MEV extraction is moving off-chain. On-chain auctions like Flashbots Auction are transparent but slow, creating a latency arms race. Private mempools and order flow auctions now dominate, shifting value capture to pre-chain coordination.
security-post-mortems-hacks-and-exploits
Blog
The Future of MEV Capture: Will It Move Completely Off-Chain?
An analysis of how intent-based architectures and off-chain order flow auctions are fundamentally reshaping the MEV landscape, moving value extraction away from public mempools and creating new security paradigms.
introduction
THE INEVITABLE SHIFT
Introduction
MEV capture is migrating off-chain, transforming from a public auction into a private negotiation between sophisticated infrastructure.
The endgame is intent-based architectures. Users express desired outcomes, not transactions. Solvers like UniswapX and CowSwap compete off-chain, internalizing MEV as a cost of execution, not a public extractable surplus.
Blockchains become settlement layers. Execution complexity migrates to specialized networks like Anoma or SUAVE. The chain verifies results, while the economic activity and its associated value occur in a separate, faster layer.
Evidence: Over 90% of Ethereum blocks are now built by builders using private order flow, and intent-based DEX aggregators already process billions in volume, proving the model works.
key-trends
THE ON-CHAIN VS. OFF-CHAIN BATTLEGROUND
Executive Summary
The fight over MEV is shifting from public mempools to private order flow networks, redefining value capture in blockchains.
01
The Problem: Public Mempools Are a Free-for-All
Open mempools expose user intent, creating a $1B+ annual market for predatory arbitrage and sandwich attacks. This results in:\n- Worse execution prices for end users\n- Network congestion from failed front-running txns\n- Centralization pressure on block builders
$1B+
Annual MEV
>90%
Searcher Profit
02
The Solution: Private Order Flow to Builders
Protocols like Flashbots Protect, BloXroute, and Titan allow users/wallets to route transactions directly to trusted builders. This shifts capture from searchers to builders and users via:\n- Guaranteed inclusion without front-running\n- Revenue sharing back to applications (e.g., Uniswap) \n- Reduced gas auctions from removed competition
~40%
Ethereum Flow
>95%
Top Builder Share
03
The Endgame: Intents and Solving
Fully off-chain MEV capture moves to an intent-centric model, as pioneered by UniswapX, CowSwap, and Across. Users submit desired outcomes, not transactions.\n- Solvers compete off-chain for best execution\n- MEV is internalized as solver profit\n- Cross-chain intents become native (see LayerZero, Chainlink CCIP)
$10B+
Intent Volume
~500ms
Solver Latency
04
The Counterforce: SUAVE and Decentralization
Ethereum's SUAVE chain aims to keep MEV markets credibly neutral and decentralized. It proposes a universal mempool and decentralized block builder network.\n- Prevents exclusive order flow deals\n- Standardizes cross-chain MEV auctions\n- Threatens the business model of private relay cartels
1 Chain
Universal Mempool
TBD
Builder Decentralization
05
The Reality: Vertical Integration Wins (For Now)
Entities controlling key vectors—wallet (MetaMask), RPC (Alchemy), relay (Flashbots)—are integrating to capture the full stack. This creates:\n- Unbeatable data advantages for in-house searchers\n- Sticky user bases through default settings\n- Regulatory risk as centralized MEV hubs emerge
3 Stack
Key Vectors
Default
User Choice
06
The Verdict: Hybrid, Not Binary
MEV capture will not move completely off-chain. A hybrid model will dominate: private order flow for retail, public auctions for niche/long-tail MEV, and intents for complex DeFi. The real battle is over who owns the routing decision and captures the rent.
Hybrid
Future State
Routing
Key Battleground
thesis-statement
THE OFF-CHAIN SHIFT
Thesis: MEV is Not Disappearing, It's Privatizing
The capture of Maximal Extractable Value is migrating from public mempools to private, off-chain execution venues controlled by sophisticated operators.
Public mempools are obsolete for high-value transactions. Frontrunning and sandwich attacks in transparent pools force users and protocols to seek private order flow. This drives MEV capture into off-chain systems like Flashbots Protect, CoW Swap, and private RPC endpoints.
Intent-based architectures privatize execution. Protocols like UniswapX and Across shift complexity off-chain by having users submit outcome-focused intents. Solvers compete privately in off-chain auctions, internalizing MEV that was once public arbitrage.
The endpoint is the new battleground. Wallet providers and applications like MetaMask and Rabby integrate private transaction routing by default. This consolidates order flow into a few centralized relays, creating off-chain MEV cartels with opaque revenue.
Evidence: Over 90% of Ethereum block space is built by builders using MEV-Boost, with Flashbots capturing the majority of this private order flow. This proves the privatization is complete at the block production layer.
market-context
THE INCENTIVE MISMATCH
The On-Chain MEV Death Spiral
The economic design of public blockchains creates a self-reinforcing loop where maximal extractable value (MEV) erodes the core value proposition of on-chain execution.
On-chain MEV is a tax on honest users, creating a direct conflict between searcher profit and network utility. Every dollar extracted via sandwich attacks or DEX arbitrage is a dollar lost from a retail trader's slippage tolerance or a liquidity provider's yield.
The death spiral accelerates as sophisticated MEV tooling like Flashbots' SUAVE and bloXroute's relays commoditize extraction. This pushes protocol developers to build off-chain or in private mempools, fragmenting liquidity and reducing the public chain's value as a settlement layer.
The counter-force is intent-based architectures like UniswapX and CowSwap, which shift execution logic off-chain. These systems batch and solve user intents via off-chain solvers, moving the MEV competition to a dedicated layer and returning a guaranteed price to the user.
Evidence: Over 70% of Ethereum blocks are now built by MEV-Boost relays, proving the centralization pressure. Protocols like dYdX have already migrated core matching engine logic to a centralized off-chain sequencer to eliminate on-chain frontrunning.
THE FUTURE OF MEV CAPTURE
On-Chain vs. Off-Chain MEV: A Protocol Comparison
Compares the architectural and economic trade-offs between on-chain and off-chain MEV capture models, analyzing their viability for the future.
| Feature / Metric | On-Chain (e.g., MEV Auctions, PBS) | Hybrid (e.g., SUAVE, UniswapX) | Fully Off-Chain (e.g., Private Order Flows) |
|---|---|---|---|
Primary Execution Venue | Public Mempool | Specialized Searchers / Solvers | Private Channels / Dark Pools |
Frontrunning Resistance | |||
Extractable Value Redistribution |
| Variable (Solver/User/Protocol) |
|
Finality to User (Typical) | 12-30 seconds | 1-5 seconds | < 1 second |
Protocol-Level Revenue (e.g., MEV-Burn) | |||
Requires Trusted Third Party | |||
Integration Complexity for dApps | Low (Native) | High (SDK/Intent Integration) | Very High (Exclusive Partnerships) |
Current Market Share of MEV Flow | ~65% | ~25% (Growing) | ~10% |
deep-dive
THE OFFSHORE SHIFT
The Security Post-Mortem of a Privatized Future
The economic and security logic of MEV extraction will push its most valuable components into private, off-chain systems.
The most valuable MEV will move off-chain. Complex strategies like cross-domain arbitrage and JIT liquidity require data and execution speed impossible on public mempools. This creates a private order flow market where searchers and block builders transact directly.
Public chains become settlement layers. On-chain execution is the slow, expensive, and transparent final step. The real competition shifts to off-chain coordination between entities like Flashbots SUAVE, bloXroute, and proprietary builder networks.
This privatizes systemic risk. Security audits become impossible when the critical auction logic for blocks exists in private mempools and relay networks. A bug in Flashbots' MEV-Share or a centralized relay becomes a single point of failure for chain liveness.
Evidence: Over 90% of Ethereum blocks are now built by a handful of entities using MEV-Boost, demonstrating the centralizing pressure of off-chain coordination. The next step is moving the auction logic itself off-chain.
protocol-spotlight
THE FUTURE OF MEV CAPTURE
Architect Spotlight: Who's Building the Off-Chain Stack
The battle for MEV is shifting from public mempools to private off-chain networks, redefining value flow and protocol architecture.
01
The Problem: Public Mempools Are a Free-for-All
Public mempools broadcast user intent, creating a toxic, zero-sum game of frontrunning and sandwich attacks. This extracts ~$1B+ annually from users and creates systemic instability.\n- Value Leakage: Searchers and validators capture most value.\n- User Harm: Failed transactions and degraded slippage are common.\n- Inefficiency: Network congestion is artificially inflated.
$1B+
Annual Extract
>50%
Failed TXs
02
The Solution: Private Order Flow Auctions (OFAs)
Protocols like UniswapX and CowSwap route user orders off-chain to a competitive network of solvers. This flips the model: searchers now compete to give users the best price, not exploit them.\n- Value Redistribution: MEV is internalized as better execution for the user.\n- Intent-Based: Users specify what they want, not how to do it.\n- Privacy: No frontrunning on visible transaction paths.
~99%
Fill Rate
$10B+
Volume
03
The Enforcer: SUAVE by Flashbots
SUAVE is a dedicated blockchain attempting to become the preferred mempool and decentralized block builder for all chains. It aims to commoditize the MEV supply chain.\n- Universal Preference: A single, neutral venue for order flow.\n- Decentralized Censorship Resistance: Mitigates the centralization risks of OFAs.\n- Modular Design: Separates expression, execution, and settlement layers.
0
Public Mempool
All
Chain Agnostic
04
The Integrator: Cross-Chain Intents & Bridges
The off-chain stack extends to cross-chain. Across and LayerZero's DVN model use intents and competitive solvers to optimize bridging, capturing cross-chain MEV.\n- Unified Liquidity: Solvers tap into the best liquidity source across chains.\n- Optimal Routing: Finds the fastest/cheapest path, not just a single bridge.\n- Atomic Composability: Enables complex cross-chain arbitrage as a service.
~500ms
Quote Latency
30%+
Cost Savings
05
The Risk: Centralization of the Off-Chain Layer
The shift to off-chain execution creates new choke points. Dominant solver networks or OFA aggregators could become the new rent-extracting intermediaries.\n- Solver Cartels: Reduced competition leads to worse prices.\n- Protocol Capture: A single intent network could dictate terms to L1s/L2s.\n- Regulatory Attack Surface: Centralized order flow is a clear target.
1-3
Dominant Solvers
High
Regulatory Risk
06
The Endgame: MEV as a Protocol Revenue Stream
Forward-thinking L1/L2s like Canto and Aevo are building native intent-based systems to capture and redistribute MEV at the protocol level. This turns a parasitic externality into a sustainable public good.\n- Protocol-Owned Liquidity: MEV funds treasury or staker rewards.\n- Subsidized Transactions: Negative fee transactions become possible.\n- Aligned Incentives: Value accrues to the network, not third parties.
100%
Internalized
Protocol
Value Accrual
counter-argument
THE MEV FRONTIER
The Bull Case for On-Chain Resilience
The long-term value accrual for blockchains depends on capturing MEV on-chain, not outsourcing it to off-chain networks.
MEV will not move off-chain. The economic security of a blockchain is its ability to monetize its ordering rights. Projects like Flashbots' SUAVE aim to bring auction logic on-chain, ensuring value is captured by the protocol and its validators, not external searchers.
On-chain auctions create composable revenue. Protocols like EigenLayer and Espresso Systems are building infrastructure for shared sequencers and decentralized proposers. This creates a verifiable revenue stream that can be restaked or used to subsidize user transactions, unlike opaque off-chain deals.
Intent-based architectures are the compromise. Systems like UniswapX and CowSwap abstract complexity for users but ultimately settle and pay fees on-chain. They demonstrate that user experience and on-chain value capture are not mutually exclusive.
Evidence: Ethereum's PBS (proposer-builder separation) framework is a canonical example. It formalizes the builder role on-chain, creating a transparent market for block space that directly funds validator staking yields and protocol treasury.
risk-analysis
MEV INFRASTRUCTURE RISKS
The New Risk Matrix: What Could Go Wrong Off-Chain?
As MEV capture shifts to off-chain systems like private mempools and intents, new systemic and operational risks emerge that are not present in public mempool models.
01
The Centralization of Censorship
Private order flow auctions (OFAs) like Flashbots Protect and BloXroute's MEV-Share route user transactions through a handful of centralized relays. This creates a single point of failure and control, enabling regulatory pressure and transaction blacklisting at the network layer. The decentralization of block building is undermined if order flow is monopolized.
>90%
OF Flow to Top 3
1
Gov't Subpoena Away
02
Solver Cartels & Collusion
Intent-based architectures (e.g., UniswapX, CowSwap) rely on off-chain solvers to fulfill user intents. A small group of well-capitalized solvers can form a cartel to suppress competition, extract maximal value, and censor transactions. This recreates the miner extractable value (MEV) problem it aims to solve, but in a less transparent, off-chain setting.
~5
Dominant Solvers
0%
On-Chain Proof
03
The Oracle Manipulation Attack
Cross-chain intents and generalized bridging (e.g., Across, LayerZero) depend on off-chain agents and oracles for state verification. A compromised or malicious oracle can falsify settlement proofs, leading to double-spends and fund theft across chains. The security collapses to the weakest link in the off-chain attestation network.
$2B+
Bridge Hack Volume
Minutes
To Finality Fraud
04
Liquidity Fragmentation Death Spiral
As MEV capture moves to private channels, the public mempool becomes a toxic dump of arbitrage leftovers. This reduces base-layer revenue for validators, potentially threatening chain security. It also creates a two-tier system where users unaware of private channels get consistently worse execution, degrading the public good of fair settlement.
-40%
Pub. Mempool Value
↑ 300%
Slippage for Normies
05
Regulatory Capture of the MEV Supply Chain
Off-chain MEV entities (relays, builders, solvers) are licensed businesses, not permissionless protocols. This makes them direct targets for KYC/AML regulations and securities laws. A regulator could mandate transaction filtering or seize control of the supply chain, effectively nationalizing a critical piece of blockchain infrastructure.
100%
OFAC Compliant
0
Decentralized Relays
06
The Time-Bandit Reorg Risk
Proposer-Builder Separation (PBS) with off-chain builders introduces the risk of time-bandit attacks. If a builder discovers a highly profitable MEV opportunity after a block is proposed, they have an incentive to collude with future proposers to reorg the chain and steal the value. This attacks the very finality of the chain for off-chain profit.
7+
Blocks Deep Risk
$100M+
Attack Incentive
future-outlook
THE HYBRID REALITY
Outlook: A Hybrid, Asymmetric Battlefield
MEV capture will not move completely off-chain; it will fragment into a hybrid model where on-chain and off-chain strategies compete asymmetrically.
On-chain searchers dominate complexity. They will retain control over complex, high-value strategies like multi-block DEX arbitrage and NFT floor sweeping, where latency and execution logic are paramount.
Off-chain solvers capture commoditized flow. Protocols like UniswapX and CowSwap will route simple, high-volume swaps to off-chain solvers, abstracting MEV from end-users through intents.
The battlefield is asymmetric. On-chain competition is a symmetric speed race. Off-chain competition, led by Across and SUAVE, is asymmetric, favoring capital and order flow aggregation over pure latency.
Evidence: Over 70% of Ethereum blocks contain MEV, but intent-based volumes on UniswapX are growing 15% MoY, proving both models are scaling simultaneously.
takeaways
THE MEV ENDGAME
TL;DR for Builders and Investors
The battle for MEV is shifting from public mempools to private order flow networks, redefining value capture and protocol architecture.
01
The Problem: Public Mempools Are a Free-for-All
On-chain auctions like Ethereum's mempool are inefficient and extractive. They leak user intent, create network congestion, and allow searchers to capture the majority of value.\n- ~$1B+ in MEV extracted annually from public mempools.\n- Front-running & sandwich attacks degrade user experience and trust.\n- Inefficient price discovery leads to suboptimal execution for traders.
~$1B+
Annual Extract
>90%
Searcher Profit
02
The Solution: Private Order Flow is the New Moat
Protocols and applications will capture MEV by routing user transactions off-chain first. This is the core thesis behind UniswapX, CowSwap, and wallet integrations like MetaMask.\n- Intent-based architectures let users specify outcomes, not transactions.\n- Off-chain auctions between professional solvers (e.g., Across, 1inch Fusion) find optimal execution.\n- Revenue sharing models allow dApps and wallets to monetize their user flow directly.
10-100x
More Efficient
0%
Sandwich Risk
03
The Infrastructure: Specialized MEV Chains & Co-Processors
The future is a dedicated off-chain execution layer. Think EigenLayer's shared sequencer, Espresso Systems, or app-specific rollups with integrated MEV capture.\n- Purpose-built chains (e.g., for dYdX, Aevo) internalize MEV as protocol revenue.\n- Proposer-Builder Separation (PBS) on Ethereum formalizes the market, but value accrues to block builders, not users.\n- Cross-chain intent networks like LayerZero's OFT and Circle's CCTP will become major MEV vectors.
$10B+
TVL at Stake
<500ms
Auction Latency
04
The Investment Thesis: Own the Flow, Own the Future
Value will accrue to entities that aggregate and route user intent, not just those who execute it. This reshapes the stack.\n- Wallets & Frontends become critical gatekeepers with new monetization paths.\n- Solver networks (e.g., CowSwap solvers, UniswapX fillers) are high-margin B2B services.\n- Builders (e.g., Flashbots SUAVE, Jito Labs) evolve into essential, vertically-integrated infrastructure.
100x
Upside for Aggregators
New Rev Stream
For dApps
ENQUIRY
Get In Touch
today.
Our experts will offer a free quote and a 30min call to discuss your project.
NDA Protected
Confidentiality24h Response
Time to ValueDirectly to Engineering Team
No Sales Fluff10+
Protocols Shipped
$20M+
TVL Overall