extra risk. sfrxETH’s boosted rate depends on the ratio of unstaked to staked frxETH, on the lending market functioning as designed, and on the Frax ecosystem staying healthy. A few tenths of a percent of extra APR is real, but it is riding on more contract surface and less liquidity than stETH. For a long-term holder who is not going to touch the position, the differences between the three base yields are small; for an active DeFi user who can compound and layer strategies, they add up over a year.

Liquid staking is not restaking

A lot of confusion in this corner of the market comes from blurring two layers. Liquid staking gives you a receipt for ETH that secures Ethereum itself. Restaking takes that receipt and re-pledges it to secure additional services, in exchange for extra yield and, crucially, extra slashing risk. The two are siblings, not the same thing, and the market spent 2026 pulling them apart rather than bundling them together; ether.fi, for one, removed restaking from its flagship token to move back toward plain staking. If you want the layer above these three protocols, our explainer on how the restaking market is unbundling covers it in depth.

The practical takeaway is simple. stETH, rETH and sfrxETH are the foundation, and restaking is an optional, riskier floor you can build on top. Judge the three staking tokens on staking merits first. Whether to then restake is a separate decision with its own risk budget, and conflating the two is how people end up with more exposure than they realize.

The risks: depegs, oracles and smart contracts

The reference event for liquid staking risk is still the 2022 stETH depeg. During the Terra collapse that May, stETH slid from about 97 cents on the ETH to the low 90s as leveraged holders rushed the exit and a large withdrawal thinned the main Curve pool, according to a Nansen analysis reported by CoinDesk. There was no hack; it was a pure liquidity and confidence event, made worse by the fact that stETH could not yet be redeemed for ETH. That last piece has since changed: withdrawals have been live since the 2023 Shapella upgrade, so an arbitrage now anchors stETH to ETH in a way it could not in 2022. Redeemability does not stop short-term price gaps, but it does put a floor under them.

Smart contract risk is the other constant. All three protocols are heavily audited and run bug bounties, but audits reduce risk rather than remove it, and the smaller and more novel a system is, the more untested surface it carries. Frax’s lending-market design and Rocket Pool’s megapools are newer than Lido’s battle-tested contracts. There is also oracle and liquidation risk in how these tokens get used downstream: in March 2026 a price-oracle safeguard on the Aave money market briefly misread wstETH about 2.85 percent low and triggered tens of millions of dollars of avoidable liquidations before the risk team pledged to reimburse affected users. None of this makes liquid staking uniquely dangerous, but it is why spreading exposure across tokens, rather than blindly maximizing yield, is the sober play.

Regulation: the SEC’s 2025 green light

For US readers the regulatory weather changed sharply in 2025. The baseline had been the SEC’s 30 million dollar settlement with Kraken in early 2023 over its staking-as-a-service product, which read as a warning to the whole industry. Then, on 5 August 2025, the SEC’s Division of Corporation Finance issued a staff statement concluding that certain liquid staking activities and the staking receipt tokens they produce do not involve the offer or sale of securities. The reasoning is that a liquid staking provider acts as an agent, holding assets, staking them per the protocol and issuing or redeeming receipt tokens, rather than as a manager promising returns through its own entrepreneurial efforts.

SEC Commissioner Hester Peirce made the point bluntly in a companion statement, arguing that providing security to a blockchain is not the same thing as offering a financial security. The practical effect was to clear a path for staked-ETH exchange-traded funds, and by March 2026 major issuers including BlackRock had launched products that bake staking in, as CoinDesk reported. The safe harbor has edges, though: a provider that guarantees a fixed return, or that exercises discretion over how and when to stake, can fall right back outside it. And this is a US framing; the anti-money-laundering and licensing layer that applies to the providers themselves runs on a separate track, as our guide to FATF crypto guidance and the Travel Rule explains. Other jurisdictions draw the lines differently.

Which one should you actually use?

There is no single winner, only a best fit for what you are optimizing. A few honest rules of thumb:

• You want the deepest liquidity and the widest DeFi and institutional reach: Lido. You are accepting that you are also feeding the concentration problem, which dual governance and the Community Staking Module are meant to soften but do not eliminate.
• You care most about Ethereum’s decentralization, or you want to run a validator with less capital: Rocket Pool. You trade some liquidity and simplicity for a token whose operator set actually reflects the network’s ideals.
• You live inside DeFi and want engineered, boosted yield: Frax’s sfrxETH, provided you are comfortable with a smaller system and with exposure to the broader Frax stack.
• You are staking a meaningful amount: consider holding more than one. Spreading across stETH and rETH, for example, hedges both smart contract risk and the moral hazard of feeding a single dominant protocol.

Whichever you pick, hold the token in a wallet you control rather than leaving it on an exchange, and understand how that wallet signs transactions before you start moving size; our comparison of MetaMask, Phantom and Rabby is a reasonable starting point. Liquid staking has matured into real financial infrastructure, but a receipt token is only ever as safe as the contracts behind it and the keys in front of it.

Frequently Asked Questions

Is Lido or Rocket Pool better for staking ETH in 2026?

It depends on your priority. Lido offers the deepest liquidity, the widest DeFi support and institutional-grade vaults, which makes stETH the more practical choice for most users and treasuries. Rocket Pool offers far stronger decentralization through several thousand permissionless node operators, which matters if you value censorship resistance or want to run your own validator with a 4 ETH bond. Neither is objectively better; they optimize for different things.

Why is stETH so much bigger than rETH and sfrxETH?

Lido moved first, integrated aggressively across DeFi and became the default liquid staking token, which created a liquidity flywheel: the more places stETH is accepted, the more people hold it, and the more useful it becomes. As of August 2026 stETH’s market cap sits near 17.9 billion dollars, versus roughly 706 million for rETH and under 300 million of total value locked for Frax Ether. Scale is now Lido’s main advantage and its main risk at once.

Does sfrxETH really pay a higher yield than stETH?

Often yes, by a modest margin. Because frxETH holders who supply liquidity do not receive staking rewards, those rewards concentrate among sfrxETH holders, and frxETH v2 adds lending-market interest on top, which can lift the effective rate above the network average. The extra yield is compensation for a smaller, more complex system and tighter liquidity, so it is not risk-free outperformance.

Is liquid staking legal in the United States?

After the SEC Division of Corporation Finance staff statement of 5 August 2025, certain liquid staking activities and the receipt tokens they issue are treated as not involving the offer or sale of securities, provided the provider acts as an agent rather than promising returns through its own efforts. That cleared a path for staked-ETH ETFs. The safe harbor has limits, and a provider guaranteeing returns or exercising staking discretion can fall outside it, so the treatment is favorable but conditional.

Can stETH, rETH or sfrxETH lose their peg?

They can trade below their expected ETH value under stress, as stETH did during the 2022 Terra collapse when it slipped into the low 90s of cents on the dollar. Since Ethereum enabled staking withdrawals in 2023, redemptions anchor these tokens to ETH through arbitrage, which limits how far and how long a discount can persist. Short-term gaps are still possible, especially for the smaller and less liquid tokens, so treat a tight peg as normal but not guaranteed.

By the HOGE Wire staking desk. Figures verified against CoinGecko, DefiLlama and protocol documentation as of 13 August 2026; this article is informational and not financial advice.