Lido vs Rocket Pool vs Frax: Who Survives the Drawdown
ETH has fallen roughly 64% from its peak, yet a record 36% of supply is now staked. We ask whether Lido, Rocket Pool, and Frax can run a staking business through the crash.
On 12 November 2026, Ethereum slipped below $2,000 for the first time since August, and by mid-month the price sat near $1,810, down about 64% from the August 2025 record of $4,946 (CoinGecko). On almost any reading that is capitulation. Yet on the same day, the share of ETH locked in staking hit an all-time high of roughly 36% of supply, about 44.6 million ETH (Bitcoinist). Holders were not fleeing. They were locking in.
That divergence, a price in freefall while staking climbs to a record, is the backdrop for the question this piece asks. Lido, Rocket Pool, and Frax are the three names most people reach for when they want a liquid version of staked ETH. We have compared them many times: on yield, on fees, on decentralization, and on how fast you can get your ETH back. This time the test is harsher and simpler. A liquid staking protocol is a business, and a 64% drawdown is the closest thing the sector has to a live fire drill. Which of these three is actually built to run a staking business through a crash?
The short version: the receipt tokens held their pegs, the deposits kept coming, and the three protocols behind them look less alike than ever. One is diversifying out of a position of strength, one is fighting for demand it openly admits it may not have, and one has quietly left the race altogether.
A 64% Crash Meets a Record Staking Rate
Start with the numbers. ETH trades near $1,810, down roughly 64% from its all-time high of $4,946.05 set in August 2025 (CoinGecko). Over the same week the token fell about 11%. And yet, on 12 November, some 44.6 million ETH was staked, around 36% of the circulating supply and the highest share in Ethereum’s proof-of-stake history, up from a prior record of 32.4% in June (Bitcoinist, CryptoTicker). The queue data told the same story: roughly 900,000 ETH was waiting to enter staking against about 250,000 waiting to exit, so even in the worst month of the year far more ETH wanted in than out.
Why lock up a falling asset? Two reasons. First, staking is denominated in ETH, not dollars: a staker is accumulating more ETH and betting the coin recovers, so the dollar price of each reward is somebody else’s problem. Second, the marginal staker of 2026 is increasingly not a retail holder but a treasury company or a fund. Reporting around the record points to Ethereum treasury firms such as BitMine and Grayscale steadily adding to staked positions (Bitcoinist), coins that, once staked and often wrapped inside an ETF or a corporate balance sheet, do not come back to market quickly.
The pattern is not new, only larger. Analysts at XWIN Research Japan, writing via CryptoQuant, noted that “Ethereum staking levels continued to climb even during the sharp corrections of 2022 and 2023” (Bitcoinist). One caveat matters, though: a record staking ratio is not a price floor. The coins that still trade set the price, and with spot Ether ETFs seeing outflows and Bitcoin itself under pressure, a high lock-up rate has not stopped the slide (CryptoTicker). For our three protocols, though, the number that counts is the deposit base, and the deposit base grew.
Liquid Staking in One Paragraph
Solo staking on Ethereum needs 32 ETH and locks it behind an exit queue. Liquid staking pools that requirement: you deposit any amount, a protocol runs the validators, and you receive a token that represents your staked ETH plus the rewards accruing on it. That token is the product. Two designs dominate. A rebasing token (Lido’s stETH) keeps a one-to-one name with ETH and grows the number of tokens in your wallet each day. An exchange-rate token (Rocket Pool’s rETH, Frax’s sfrxETH, and the wrapped version of stETH, wstETH) keeps the token count fixed and lets each unit become worth more ETH over time, which is why 1 rETH trades for about 1.18 ETH and 1 sfrxETH for about 1.17 ETH today (CoinGecko). The distinction matters for taxes and for how a token behaves as DeFi collateral, and we have covered it before. Here the point is simpler: whichever design you hold, your claim is only as good as the business running the validators behind it.
Three Very Different Businesses
The three protocols share a product category and almost nothing else. Lido is a DAO-governed protocol that is, by a wide margin, the largest: its stETH carries a market cap around $17.6 billion, representing roughly 22% of all staked ETH and more than 40% of the entire liquid staking market (CoinGecko, Crypto Briefing, DefiLlama). Rocket Pool is the permissionless alternative: anyone with 4 ETH and some of its RPL token can run a node, and its rETH is a far smaller $640 million or so (CoinGecko). Frax is the outlier: its frxETH and sfrxETH together are worth under $200 million (CoinGecko), and, as we will see, liquid staking is no longer the point of Frax at all.
| Protocol | Receipt token(s) | Model | Receipt market cap | Share of staked ETH | Governance token | 2026 direction |
|---|---|---|---|---|---|---|
| Lido | stETH / wstETH | Permissioned DAO, curated plus community operators | ~$17.6B | ~22% | LDO ~$220M (down ~96%) | Diversifying beyond staking |
| Rocket Pool | rETH | Permissionless, 4 ETH bond plus RPL | ~$640M | ~1% | RPL ~$28M (down ~98%) | Fighting for rETH demand |
| Frax | frxETH / sfrxETH | Team-run, now a side product | <$200M | <0.2% | FRAX (North Star rebrand) | Pivoted to stablecoins |
Why a Bear Market Is the Real Audit
Strip away the branding and a liquid staking protocol earns money one way: it takes a cut of the staking rewards its validators produce. Those rewards are paid in ETH. When ETH falls by two-thirds, the dollar value of that revenue falls by two-thirds with it, even if not a single depositor leaves. Meanwhile the costs of running the business, smart-contract engineers, audits, oracle infrastructure, legal and compliance work, relayer relationships, are mostly priced in dollars and do not fall. Worse, the asset each DAO was meant to use as both treasury and incentive currency, its governance token, has fallen harder than ETH: LDO is down about 96% from its peak and RPL about 98% (CoinGecko, CoinGecko).
So the audit a bear market performs is not “did the receipt hold its peg.” It is “can the business behind the receipt still pay its bills, keep its operators honest, and fund the next upgrade on a fraction of the revenue it modeled in the bull market.” A staking token can trade at a flawless peg right up until the protocol issuing it quietly runs out of runway. That is the lens for the rest of this piece. It is the same lens we apply to miners when Bitcoin trades below the cost of production, the subject of our look at the price floor stock-to-flow missed: security that is subsidized is only as durable as the subsidy behind it.
Lido: Scale as a Moat, Diversification as Insurance
Lido enters the drawdown from the strongest revenue position by far. A roughly 10% fee on the rewards of about 9.76 million ETH worth of deposits is still a large, recurring, ETH-denominated income even after the price has halved, simply because the base is so large (CoinGecko). That scale is the moat.
But scale in plain staking is a shrinking prize. Base rewards have compressed to around 2.4% to 2.6%, the lowest in Ethereum’s proof-of-stake history (CryptoTicker), and Lido’s own share of staked ETH has slipped from about 23.9% in January to roughly 22% by mid-November (Crypto Briefing). Lido’s answer has been to stop being only a staking protocol. In its 2026 budget the DAO approved about $60 million in spending (roughly $43.8 million for core operations and $16.2 million for liquidity and institutional development) and shipped products beyond stETH: stVaults, which let institutions run customized staking setups, and Lido Earn, a suite of structured yield vaults branded EarnETH and EarnUSD (Bitcoin.com). Institutional staking on Lido rose from 25.9% to 35.3% of its base in the first half of 2026, and the DAO has framed the shift as a move toward what it calls “real-business DeFi” (Bitcoin.com). It is the same reach into tokenized, off-chain-linked yield we traced in real yield’s pipe into tokenized treasuries and stocks.
The other half of Lido’s 2026 story is engineering. Through a year-long Curated Module v2 migration, nicknamed Lido Core, the protocol has been consolidating validators onto Ethereum’s new compounding (0x02) design, moving more than 8 million ETH, worth about $16 billion, and cutting its validator count by roughly a third (The Block, Crypto Briefing). Network-wide, more than 40% of staked ETH now runs on these larger validators, and Lido pushed its own compounding share from about 32% to 52% (Crypto Briefing). This is the single biggest reason Ethereum’s total validator count has fallen from around 975,000 early in the year to under 800,000 now. In a bear market the relevance is cost: fewer, larger validators are cheaper to run per dollar staked, which is exactly the efficiency a protocol wants when its revenue is shrinking.
Bigger is not automatically safer for the network, though. Vitalik Buterin has repeatedly flagged staking concentration as one of the biggest risks to Ethereum’s base layer (The Block), and a single protocol sitting near a quarter of all staking has long been the sector’s uncomfortable centerpiece. For Lido, the live risk in this drawdown is political, not financial.
Rocket Pool: Decentralized, and Fighting for Demand
Rocket Pool does decentralization honestly. There is no curated operator set; anyone can run a node by posting a bond, now 4 ETH after the Saturn One upgrade in February 2026, alongside some RPL as additional collateral (Crypto Briefing). Around 2,000 independent operators run the network. If you care who actually holds the keys to your staked ETH, rETH is the cleanest mainstream answer.
The problem is demand, and Rocket Pool’s own governance says so out loud. rETH is a $640 million token (CoinGecko), a rounding error next to stETH, and the Saturn 2 upgrade, whose votes closed on 13 October 2026, was shaped around that fact. The headline change, RPIP-71, adds a protocol-level withdrawal queue so rETH can be redeemed directly rather than only sold on the market; the node bond stayed at 4 ETH after the community rejected both higher and lower options (Crypto Briefing). Most telling, the protocol DAO deprioritized new mechanisms to funnel value to RPL holders in favor of what it bluntly called rETH demand work, writing that it is “unclear whether rETH demand will be sufficient” (RPIP-86, Rocket Pool governance).
Rocket Pool has kept shipping, to be clear. The Saturn One upgrade in February 2026 cut the node bond from 8 ETH to 4 ETH and introduced megapools that batch many validators into a single contract, and node participation grew by roughly a third in the quarter that followed (Crypto Briefing). The engineering is not the weak point; the demand for the end product is.
For a bear market that is a double bind. RPL, the token that both aligns operators and acts as a first-loss cushion behind rETH, is down about 98% from its peak to a $28 million market cap (CoinGecko). When the collateral token craters, the economic insurance behind the receipt thins, and the DAO has less token value to spend on growth. Rocket Pool is not in danger of disappearing; its model is sound and its operators are real. But it enters 2027, when Saturn 2 reaches mainnet, needing to manufacture demand for a product the market has so far treated as niche.
Frax: The One That Already Left
Frax answered the demand question by changing the subject. Through a rebrand it calls North Star, Frax renamed its FXS token to FRAX and repositioned the whole project around being a stablecoin operating system, with FRAX as the gas token of its Fraxtal chain and a GENIUS-Act-oriented stablecoin, frxUSD, as the centerpiece (Frax governance, FIP-428). Founder Sam Kazemian has described Frax as “a scalable stablecoin operating system, not just a single protocol” (4pillars). Liquid staking, once a flagship, is now a side product.
The numbers show it. frxETH and sfrxETH together are worth under $200 million (CoinGecko), and frxETH’s old job as Fraxtal’s gas token was handed to FRAX in the rebrand. The one genuinely new piece of staking engineering, frxETH v2, reached a capped mainnet around mid-October 2026, but it launched small: roughly ten validators, most of them run by Frax itself, in a design that turns the protocol into a lending market between itself and its node operators, where operators post ETH collateral and borrow against an escrowed exit message while an off-chain health bot force-exits unhealthy positions (Frax project update).
For a holder weighing the three, the honest read is that Frax has made a strategic retreat, and in a bear market a retreat can be the smart move: better to concentrate scarce resources on the business you can win (stablecoins) than to bleed money defending a distant third place in staking. But it also means frxETH is the thinnest, least liquid, and least strategically supported of the three receipts, and should be treated accordingly.
Did the Pegs Hold? Stress-Testing the Receipt
Here is the good news for holders: through the worst ETH drawdown since 2022, the receipt tokens behaved. stETH traded within a whisker of ETH, both near $1,806 in mid-November (CoinGecko). rETH and sfrxETH, being exchange-rate tokens, simply kept climbing against ETH, now worth about 1.176 and 1.173 ETH respectively, because their value comes from accrued rewards rather than a market peg (CoinGecko, CoinGecko). Crucially, the exit queue stayed far smaller than the entry queue, roughly 250,000 ETH waiting to leave against 900,000 waiting to join (CryptoTicker). There was no stampede for the door.
But a peg is only as strong as the liquidity under it, and that is where the drawdown left a mark. The deep on-chain markets that once let anyone swap large size of stETH for ETH at par have thinned dramatically; the main Curve stETH pool, which held billions before the 2022 Terra collapse, is now a small fraction of that depth (DefiLlama). In practice that means small redemptions clear at par while a large, forced seller, the kind a cascading liquidation produces, would still face real slippage. The lesson of 2022 has not been repealed, only deferred. In June 2022 stETH fell to about 0.94 ETH when Celsius (holding some 445,000 stETH, about $1.4 billion at the time) and Three Arrows Capital were forced to dump into shallow pools, before withdrawals even existed (CoinDesk). Withdrawals exist now, which is the real improvement, but they are gated by the same validator exit queue for everyone.
There is a deeper point about what the word liquid means in a true rush. Secondary-market selling is only one exit; the other is protocol redemption, and all three receipts ultimately draw on the same Ethereum validator exit queue, which processes only a limited amount of ETH per epoch. In calm conditions that queue clears in days. If a large share of stakers tried to leave at once, redemptions would back up behind one another regardless of which receipt they held, and liquid staking would reveal itself as eventually liquid. The 2026 drawdown never triggered that scenario, since the exit queue stayed modest, but it is the tail risk every one of these tokens shares, and no amount of protocol engineering removes it, because the bottleneck lives in Ethereum itself.
| Receipt | Peg or rate | Market cap | What a large exit costs |
|---|---|---|---|
| stETH (Lido) | ~1.00 ETH (rebasing) | ~$17.6B | Small sells near par; big sells hit thinned Curve depth, then the shared exit queue |
| rETH (Rocket Pool) | ~1.176 ETH (exchange rate) | ~$640M | Secondary swaps or the new RPIP-71 redemption queue |
| sfrxETH (Frax) | ~1.173 ETH (exchange rate) | ~$78M | Thin secondary liquidity; redeem via the vault, then the exit queue |
The Governance-Token Massacre
If you want a single picture of how the bear market hit these protocols as businesses, look at their tokens, not their receipts. LDO is down about 96% from its 2021 high to roughly $220 million (CoinGecko). RPL is down about 98% to roughly $28 million (CoinGecko). The receipts held; the equity-like claims on the protocols did not.
Does it matter? It depends on how the token is wired into the product. For Lido, LDO is a governance and incentive token: stETH’s safety does not depend on LDO’s price, and a $220 million token is still enough to govern a protocol securing tens of billions. For Rocket Pool the link is tighter and the damage more real, because RPL is posted by node operators as collateral and functions as a first-loss cushion behind rETH, so a 98% decline genuinely thins the economic insurance and shrinks the budget the DAO can deploy. Frax sidestepped the question by making its token the center of a different business entirely.
This is where the health of a staking protocol starts to rhyme with the broader 2026 question of what happens when a crypto project’s token dies. A near-worthless governance token does not automatically kill the staking product, but it removes the war chest, dilutes the incentives, and, for designs that lean on the token as collateral, erodes the safety margin at exactly the moment it is most needed.
The Economics: Who Can Fund a Staking Business at $1,800 ETH
Put the revenue question in a table. The variable that decides survival is not the headline fee rate; it is the fee rate times the size of the base, in dollars, minus the cost of running the machine. Lido’s roughly 10% cut of the rewards on about 9.76 million ETH of deposits is a large recurring income even at $1,800 ETH, and it is split with operators and the DAO. Rocket Pool takes a commission on the ETH that node operators stake on behalf of rETH, but its revenue base, around 356,000 ETH behind rETH, is roughly one twenty-seventh of Lido’s. Frax’s staking revenue is now immaterial to a company whose main business is stablecoins.
| Protocol | Fee model | Revenue base (staked ETH behind the receipt) | Treasury or runway signal | Backstop in a crash |
|---|---|---|---|---|
| Lido | ~10% of rewards | ~9.76M ETH (~$17.6B) | $60M 2026 budget; diversifying into Earn and stVaults | Scale plus diversified revenue |
| Rocket Pool | Node commission plus RPL | ~356k ETH (~$640M) | DAO funded partly by RPL issuance; RPL down ~98% | RPL first-loss bond (thinned) |
| Frax | Vault spread (sfrxETH) | <$200M frxETH plus sfrxETH | Funded by the wider Frax stablecoin business | Not a priority; smallest and thinnest |
The takeaway: in a prolonged trough, revenue durability decides who keeps shipping upgrades and who stalls. Lido has the base and is building more on top of it; Rocket Pool has a sound model but a thin base and a cratered backstop token; Frax has removed itself from the equation. None of that shows up in the peg, which is precisely why the peg is the wrong thing to watch.
The Risk Stack a Bear Market Also Tests
Price is not the only thing a drawdown stresses. Falling collateral values trigger liquidations, and liquid staking tokens are among the most-used collateral in DeFi, a dynamic we unpacked in how receipt tokens run DeFi lending. stETH in particular backs billions of dollars in loans on Aave and elsewhere, which means an oracle hiccup or a sharp discount can cascade. We got a preview in March 2026, when a glitch in Aave’s price feed for wstETH briefly misread the token by about 2.85% and triggered roughly $27 million in liquidations across 34 accounts, although the protocol took no bad debt (CoinDesk).
Other risks stay live in a crash too: slashing (a validator penalty that, in a pooled protocol, is socialized across holders), smart-contract bugs (the reason all three lean on repeated audits and bug bounties), and plain operator failure. The structural mitigations differ sharply. Lido spreads risk across dozens of curated operators, 82 distributed-validator clusters, and a 644-operator permissionless Community Staking Module (Crypto Briefing); Rocket Pool spreads it across roughly 2,000 bonded operators; Frax concentrates it in a handful of mostly in-house validators. In a crash, diversification of operators is itself a form of safety, and on that axis Lido and Rocket Pool are structurally ahead of a Frax that has pulled staking in-house.
Regulation Did Not Blink
One thing that did not get worse in 2026 was the legal picture, at least in the United States. In an August 2025 statement that still governs the space, the SEC’s Division of Corporation Finance said that “Liquid Staking Activities … do not involve the offer and sale of securities within the meaning of Section 2(a)(1) of the Securities Act of 1933” (SEC). The safe harbor is conditional: the provider must stay “administrative or ministerial,” and cannot decide “whether, when, or how much” to stake or “guarantee or otherwise set the amount of rewards” (SEC).
Read as a survival test, that guidance rewards exactly the structures our three protocols use: you deposit, the protocol mechanically stakes, and you get a receipt. It warns against the discretionary, yield-guaranteeing models a stressed operator might be tempted to adopt to retain deposits. For US holders it removes one tail risk, the fear that a receipt token is a security in disguise, at a moment when every other risk is elevated. It is also part of why the institutional deposit base has felt comfortable growing into the drawdown.
Institutions Are the New Deposit Base
Follow the record staking number back to its source and you find funds, not retail. The biggest marginal stakers of 2026 have been Ethereum treasury companies and asset managers; reporting around the record cites BitMine and Grayscale among the firms steadily adding staked ETH (Bitcoinist). Spot Ether ETFs, even while seeing price-driven outflows, have normalized the idea of staked-ETH exposure inside a regulated wrapper, a theme we followed as the ETF machine hit its first real bear market.
This reshapes the three-way contest. Institutions want compliance, custody integrations, reporting, and scale, which is precisely the market Lido is building for with stVaults and its institutional push (its institutional share rose to 35.3% in the first half of 2026, per Bitcoin.com). Rocket Pool’s permissionless ethos is a harder sell to a compliance desk, though its decentralization is a genuine draw for those who want it. Frax is not really competing here. If the deposit base of the next cycle is institutional, the protocol that wins institutions wins the decade, and that race currently has one clear front-runner.
So Who Survives the Drawdown?
The bear market does not change which token holds its peg. It changes which business behind the token can keep running. On that measure, the three have separated.
- Lido survives from strength and is using the downturn to diversify: the largest base, the most revenue, an engineering lead on validator efficiency, and a deliberate move beyond staking into Earn and institutional products. Its risk is political, not financial, because it is big enough that Ethereum’s own community worries about its size. Verdict: built to survive, and widening the moat while others retrench.
- Rocket Pool survives because its model is sound and its operators are real, but it is fighting a demand problem its own governance has stopped sugar-coating, with a first-loss token down 98%. Verdict: survives, but 2027 is make-or-break for rETH demand.
- Frax has already survived by leaving. Liquid staking is a legacy product inside a company that has decided its future is stablecoins. Verdict: still there, no longer trying to win.
For a holder, the drawdown clarifies the choice. If you want the deepest liquidity, the widest DeFi acceptance, and a protocol with the resources to keep shipping, stETH is the default, and the bear market reinforced it. If you want maximum decentralization and can accept thinner liquidity and a protocol in the middle of a demand fight, rETH is the principled pick. If you are in frxETH, understand that you hold the smallest receipt of a protocol that has moved on. And whichever you choose, keep the right number in view: staking pays about 2.4% in ETH terms, barely a coupon on a falling asset, so the only reason to lock it up is a conviction that ETH itself comes back.
Frequently Asked Questions
Is my staked ETH safe if the Lido, Rocket Pool, or Frax token crashes?
Largely yes for the receipt itself, with one caveat. Your stETH, rETH, or sfrxETH represents ETH staked by validators, and its value tracks that staked ETH, not the price of the governance token (LDO, RPL, or FRAX). A token crash mainly hurts the protocol as a business: its treasury, its incentives, and, in Rocket Pool’s case, the RPL collateral that sits as a first-loss cushion behind rETH. So a falling token is a slow-burn risk to the protocol’s longevity, not an instant threat to your deposit.
Why is Ethereum staking at a record high while the price is falling?
Staking rewards are paid in ETH, so stakers accumulate more ETH regardless of the dollar price, and the biggest marginal stakers in 2026 have been treasury companies and funds (names like BitMine and Grayscale) that lock coins up for the long term. As of 12 November 2026 about 36% of all ETH, roughly 44.6 million coins, was staked, a record. Analysts note staking also kept rising through the 2022 and 2023 downturns. A high staking rate is not a price floor, though: the coins that still trade set the price.
Did stETH or rETH lose their peg in the 2026 drawdown?
No. Through mid-November 2026 stETH traded within a fraction of a percent of ETH, and rETH and sfrxETH, which are exchange-rate tokens, kept rising against ETH (about 1.18 and 1.17 ETH per token). The exit queue stayed far smaller than the entry queue. The real weak spot is depth: the large on-chain pools that absorb big stETH sells are much thinner than before 2022, so a very large forced seller could still face a discount.
Which is best for a beginner, Lido, Rocket Pool, or Frax?
For most beginners, Lido’s stETH is the path of least resistance: it is the most liquid, the most widely accepted across DeFi and exchanges, and backed by the largest and best-funded protocol. Rocket Pool’s rETH suits those who prioritize decentralization and are comfortable with thinner liquidity. Frax’s frxETH is now a minor product of a project focused on stablecoins, so it is the least suitable default choice. None of this is financial advice, and your tax situation (rebasing versus exchange-rate receipts) can also tip the decision.
What is the difference between Lido Core, Saturn, and North Star?
They are the three protocols’ big 2026 upgrades. Lido Core (its Curated Module v2 migration) consolidated more than 8 million ETH onto Ethereum’s new compounding validators, cutting validator counts to run staking more efficiently. Rocket Pool’s Saturn (One shipped in February 2026, Two was approved in October and reaches mainnet in 2027) cut the node bond to 4 ETH and added a direct rETH redemption queue. Frax’s North Star rebranded its token to FRAX and refocused the whole project on stablecoins, leaving liquid staking as a side product.
By Yuki Tanaka, staking and mining-infrastructure correspondent at HOGE Wire.