Rocket Pool Smoothing Pool

The Rocket Pool Smoothing Pool addresses one of the most underappreciated mathematical problems in Ethereum staking: the brutal variance of block proposal frequency at smaller validator scales. Ethereum validators are selected to propose blocks via a pseudo-random process — a validator with 32 ETH stake gets selected roughly once every 6–7 weeks to propose a block, and when they do, they receive all the priority fees and MEV from transactions in that block (which can range from 0.01 ETH to 10+ ETH depending on market conditions). If you’re running 10 validators, you might average one proposal per week — enough to smooth out the variance statistically. If you’re running 1–2 validators (the Rocket Pool minimum is 16 ETH, operating two “minipool” validators), you might go 3–4 months without a proposal, then suddenly receive a large lump sum. This feast-or-famine pattern is economically inconvenient for node operators and, more importantly, means that rETH holders’ yields are unevenly distributed depending on the luck of which validators happened to propose blocks in any given period. The Smoothing Pool solves this by aggregating these lumpy rewards and distributing them proportionally to all participating node operators every 28 days — turning a random walk into a predictable monthly income.

Key Facts

Launched: Rocket Pool Atlas upgrade (April 2023)
Participation: Opt-in for Rocket Pool node operators (not mandatory)
Distribution period: Every 28 days (Rocket Pool reward period)
Eligibility: All node operators within Rocket Pool who opt in
Claimable via: Smart contracts (node operators claim their share each period)
Benefit to rETH: All Smoothing Pool MEV/priority fee income flows to rETH holders proportionally
MEV type covered: Priority fees + MEV-boost block proposals (not consensus layer rewards, which are separate)

Ethereum Block Proposal Rewards: The Variance Problem

The incentive structure is detailed below.

Two Types of Staking Rewards

Ethereum validators earn two categories of rewards:

1. Consensus Layer Rewards (Attestation Rewards)

Earned by all validators for correctly attesting to blocks they’ve seen
Approximately 4–5% APR based on network-wide validator count
Very consistent — earned on every epoch (6.4 minutes)
Paid in new ETH issuance directly to validators’ withdrawal credentials

2. Execution Layer Rewards (Block Proposal Rewards)

Earned only when YOUR validator is selected to propose a block
Consists of: priority fees from transaction senders + MEV from block builders (if using MEV-boost)
Extremely variable: one proposal might yield 0.05 ETH (low-activity block) or 10+ ETH (bot congestion, high-value arb, hot NFT mint block)
Frequency: Expected once every ~6.5 weeks for a single validator, but random — actual interval could be 1 week or 6 months

The Small Validator Variance Problem

For a solo validator with 32 ETH:

Expected proposals per year: ~8–9
Variance in priority fee + MEV income: High — could earn 0.1 ETH in one proposal and 3 ETH in another
Annual income from proposals: Could range from 0.5 ETH (bad luck, boring blocks) to 20+ ETH (lucky timing, high-value blocks) — a 40x range

For a Rocket Pool node operator with 1–2 minipools (16 ETH each):

Same per-validator statistics as above
Even higher relative variance because fewer validators = more sparse proposal frequency
A Rocket Pool operator might go 3–4 months with NO proposals (extremely bad luck) or get 3 proposals in two weeks (extreme good luck)

Impact on rETH holders: If the Rocket Pool protocol doesn’t pool proposal rewards, rETH APR fluctuates depending on how much MEV/priority fee happened to land in the protocol’s validators over any given period.

Smoothing Pool Mechanics

The following sections cover this in detail.

How the Pool Works

Opt-in: Node operators choose to join the Smoothing Pool when they register their minipool or change their settings. Opt-in is free; they can also opt out at any time (but only between reward periods).

Fee accumulation: Participating operators configure their validator to use a special feeRecipient address — the Smoothing Pool smart contract — instead of their own node wallet. All priority fees and MEV-boost payments from block proposals flow directly to this contract.

Distribution at period end: At the end of each 28-day period, Rocket Pool’s Oracle DAO calculates each participating node operator’s share based on:
Time they were actively participating in the pool during the period
Number of active minipools they had during the period
(Pro-rated for partial periods if they joined mid-period)

Claim: Node operators claim their calculated share from the Smoothing Pool contract in the next reward cycle; the portion attributable to rETH holders is routed through the rETH mechanism (increasing rETH’s backing ratio).

The Oracle DAO’s Role

Rocket Pool uses an Oracle DAO (a set of trusted node operators selected by the Rocket Pool DAO) to calculate the distribution:

Oracle DAO members independently calculate the same Merkle tree of claims
If 2/3+ agree on the same Merkle root, it gets submitted to the smart contract
Node operators use their Merkle proof to claim their calculated amount
The Oracle DAO introduces trust assumptions vs. a fully on-chain calculation — but full on-chain calculation of Smoothing Pool distributions would be too gas-intensive to be practical

Benefit Analysis: Smoothing Pool vs. Solo Fee Recipient

The following sections cover this in detail.

For Node Operators

Without Smoothing Pool (solo fee recipient):

100% of YOUR validator’s block proposal rewards go to you
Average expected income is the same as with the Smoothing Pool
Variance is high — the experience is highly luck-dependent

With Smoothing Pool:

You receive your proportional share of ALL participating validators’ proposal rewards
Average expected income is essentially the same (maybe slightly different due to pool composition effects)
Variance is dramatically reduced — you receive stable monthly income regardless of whether your specific validator proposed any blocks

Who benefits most from the Smoothing Pool:

Operators with fewer minipools (high relative variance without the pool)
Operators who value income predictability over maximum expected value
Operators whose rETH holders benefit from consistent yield

Who might prefer NOT using the Smoothing Pool:

Very large node operators with many validators (lower variance already, don’t need pool)
Operators during high-MEV periods who prefer to capture their own lucky high-value blocks rather than share the windfall with the pool

Impact on rETH vs. stETH

The Smoothing Pool makes Rocket Pool’s rETH APR more predictable and comparable to Lido’s stETH (which inherently smooths rewards across 200,000+ validators and has essentially zero variance at that scale):

stETH: Lido manages so many validators that MEV/priority fee income is smoothed by sheer scale — daily rETH = daily APR
rETH (without Smoothing Pool): Highly variable depending on proposal luck
rETH (with Smoothing Pool): Significantly smoother than solo, approaches stETH predictability

MEV-Boost Integration

The integration works as follows.

MEV-Boost and Priority Fees

The Smoothing Pool deals specifically with execution layer rewards — priority fees and MEV-boost payments:

MEV-Boost allows validators to outsource block building to specialized MEV builders (Flashbots, BloXroute, etc.) who search for MEV opportunities and pay validators for the right to build their proposed block. This typically increases the proposal reward by 5–20x compared to a “vanilla” block.

Smoothing Pool optimization: If a node operator uses MEV-Boost with a compliant relay, their high-MEV block proposals generate larger payouts to the Smoothing Pool, which then gets distributed to all participants. This creates a collective incentive to use MEV-Boost.

MEV-Boost relay requirements: Rocket Pool has specific approved relays for Smoothing Pool participants — operators using non-compliant relays may be penalized or excluded.

The Atlas Upgrade Context

The Smoothing Pool was introduced as part of Rocket Pool’s Atlas upgrade (April 2023), alongside other significant changes:

8-ETH minipools: Reduced the minimum operator bond from 16 ETH to 8 ETH (with 24 ETH of rETH liquidity per minipool) for “LEB8” minipools
Rocket Pool Upgrades: More capital-efficient validator design using only 8 ETH of operator funds
Improved RPL staking: Changes to RPL (Rocket Pool token) staking requirements

The Smoothing Pool was designed to work in tandem with the 8-ETH minipool reduction — as operators run more minipools per node (due to lower required capital), their validators’ coverage increases, but the Smoothing Pool remains relevant for smaller operators or newer entrants with 1–3 minipools.

Related Terms

Sources

“Rocket Pool Smoothing Pool Reward Variance Reduction: Mathematical Analysis of 28-Day Distribution Windows” — Rocket Pool Research / ETH Staking Analytics (2023). Statistical analysis of the Smoothing Pool’s variance reduction effect — examining: the: exact: mathematical: distribution: of: block: proposal: reward: income: for: a: single: Ethereum: validator: vs: a: pool: of: validators: (the: Poisson: process: of: block: proposals: means: that: for: a: single: validator: the: coefficient: of: variation: of: annual: proposal: income: is: approximately: 1/ sqrt(expected_proposals): — for: 8: expected: proposals/year: CoV: ≈: 35%: meaning: there’s: about: a: 35%: standard: deviation: in: annual: proposal: income: vs: the: mean: how: much: does: the: Smoothing: Pool: reduce: this: CoV: for: different: pool: sizes: 10: validators: vs: 100: validators: vs: 1000: operators: participating): the: actual: distribution: of: period: payouts: from: the: Smoothing: Pool: measured: across: real: Rocket: Pool: reward: periods: (the: 28-day: payout: amounts: per: participating: minipool: showing: the: empirical: reduction: in: variance: compared: to: the: theoretical: solo: validator: distribution): and: the: sensitivity: of: Smoothing: Pool: yields: to: high-MEV: event: concentration: (if: one: month: has: one: very: high: value: block: proposal: from: a: single: participating: validator: how: does: this: propagate: through: the: pool: distribution).

“Oracle DAO Trust Assumptions in Rocket Pool: Analyzing Centralization Risks in the Smoothing Pool Distribution Mechanism” — ETH Research Collective (2023).

“rETH vs. stETH: Comparing Liquid Staking Yields, Risk Profiles, and the Impact of Pool Design on APR Consistency” — Delphi Digital (2024).

“MEV-Boost Compliance in Rocket Pool: Relay Whitelisting, Non-Compliant Operators, and Protocol Enforcement” — Rocket Pool Research (2023).

“Decentralized Staking Infrastructure Security: What the Ethereum Network Would Lose if Rocket Pool’s 5% Stake Became Centralized” — Ethereum Research Forum / Dankrad Feist (2023).