The Ethereum staking landscape is undergoing a pivotal transformation, and one of the most anticipated upgrades—Shapella (comprising the Shanghai and Capella upgrades)—is set to unlock a new era of flexibility and decentralization. With over 18 million ETH currently staked—representing roughly 15% of the total supply—investors are preparing for the first time to withdraw their staked assets and rewards. This shift isn’t just technical; it’s reshaping investment strategies across the crypto ecosystem.
Among those reevaluating their positions is Arthur Hayes, co-founder of BitMEX and a well-known macro trader with deep involvement in digital assets. Recently, Hayes revealed that he sold his entire $LDO (Lido DAO token) position at a loss, sparking widespread discussion in the blockchain community.
Hayes had previously been bullish on Ethereum’s transition to proof-of-stake and the growing role of Liquid Staking Derivatives (LSDs). So why exit now?
"I’ve always believed in Ethereum’s long-term potential," Hayes wrote in a recent post. "But I’ve also harbored concerns about Lido’s level of decentralization. When my team invested in non-custodial alternatives like Obol and ether.fi, I knew it was time to move on from $LDO."
The Hidden Risks Behind Lido’s Dominance
Lido Finance remains the dominant player in the LSD space, controlling around 75% of all ETH locked in liquid staking protocols and approximately 30% of total staked ETH. Its success stems from offering users stETH—a liquid derivative that represents staked ETH—allowing them to earn staking yields while maintaining liquidity.
However, as Akshat Vaidya, former Head of Corporate Development at BitMEX and now leading investment research for Hayes’ Maelstrom Fund, explains, Lido’s model carries inherent custodial and operational risks that many investors overlook.
👉 Discover how next-gen staking protocols are eliminating single points of failure
Why “Not Your Keys, Not Your Crypto” Still Applies
Despite branding itself as decentralized, Lido operates under a model where:
- Validators deposit ETH, but node operators contribute no capital—unlike Rocket Pool, where operators co-stake 16 ETH per 32 ETH validator.
- Node operators generate private keys, including BLS and withdrawal keys—not the stakers themselves.
- Withdrawals depend on node operators voluntarily initiating the exit process post-Shapella.
This means users don’t have direct control over their staked assets or rewards. Instead, they rely on a hypothetical asset pool managed by the protocol—similar in risk structure to centralized bridges like Wormhole, which suffered a $320 million hack in 2022.
In essence, Lido functions because node operators choose to cooperate—but there’s no technical guarantee they will.
“If a regulatory action, technical failure, or malicious actor prevents node operators from processing withdrawals,” Vaidya warns, “your staked ETH could become temporarily illiquid or inaccessible.”
And remember: the full withdrawal mechanism has never been stress-tested at scale. April 12, 2023, marks the first real-world trial.
A New Wave of Non-Custodial Staking Solutions
Recognizing these vulnerabilities, Maelstrom Fund has shifted focus toward next-generation staking infrastructure that prioritizes true ownership, security, and decentralization.
Two projects stand out:
1. Obol Labs – Decentralizing Node Operations
Obol introduces Distributed Validator Technology (DVT), an innovative middleware that splits validator keys across multiple nodes using threshold cryptography. This creates a “multisig” validator that runs simultaneously across different machines but behaves as one entity on the Ethereum beacon chain.
Benefits include:
- Elimination of single points of failure
- Increased network resilience against downtime or slashing
- Compatibility with existing staking providers (including Lido)
Rather than competing directly with current services, Obol enhances them—making even centralized or semi-centralized protocols more robust and distributed.
2. ether.fi – True Self-Custody Staking
ether.fi aims to build one of the first fully non-custodial liquid staking protocols for Ethereum. Here’s what sets it apart:
- Users generate and control their own private keys from day one.
- Full autonomy to exit validators and withdraw ETH anytime.
- No reliance on trusted intermediaries for coordination or withdrawals.
This aligns perfectly with the crypto ethos: “Your keys, your crypto.”
With Shapella enabling withdrawals, users now have the freedom to migrate between staking services in pursuit of higher yields or better security. Projects like ether.fi are positioned to capture this demand by offering not just returns—but real ownership.
👉 See how decentralized staking protocols are redefining asset control
Why Arthur Hayes Exited $LDO
Hayes’ decision wasn’t driven by short-term price movements. He acquired roughly 758,000 $LDO tokens at an average price near $2.53, only to sell all holdings in March 2023 at $2.42, accepting a small loss.
His rationale? A structural concern about centralization risk in Lido’s node operator ecosystem.
While Lido has taken steps to improve governance and reduce risks—such as transitioning to a decentralized withdrawal key setup—the underlying model still depends heavily on trusted actors.
As newer, more secure alternatives emerge, Hayes believes capital will naturally flow away from legacy systems toward trustless, user-owned solutions.
“We’re not betting against Ethereum,” Hayes clarified. “We’re betting for better architecture.”
The Future of Ethereum Staking: Trustless & User-Owned
The Shapella upgrade is more than a technical milestone—it's a paradigm shift. For the first time, stakers can exit freely, creating competitive pressure on service providers to offer better terms, transparency, and control.
This opens the door for:
- Greater decentralization across validator networks
- Innovation in yield optimization and risk distribution
- Mass adoption of self-custody models
As Akshat Vaidya puts it:
“Giving up control of your private keys shouldn’t be the price of participation.”
Projects like Obol and ether.fi represent the next evolution—where security, scalability, and sovereignty coexist.
Frequently Asked Questions (FAQ)
Q: Did Arthur Hayes profit from his $LDO investment?
A: While some reports suggest his average purchase price was as low as $2.26–$2.35 (potentially yielding a small gain), Hayes confirmed he sold below cost and accepted a loss due to structural concerns about Lido’s decentralization.
Q: Can Lido users withdraw staked ETH after Shapella?
A: Yes. After the Shapella upgrade on April 12, 2023, users can request withdrawals. However, processing occurs gradually—up to 6,500 validators per epoch (~22,000 ETH/day).
Q: What makes Obol’s DVT technology different?
A: DVT allows multiple parties to run a single validator together without any one party having full control. This improves uptime and reduces reliance on individual node operators.
Q: Is ether.fi live yet?
A: ether.fi launched its testnet in early 2023 and plans mainnet deployment shortly after Shapella. It emphasizes full user custody throughout the staking lifecycle.
Q: Are all LSDs risky like Lido?
A: Not all. While many early LSDs rely on centralized or semi-trusted node operators, new protocols are emerging with non-custodial designs that give users full control over keys and withdrawals.
The era of passive acceptance is ending. In a world where true ownership matters, the next wave of Ethereum staking won’t just offer yield—it will return power to the user.
👉 Explore how you can stake ETH without surrendering control