Lido vs Rocket Pool vs Frax: The Demand Problem

The Three-Way Race Quietly Became Something Else

For three years the standard way to compare Ethereum’s liquid staking market was a straight line from Lido to Rocket Pool to Frax: a giant, a challenger, and a clever experiment. In late 2026 that framing has stopped describing reality. All three shipped their biggest upgrades to date this year, and yet the distance between them grew rather than shrank. Lido’s stETH is worth about 21.74 billion dollars and sits at number 11 by market capitalization across all of crypto. Rocket Pool’s rETH is worth roughly 809 million dollars. Frax’s frxETH and its staked version together hold around 230 million dollars. One of these is a monetary asset. The other two are products.

The usual explanation for that gap is design: Lido got there first, Rocket Pool is more decentralized, Frax is more financially engineered. Those differences are real, and this piece walks through every one of them. But they are no longer what separates the three. Each protocol now runs modern, audited, upgraded machinery. What separates them is demand for the token that comes out the other end. Lido has it in abundance, Rocket Pool is spending its entire 2026 governance cycle trying to manufacture it, and Frax has effectively stopped competing for it. The comparison most readers are searching for has quietly become a case study in why demand, not technology, decides a liquid staking market.

Staking belongs to the same family of questions as mining economics. Both pay a thin, volatile yield for helping secure a chain, and both have watched that yield compress as more capital crowds in. Bitcoin’s miners spent 2026 watching difficulty eat their margins; Ethereum’s stakers have watched issuance do the same. Ether’s base staking rate has fallen to roughly 2.63 percent, down from above 5 percent in early 2023. When the reward is that small, the question is not who pays a little more. It is whose receipt you can actually use.

What a Liquid Staking Token Actually Is

Staking on Ethereum means locking ETH to run a validator, the software that proposes and attests to blocks and earns newly issued ETH plus a share of transaction tips and MEV for doing so honestly. The catch is that a full validator takes 32 ETH, and once deposited that ETH is illiquid and subject to an exit queue. Liquid staking solves the lockup by pooling deposits, running the validators for you, and handing you a transferable receipt token that represents your stake plus its accruing rewards. You keep exposure to the yield while holding something you can trade, lend, or post as collateral.

Two barriers make this worth paying a fee for. The first is the 32 ETH minimum, which prices most people out of running their own validator; pooling lets someone stake a fraction of that. The second is operational risk. A validator that goes offline bleeds small penalties, and one that double-signs or otherwise misbehaves can be slashed, losing part of its stake outright. Spreading deposits across many professionally run validators turns a single point of failure into a diversified basket, which is the quiet reason even large holders who could afford 32 ETH often choose a liquid staking token instead.

The receipts come in two shapes, and the difference matters for taxes and for how the yield looks. A rebasing token like stETH keeps a one-to-one peg to ETH and pays you by increasing the number of tokens in your wallet each day. A reward-bearing token like Rocket Pool’s rETH, Frax’s sfrxETH, or Lido’s own wrapped wstETH keeps a fixed token count and lets each unit appreciate against ETH instead. One rETH is now worth about 1.178 ETH, one sfrxETH about 1.177 ETH. Same economics, different plumbing: the rebasing version sprays a daily stream of small rewards you may have to account for, while the reward-bearing version buries the yield in an exchange rate you realize only when you sell.

Where does the yield come from? The same place a miner’s revenue comes from, just on the other side of Ethereum’s move to proof of stake: the protocol pays validators to secure it. That payment is consensus issuance plus execution-layer tips and MEV. It is not a company’s cash flow and it is not a government coupon, which is why it floats with network activity and with the amount of ETH competing for it. As of early November 2026 roughly 36 percent of all ether is staked, about 44 million ETH across some 813,000 validators, down from nearly 975,000 validators at the start of the year as the network consolidates. The more ETH chasing the same issuance, the thinner the slice for each validator.

Three Receipts, Three Machines

Before the demand question, the mechanics. All three protocols pool ETH and run validators, but they disagree on who runs those validators, how much the operators must put at risk, and what the receipt looks like. The table is the quick map; the sections after it fill in the detail.

Two things jump out. Lido is the only one whose receipt is a rebasing, number-goes-up token that most of DeFi treats as a cash equivalent. And the three governance tokens, LDO, RPL, and the rebranded FRAX, now play very different roles, which turns out to be the clearest readout of each protocol’s demand.

Lido: The Network That Keeps Compounding

Lido is the incumbent, and it is not close. Its stETH represents close to 9.7 million ETH, somewhere around 22 percent of all staked ether and roughly three fifths of the liquid staking segment. It charges a flat 10 percent fee on staking rewards, split between its node operators and the DAO treasury. The operator set is curated rather than open, which is the source of both its reliability and its central criticism.

In 2026 Lido rebuilt its back end. Through the Curated Module v2, launched 27 July 2026, it began consolidating more than 8 million ETH, worth about 16 billion dollars when the plan was set, onto Ethereum’s larger post-Pectra 0x02 validators. The share of Lido stake on those validators has climbed from about 32 percent toward 52 percent, and the migration is expected to cut the network’s validator count from roughly 880,000 toward 628,000, easing load on the beacon chain. For stakers the change is invisible; for Ethereum it is one of the largest single reorganizations of the validator set to date.

Lido’s answer to the concentration worry is dual governance, live since 4 July 2025. stETH holders can escrow their tokens to object to a DAO decision: reach 1 percent of staked ETH and a proposal is delayed, reach 10 percent and the protocol enters a rage-quit state that freezes new proposals and lets dissenters exit before anything contested executes. It does not decentralize the operator set, but it hands stakers a brake on the LDO holders who govern the protocol.

The most telling 2026 move is that Lido is no longer only a staking protocol. The DAO approved a roughly 60 million dollar budget to push beyond staking into stablecoin and institutional yield, launching pooled vaults it brands EarnUSD and EarnETH as base staking returns compress. The market leader is using its distribution to sell something other than staking, which tells you how it reads the next few years.

Rocket Pool: The Decentralization Bet With a Demand Problem

Rocket Pool is the permissionless alternative. Anyone can run a node by putting up their own ETH bond alongside pooled ETH, which is why its operator set numbers in the thousands rather than the dozens. Ordinary stakers hold rETH, a reward-bearing token worth about 809 million dollars in total, with protocol value locked near a billion dollars. That makes Rocket Pool the most credibly decentralized of the three and, by market size, a distant second.

The model works because of that bond. A Rocket Pool operator posts their own ETH plus RPL as collateral, then runs validators topped up with ETH from the rETH pool; if they misbehave, their own stake is penalized first, so rETH holders are insulated. Saturn 1 moved the protocol to megapools, which let one operator run many validators under a single contract and lowered the required bond, with the stated aim of making node operation cheaper and pulling in more ETH. The design is elegant and genuinely permissionless. The trouble is that cheaper node operation only helps if there is rETH demand to match it, and that is precisely the lever the protocol has struggled to move.

Its 2026 was dominated by the Saturn upgrades. Saturn 1 arrived in February and introduced megapools and a lower bond; the Saturn 2 package went through governance in the autumn. Votes closed 13 October 2026, approving a protocol-level rETH withdrawal queue that lets holders trigger validator exits for a full redemption, keeping the node bond at 4 ETH after voters rejected both a cut to 1.5 ETH and a rise to 6 ETH, and adopting a random megapool exit mechanism.

Then the tokenomics. Under Saturn 2, RPL issuance is set to fall to 2.5 percent a year, with node-operator RPL rewards eliminated entirely and the remaining inflation routed to the DAO treasuries. For years RPL was the reward that drew node operators; the protocol is deliberately turning that faucet off.

Why? Because Rocket Pool has diagnosed its own bottleneck, and it is demand for rETH. The Saturn 2 specification states plainly that the protocol DAO “views rETH demand-related work as a higher priority than implementing complex new RPL value-capture mechanisms,” and that it is “unclear whether rETH demand will be sufficient” to support smaller-bond validators. Governance even hosts a proposal bluntly titled increase rETH demand by reducing the voter share. The engineering is excellent; the problem is that not enough people want the token it produces, and the protocol knows it.

Frax: The Protocol That Walked Off the Field

Frax built the most financially clever staking product of the three, and then, over 2025 and 2026, decided staking was not its main business. Its design is a two-token system: frxETH is a liquid token that pays no yield on its own, while sfrxETH is a vault that concentrates all the staking rewards for holders who lock in. Because non-stakers (for example, liquidity providers holding frxETH in a pool) forgo their share, the yield piled into sfrxETH has often run higher than a plain validator rate. One sfrxETH is now worth about 1.177 ETH.

The newest iteration, frxETH v2, recasts staking as a lending market: a node operator deposits ETH into a pool contract under their own control and borrows additional ETH from the protocol to fund validators, using the validators themselves as collateral. Frax moved it to a capped mainnet around mid-October 2026, a deliberately small pilot of roughly ten validators mostly run by Frax while the team watched performance before lifting the cap. It is sophisticated, and it is tiny: frxETH and sfrxETH together are worth around 230 million dollars, a rounding error next to stETH and roughly flat in ETH terms for a year.

The two-token split was clever engineering, but it also split demand. Where Lido offers one asset that is both liquid and, once wrapped, yield-bearing, Frax asks users to choose between a liquid token that earns nothing and a vault token that earns more but circulates less. That trade can beat the market on raw yield, yet it never produced a receipt that DeFi wanted to hold as a reserve, and without that anchor the product stayed small. By the time frxETH v2 arrived, the team had already concluded its energy was better spent on stablecoins.

The reason is strategic, not technical. Under the North Star upgrade, Frax renamed its FXS governance token to FRAX and made it the gas token of its Fraxtal network, spun the old Frax stablecoin into frxUSD, and reoriented the entire project around being, in founder Sam Kazemian’s words, “a scalable stablecoin operating system, not just a single protocol.” frxUSD is built to fit the new United States stablecoin rulebook, which America finished legislating in 2026. Liquid staking is now a feature of that system rather than the headline. Frax did not lose the staking race so much as leave it.

Why DeFi Treats stETH as Money

Here is the mechanism behind the gap. A liquid staking token is only as useful as the places that accept it, and acceptance compounds. stETH and its wrapped form wstETH are the deepest-integrated staking assets in DeFi: the Curve stETH pool is one of the largest in the ecosystem, Aave and other lending markets take wstETH as prime collateral, and a long tail of vaults, restaking systems, and structured products are built to hold it. Every integration makes stETH easier to use, which pulls in more holders, which justifies the next integration. That flywheel is the real moat, and it is a demand-side moat, not a technical one.

The clearest sign of that moneyness is leverage. Because wstETH is accepted as collateral, holders routinely borrow stablecoins or ETH against it and restake the proceeds, a loop that only works when the collateral is deep and liquid enough that lenders trust it. Whole strategies, and at times whole protocols, have been built on stETH as a base reserve asset. That is not a status you can engineer into a token; the market confers it on the one that is already the most used, which is why it is so hard for a smaller receipt to catch up even with identical yield and better decentralization.

rETH and sfrxETH are accepted in many of the same venues, but the liquidity is thinner and the integrations fewer, so they trade less like money and more like a position. The practical consequence is that a stETH holder can move size in and out at a tight spread, while a large rETH or sfrxETH holder has to think about slippage. For an individual staking a few ETH the difference is minor. For a treasury or a protocol, it is the whole decision.

This is why the comparison is really about demand. The three protocols produce receipts with nearly identical underlying yield, but only one of those receipts has become a monetary primitive. The other two are competing for a network effect that, by its nature, tends to concentrate in a single winner.

The Yield Is Almost Identical, So It Is Not the Tiebreaker

Strip away the branding and the headline numbers nearly touch. Ethereum’s base staking rate is about 2.63 percent as of early November 2026, with MEV adding perhaps half a point to a full point for well-run validators. After fees, Lido’s stETH, Rocket Pool’s rETH, and Frax’s sfrxETH all land within a few tenths of a percent of one another. The figures below are approximate and move daily.

Two honest caveats. First, sfrxETH’s edge comes from other frxETH holders giving up their yield, so it depends on how much frxETH sits idle; it is real, but it is not free. Second, and more important, none of these rates clears the dollar yield on short-term United States Treasury bills, which stayed higher than ether’s staking rate through most of 2026. Staking is an ETH-denominated coupon layered on top of ETH price exposure, not a substitute for cash yield. That is why value-accrual and fees, not headline APR, are where these protocols actually differ.

What moves the base rate is plain supply and demand for stake. The more ETH staked, the more validators split the same issuance, so the rate drifts down as the staking ratio climbs toward the level it holds today. MEV cuts the other way, lifting returns when markets are busy and arbitrage is plentiful, which is why a well-connected validator can out-earn a passive one by a meaningful margin. All three protocols capture MEV through relays and pass most of it through to holders, so the real differences there are smaller than the marketing sometimes implies.

Fees, and Where Your Spread Actually Goes

When the gross yield is a wash, fees decide the net, and the three take their cut differently. Lido keeps a flat 10 percent of rewards. Rocket Pool does not charge rETH holders a protocol fee directly; instead its node operators take a commission on the pooled ETH they run, and rETH holders receive the blended rate net of it. Frax routes all yield to sfrxETH and takes a 10 percent vault fee, with frxETH holders earning nothing unless they stake into the vault.

The deeper question is where the protocol’s own revenue goes, because that is what separates real yield from token emissions. Lido’s fee funds a treasury and, increasingly, its expansion into non-staking products. Rocket Pool is in the middle of rewiring its economics so that protocol revenue, not inflation, rewards participants, which is the whole point of cutting RPL issuance and ending operator rewards. Frax has folded staking revenue into a broader stablecoin-and-DeFi income stream that it pays out to holders who lock the FRAX token. Three different theories of how a staking business should make money, converging on the same realization that paying people in your own freshly minted token is not a durable model.

The Governance Tokens Are a Demand Readout

If you want a single view that tells the story, look at the governance tokens. They do not represent a claim on the staked ETH; they govern the protocols and, in theory, capture a slice of fees. In practice their prices track how much the market believes in each protocol’s future demand, and the readout is stark.

LDO trades near 0.32 dollars, down about 95.6 percent from its 2021 peak, which sounds grim until you see RPL at about 1.38 dollars, down nearly 97.8 percent from its high with a market cap under 35 million dollars. Even the leader’s token is deeply out of favor; the challenger’s is an afterthought. FRAX, meanwhile, is no longer really a staking token at all, so its value tracks Fraxtal and frxUSD rather than frxETH. The message across all three is the same one the TVL figures send: capital is not betting on a liquid staking land-grab anymore. It is betting on whichever receipt already won.

Decentralization: The Trade Only Rocket Pool Is Still Making

The strongest case against Lido has nothing to do with yield. It is that a single protocol controlling around a fifth of all staked ETH, run by a curated operator set, is a concentration risk for Ethereum itself. If any one entity approaches a third of stake it can threaten the chain’s finality guarantees; past two thirds it could finalize unilaterally. Ethereum co-founder Vitalik Buterin has repeatedly flagged staking concentration as one of the biggest risks to the base layer and pushed for designs that keep the validator set broad and home staking viable.

Rocket Pool is the live answer to that worry. Its permissionless design means no gatekeeper decides who runs a validator, and its node operators must bond their own ETH, aligning them with the protocol’s safety. That is the product Rocket Pool is really selling, and it is a genuine good. The difficulty, as its own governance admits, is that decentralization has not translated into demand for rETH; the market has rewarded Lido’s convenience and liquidity over Rocket Pool’s trust-minimization. Lido’s dual governance is a partial concession to the same concern, giving stakers a veto without opening the operator set. Frax sits furthest from the decentralization ideal, with a small, largely in-house validator set that it clearly no longer treats as a selling point.

History sharpens the point. Back in 2023, when Lido’s share of staked ETH was nearing a third, the DAO was asked to vote on whether to self-limit its own growth for the good of the network. It declined, overwhelmingly, which told everyone that no large staking protocol will voluntarily cap itself and that the fix, if there is one, has to come from Ethereum’s design rather than from goodwill. There is also the censorship question: a dominant operator set choosing which transactions to include would be its own kind of problem, which is why inclusion lists and relay diversity sit in the same debate.

For an individual this is a values choice as much as a financial one. If you believe Ethereum’s long-run health depends on not letting one protocol dominate, rETH is the vote that matches the belief. If you only care about the liquidity and integrations of your receipt, the market has already told you where those are.

Getting Your ETH Back

“Liquid” means two different things, and conflating them is the most common staking mistake. The first meaning is the secondary market: you can swap stETH, rETH, or frxETH for ETH on an exchange or a decentralized pool instantly, paying whatever spread the market demands. The second is primary redemption: you ask the protocol to unstake and return your ETH at full value, which means joining Ethereum’s validator exit queue and waiting. Speed costs price; price costs time.

Lido’s stETH has the deepest secondary liquidity of any staking token, so swapping is usually cheap even in size, and its direct withdrawal path returns ETH at par subject to the exit queue. Rocket Pool historically relied on a withdrawal buffer plus the rETH market; its new rETH withdrawal queue, approved in the Saturn 2 votes, adds a protocol-level redemption path that triggers validator exits when enough rETH requests accumulate, closing a real gap. Frax’s sfrxETH unwinds through its vault and the frxETH market. Across all three, a calm market makes every route cheap and a panicked one makes every route expensive, which is exactly when the depth of stETH’s liquidity earns its reputation, as it did when stETH held its peg through 2026’s sharp sell-off.

The exit queue is a shared constraint, not a protocol feature: when many validators try to leave at once, everyone waits, regardless of whose receipt they hold. That is worth remembering before assuming any liquid staking token is a same-day exit in a crisis.

What US Regulators Say About Holding the Receipt

For United States readers the regulatory picture got clearer in 2025 and has held. The SEC’s Division of Corporation Finance stated on 5 August 2025 that, in its view, “Liquid Staking Activities … do not involve the offer and sale of securities within the meaning of Section 2(a)(1)” of the Securities Act. That removed a cloud that had hung over the whole category and helped open the door to staking-linked exchange-traded products.

The guardrail is specific, and worth understanding, because it is also a control test. The safe harbor holds only if the provider stays administrative and ministerial: it must not decide “whether, when, or how much” of your assets to stake, and it must not “guarantee or otherwise set the amount of rewards.” A protocol that merely pools, stakes, and hands back a proportional receipt fits inside the line; a custodian promising a fixed return or exercising discretion falls outside it. That framing favors transparent, non-discretionary designs, and it is one reason institutions have grown comfortable holding staking receipts through regulated custodians, including names like BitGo, which went public and chartered in 2026. None of this is tax advice, and the rebasing-versus-reward-bearing distinction still changes how your rewards are reported, so check your own situation with a professional.

That clarity has a demand dimension too. Once the securities cloud lifted, the door opened to regulated, staking-linked exchange-traded products that let traditional investors hold staked-ETH exposure without touching a wallet, and the receipts those products accumulate flow disproportionately to the most liquid, most institutionally legible token. Regulatory certainty, in other words, has tended to reinforce the same concentration the decentralization camp worries about, by making the incumbent receipt the easy default for the largest pools of capital.

How to Choose in Late 2026

Strip it down and the decision follows the demand story. If you want the most liquid, most widely accepted staking receipt, one you can lend, collateralize, and exit at a tight spread, stETH is the default and the network effects are not close. If you care more about Ethereum’s decentralization than about maximum composability, and you are willing to hold a thinner-traded token to back that belief, rETH is the honest choice, and Rocket Pool is the only one of the three still building its whole pitch around it. If you already live in the Frax ecosystem or want the concentrated vault yield of sfrxETH, Frax still works, but understand that you are holding a product its own maker now treats as a side feature of a stablecoin network.

Whichever you pick, hold the receipt in a wallet you control rather than leaving it on a venue, because the token is a bearer claim on your stake. And watch the demand side, not the feature list. The protocols will keep shipping upgrades; the one that matters is whether the market still wants to hold what each of them mints. On that measure late 2026 has already returned a verdict, and the open questions now are whether Rocket Pool can change it, and whether Lido’s lead becomes a problem Ethereum itself has to manage.

Frequently Asked Questions

Is Lido, Rocket Pool, or Frax the best liquid staking protocol in 2026?

There is no single best; it depends on what you value. Lido’s stETH offers the deepest liquidity and the widest DeFi acceptance, Rocket Pool’s rETH offers the strongest decentralization, and Frax’s sfrxETH offers a concentrated vault yield inside the Frax ecosystem. For most users who want to use their receipt as collateral or exit at a tight spread, stETH’s network effects make it the practical default.

Why is Rocket Pool so much smaller than Lido?

Mainly demand for the receipt token, not technology. Rocket Pool is more decentralized, but rETH has far thinner liquidity and fewer integrations than stETH, so capital has concentrated in Lido. Rocket Pool’s own 2026 governance names rETH demand as its top priority, which is a candid admission that the gap is a demand problem rather than an engineering one.

Did Frax quit liquid staking?

Not formally, but close in practice. frxETH and sfrxETH still run, and frxETH v2 reached a small capped mainnet in October 2026, yet the combined product is worth only around 230 million dollars and is roughly flat in ETH terms. Under its North Star upgrade Frax reoriented around stablecoins and its Fraxtal network, making liquid staking a side feature rather than its headline business.

Is staked ETH a security in the United States?

The SEC’s Division of Corporation Finance said in August 2025 that liquid staking activities, as described, do not involve the offer and sale of securities, provided the staking provider stays administrative and does not exercise discretion over your assets or guarantee rewards. It is staff guidance rather than a law, but it removed a major source of uncertainty for United States holders.

What is the difference between stETH and rETH?

stETH is a rebasing token that stays pegged near one ETH and pays you by adding tokens to your balance each day. rETH is a reward-bearing token with a fixed count whose value rises against ETH over time; one rETH is worth about 1.18 ETH today. The economics are similar, but the rebasing versus exchange-rate design changes how rewards are tracked and taxed, and stETH is far more liquid.

Yuki Tanaka covers staking, validators, and the economics of proof-of-stake for HOGE Wire.