Detailed explanation of MEV principles, latest trends and evaluation methods

Author: Patrick Mccorry; Compiler: Huohuo/Vernacular Blockchain

1. Related concepts

Let's start with the definition of the MEV abbreviation:

**MEV (Maximizing Extractable Value) Maximum Extractable Value: **During the block production process, the maximum value that an agent can extract by including, excluding, or changing the order of transactions.

More specifically, the concept of Miner Extractable Value (MEV) refers to a scenario where an agent examines a user’s recent transactions, develops a strategy to profit from them, and subsequently implements that strategy to capture any potential gains.

Make money by interfering with the execution of pending transactions This very simple act can have huge consequences for users, DeFi protocols, and the underlying blockchain network.

Before getting into the core discussion, we'll give you some background information on MEV - What are the ethics of MEV? Both in the context of agents exploiting MEV and actors trying to defend against it.

1) Agents in MEV game

In addition to users seeking to participate in smart contracts, there are two key roles closely related to the MEV concept:

• **Searcher. **The agent discovers an opportunity to profit from user transactions, creates a bundle of transactions to exploit it, and proposes the bundle to the proposer.

• **Proposer. **Agent with the authority to determine the order of transactions.

Searchers can be trading companies with extensive expertise or hobbyists coding in their bedrooms.

Searcher is a permissionless role.

**The only obstacle is the searcher's ability to find alpha, build competitive MEV bots, and exploit the opportunity. **Access to capital will help, but it’s not yet a significant hurdle.

On the other hand, the proposer has the right to decide the order of transactions and thus the execution results of the transactions. This key role can be filled by various entities, including miners (in a proof-of-work system), stakers (in a proof-of-stake system), or sequencers (in a rollup).

Although the proposer pool is usually restricted, it can be open.

There are compelling reasons to impose restrictions on who can assume the role of proposer:

1. Consensus Protocol: Many blockchain systems require proposers to participate in a round-based protocol, requiring the cooperation of a majority (or an overwhelming majority) of proposers in each round. Coordinating communication among all N participants often proves to be a bottleneck.

2. Transaction routing: Users must have a reliable way to transmit transactions to the proposer. This can be achieved by forwarding transactions to a public mempool or directly to a designated proposer.

3. Verifiable Integrity: The wider community may seek an objective metric to verify that all proposers work together as a collective to decide the order of transactions, and that they always extend the latest order. For example, the network could implement a fork-choice rule where proposers build on top of the heaviest chain (stake/work).

4. Risks of MEV: In some blockchain systems, especially most of today's Rollups, proposers are given the responsibility not to take advantage of MEV opportunities. Therefore, it is crucial that communities have the right to decide who to trust.

Simply put, there must be a mechanism that allows users to independently verify that the proposer has the authority to determine the ordering of recent transactions. Without this guarantee, malicious actors could flood the system with fake trading orders, leaving users unable to discern the truth.

To keep the explanation of agency simple, we incorporate the role of builder into the proposer and assume that the proposer both builds a block and has the authority to publish it.

2) Agent interactions and relationships

What is the interaction between searchers and proposers?

**A focus of MEV research is understanding the interaction between seekers and proposers. **In addition, it must be determined whether these roles can be fulfilled by the same entity, or whether different agents are required:

1. **Same broker. **The searcher can be a proposer in the system,

2. **Separate agent. **There is one (or more) seeker(s) who are not proposers, and they are all competing to influence the proposer.

In other words, it is important to determine whether searchers have complete and undisputed control over the deal ranking strategy. **If the searcher is also a proposer, then it may grant the searcher additional power to observe the strategies employed by other searchers, allowing them to steal opportunities from more competitive searchers. **

On the other hand, if searchers cannot become proposers or collude with proposers, then it allows us to assume an environment in which searchers must compete with each other. Their goal is to influence proposers and convince them to rank the deal list according to the preferences of the winning searcher.

Outsourcing vs. Monopoly in MEV Profit Maximization: There is an interesting debate whether it is more profitable for the proposers to subcontract the task of identifying and exploiting MEVs to an open market of searchers, or whether they Should it be aimed at monopolistic means.

We assume that proposers will act as honest parties and adhere to their committed transaction ordering policies. Furthermore, searchers and proposers are always independent agents.

Our focus is on understanding the strategies searchers use to influence proposers' ranking strategies and hopefully beat all other competitors to the same opportunity.

3) Transaction ordering policy

MEV’s pursuit and defense focuses on the searcher’s ability to influence individual components of a blockchain system:

• **Transaction ordering policy. **Given a list of pending user transactions, which sorting algorithm is applied to output the final list of sorted transactions (i.e. confirmed blocks).

Blockchain systems can implement a variety of ordering strategies, with the goal of providing fairness to all users who may want to conduct transactions.

**This begs the question: **What is the definition of fairness?

• Should users pay the same fees and transact on a first-come, first-served basis?

-Should users pay fees based on the priority of transactions and have all transactions sorted based on the fees paid?

Both cases follow the general principle that users can conduct transactions as long as they have the ability to pay. **It does not stipulate that the user's transaction will occupy the promised position in the total sorting, but that it will eventually be sorted and executed in time.

**This concept of fairness is interesting and is fundamental to the censorship resistance of blockchain networks. **

It outlines that users’ ability to transact should be determined solely by their ability to pay, and that they should not be discriminated against based on geography, identity, gender or belief system. It originated in the Bitcoin space and can be easily applied since the network only supports payments.

However, the ability to guarantee transaction inclusion falls short when we try to understand fairness in smart contract-enabled systems. For networks like Ethereum, we must expand the scope of fairness beyond incorporating transactions into global ordering. It should also consider the intent of the user who signed the transaction and whether the user's desired outcome was achieved after the transaction was executed.

**Recognize the important role of user intent in fairness. ** Users measure fairness not only by the ability to incorporate transactions in a timely manner, but also by assessing the actual outcome of the transaction and whether it meets their original expectations when signing the transaction.

This can lead to a new and interesting definition of what we mean by censorship:

• **Weak censorship resistance. **As long as users are willing to pay the appropriate fees, they can order execution transactions at any time.

• **Powerful censorship resistance. ** Users can force the desired outcome of a transaction and they only need to pay the appropriate fees.

Keep this in mind as it will become important as we learn how MEV can be exploited to interfere with a user's transactions and force their execution to fail. Therefore, even if the user's transactions can be forced to be included in the overall ranking, the user's desired result (intention) cannot be achieved.

As far as we know, if we want to build a system that is robust to censorship, ranking strategies must prevent searchers from selectively interfering with users' ability to transact. This remains an open research question.

Enforcing Sanctions Through Relays OFAC is actively testing whether blockchain networks can continue to treat users fairly based on their ability to be included in payment transactions.

2. Activation of MEV

To delve deeper into the technical aspects of Miner Extractable Value (MEV), we must examine the following:

1. **Position MEV Opportunities: **Understand how searchers discover users’ recent transactions in the blockchain system.

2. **Execution Environment: **Check the technical environment for all transaction executions.

3. Exploit Strategies: Research the various strategies searchers can use to exploit MEV opportunities, such as trade reordering, front-running, and arbitrage.

4. Influence ranking: Explore how searchers influence proposers to prioritize their MEV related transactions.

Once we have a firm grasp of these basic components, we can move on to evaluating the ethical implications and moral considerations of MEV.

1) Looking for MEV opportunities

Searchers need to access recent user transactions to find new MEV and money-making opportunities.

There are two ways to find transactions:

• **Gossip Protocol. **Users submit their transactions to the peer-to-peer network and the transactions are propagated to all nodes in a very fast timeframe (<1 second).

• **Proposer feed. ** Proposer posts pending and/or recently ordered transactions.

Most users send transactions on the gossip protocol in the hope that proposers will discover their transactions and include them in their blocks. Meanwhile, anyone, including searchers, can join the gossip protocol and listen for pending transactions.

It has led to the 'Dark Forest' nickname as searchers are almost guaranteed to find a user's transactions and interfere with their execution if a money-making opportunity exists. **For example, in the Dark Forest post, the author failed to recover the funds at risk while the searchers discovered their trade, evaluated it, and collected funds for themselves. **

So far, the only way to defeat the Dark Forest is to avoid sending transactions to the peer-to-peer network. In a subsequent post, the author escapes the dark forest by sending transactions directly to Ethereum miners. This, along with other instances, eventually led to Flashbot offering a direct transaction feature, allowing users to send transactions directly to trusted miners (as a service).

If the blockchain undergoes a reorganization and a user's transaction is temporarily unconfirmed and placed in the mempool, there is still a risk of MEV bots exploiting direct transactions. However, reorganization events are relatively rare in Ethereum Proof-of-Stake compared to 7% of all blocks in PoW Ethereum.

**The same risk does not apply to Rollup (as implemented today). Almost all transactions are direct transactions because the user has a direct communication connection with the proposer (sequencer). **Searchers have little opportunity to eavesdrop on the channel, which significantly increases the difficulty of exploiting MEV opportunities for pending transactions.

This has led to the belief that Rollup has defeated searchers. **Any success thus far depends on the credibility of the proposer and not on exploiting MEV for personal gain. **Of course, that's not the whole story, searchers can still find MEV opportunities.

**In Rollup, due to direct trades, searchers turn their attention to finding recently confirmed trades in the hope of finding arbitrage-like opportunities. **

For example, in Arbitrum, proposers maintain a feed that publishes recently ordered transactions. It is published every 250 milliseconds, mainly to help infrastructure providers such as Infura and Etherscan obtain the latest data. This allows users to send transactions to Sequencer and then check their status on Etherscan. Additionally, it allows anyone to run an Arbitrum node with Sequencer confirmed status.

Unfortunately, this feed was discovered by MEV bots. Searchers will connect to the source and take advantage of arbitrage opportunities in recently ordered trades.

**Any information about a transaction can be MEV enabled. **Most discussions about MEV focus on the ability of searchers to locate and interfere with the execution of pending transactions. However, even if the proposer is fully trusted and does not allow the searcher to find pending transactions, it is still possible for the searcher to exploit any information provided by the proposer.

2) Shared database status

The final execution of a transaction may differ from the execution expected at the time the transaction was signed.

**Every blockchain system operates as a finite state machine, in which case there is a state transition function (STF), which requires: **

-Latest database status

- User input

After execution, STF will output the new state of the database. We can summarize it as follows:

• STF(database_state, user’s input) = new_database_state

When users initiate a transaction, they target a specific state transition function and its inputs. Note that transactions are not committed to the current database state; the latest database state is only known at execution time.

**In a blockchain system, the state transition function includes many components that may affect database updates. **

For simplicity, it is mainly defined by virtual machines such as EVM, WASM, MIPS or Cairo. Taking it a step further, when developers deploy a smart contract to a virtual machine, they lock the entries in the database for the exclusive use of the smart contract. Database entries can only be updated when the smart contract is executed.

Smart contracts define access rights to specific database entries.

**So when a user initiates a transaction and targets a smart contract, they intend to update a specific entry in the database or any database entry that the smart contract also has write access to. Since the smart contract defines write access, it can define who is allowed to perform the operation. **

**In most cases, smart contracts operate with an inclusive policy, allowing anyone to execute it as long as they meet certain predefined criteria. **Unless smart contract functionality is used for administrative purposes, the standard will not depend on the identity of the transaction initiator, but rather on maintaining the rules governing the smart contract. **For example, before performing the exchange of Token X → Token Y, check whether the user has sufficient Token X balance. **

To summarize, we must consider two key aspects of trading:

1. No output commitment: When users sign a transaction, they are not locked to a specific execution result. Their signature covers the input and target smart contracts but does not dictate the exact execution.

2. Smart Contract Prerequisites: A smart contract specifies the conditions that must be met for successful execution. These conditions typically revolve around enforcing protocol rules (such as exchange logic) rather than the identity of the caller.

Both components are necessary to facilitate users to simultaneously issue transactions and handle race conditions. Otherwise, as we saw with the launch of exchanges on Cardano a few years ago, it can cause terrible usability issues.

At the same time, it leads to MEV being enabled on any smart contract platform, as it allows bots to interfere with the execution of user transactions and potentially generate profits by doing so.

3) Transaction bundling and interference methods

Thanks to the public nature of:

• User transactions,

• Shared database state.

Searchers can simulate pending transactions and gain complete visibility into future database states. Their task is to simulate transactions and determine whether there is a future database state that is favorable to them. If so, then they should work to enable future database states to occur and capture profit opportunities.

Once they find a pending deal that is profitable for them, searchers can execute one of two strategies:

• **Don't interfere. **Allows the user's transactions to execute as expected, and searchers will leverage the resulting database state to follow up on their own transactions.

• **Must intervene. **Searchers must issue transactions that set the ideal conditions before executing user transactions.

The hands-off approach is simple. The searcher has essentially pre-calculated what the database will look like after the user's trade is executed, and they can issue a trade that is executed after the fact and capture the resulting profits. For example, searchers can chase arbitrage opportunities by running the user's trades in reverse.

A searcher may issue two trades, sandwiching the user's trade and interfering with its execution in order to gain profit.

The do interfere method requires the searcher to issue a transaction and strives to rank their transactions before the user's transactions. This will affect the execution of user transactions and hopefully produce the desired database state that is beneficial to the searcher.

Two examples of interference include:

• **Sandwich. **The searcher will emit two transactions surrounding the user's transaction. It interferes with the user's trade execution to gain profits.

• **Be first. ** Seekers will copy the user's transaction and execute it before them. It allows searchers to grab profit opportunities before users do.

In order for interference methods to work, assumptions need to be made about the transaction execution model. As mentioned before, we assume that a user's transaction has no fixed outcome at the time of signing, and its final execution depends on the shared database state.

Due to the execution model, and the fact that the user can define a set of pre/post conditions that must be met before a transaction can be successfully executed, it can be said that the user defines a set of acceptable outcomes, even though it may be looked at by chance Profitable searchers use it against them.

When assessing the ethics of MEV, the idea that users have the right to approve a range of acceptable outcomes is important.

4) Influence how proposers prioritize transaction ordering

This brings us to the final step in enabling MEV - understanding how searchers can convince proposers to prioritize their deal packages to a specific position in the overall ranking.

The approach taken depends on the ordering strategy implemented by the proposer, but generally falls into two categories:

• **Priority Auction. **Searchers must pay higher bids than all other searchers.

• **Delayed gameplay. **Searchers must send their transactions (paying the appropriate fee) to the proposer before all other searchers.

In other words, we need to think about competition among searchers, how they can outcompete, and which approach will enable an open market of searchers to participate on a level playing field.

An example of a priority gas auction, where searchers continually broadcast new transactions with higher fees. There are at least 100 transactions within a 12 second block window.

Due to the public nature of the Gossip Protocol and the fee market auction mechanism on Ethereum, as the community became aware of MEV, a new phenomenon emerged and resulted in severe network congestion.

In a dark forest, if one searcher discovers an MEV opportunity, other searchers are likely to find it as well. Only one searcher can win the MEV opportunity, resulting in a very competitive bidding war called the Priority Gas Auction.

In a priority gas auction, the searcher wants to pay higher than the minimum necessary bid of all competitors while maximizing profits. They have to monitor the current bid set (in the mempool) and issue new transactions with higher bids. All new transactions should replace previous transactions.

**Competitors repeat the above process, resulting in a flood of spam hitting the peer-to-peer network. **For example, in the image above, we can count at least 100 transactions in a 12 second window. Furthermore, only one transaction can succeed and seize the MEV opportunity. All competing transactions are still included in a block and may not be executed. Wasted bandwidth and block space.

Flashbots solves the congestion issues associated with priority gas auctions by moving the auction off-chain.

The emergence of Flashbots brings a solution that alleviates the problems associated with priority gas auctions.

Flashbot's solution: Adopt a priority auction protocol to pick the winner and move it off-chain.

We encourage all searchers to submit bundles to Relay run by Flashbots. It is up to the relay to select the winning bid and forward it to the proposer. All failed bids are dropped by the relay.

This paved the way for the development of the proposer-builder separation (BPS) framework, a concept that distinguishes between block builders who order transactions for blocks and those who have the authority to decide on blocks The block proposer for the final content of the block.

The separation of roles creates an open market for builders and searchers, who can collaborate to create profitable blocks while sharing a portion of the profits with proposers through priority auctions. The main goal is to ensure that no one party captures all the profits generated by the MEV opportunity.

For a layer 1 blockchain like Ethereum, the process of convincing proposers is very different than for a rollup blockchain like Arbitrum.

Ethereum has about 800,000 validators, a public mempool, and the process of selecting a validator to become the next proposer depends on a random beacon. Arbitrum has only one Sequencer (Proposer), which has a private memory pool that is easy to identify and users can connect to it directly.

The aggregation environment affects how searchers try to influence proposers because they no longer have access to pending instructions and only one (or a few) parties need convincing.

As mentioned before, Searchers can:

- Listen to the sequencer feed

- Find recently ordered transactions

- Use reverse run strategies to seize MEV opportunities

If the searcher is the first bot to learn about the MEV opportunity and has the fastest connection to the proposer, they can increase their chances of winning a profit from the competition. In other words, without priority auctions, the only way for a searcher to win is to play the delay game.

Searchers studying the Sequencer feed strategy discovered that the feed would randomly prioritize different web socket connections to receive transactions first.

The best strategy is to simply open as many connections as possible and be the first to receive transactions by winning the connection lottery. This resulted in over 150k connections to the Arbitrum Sequencer.

Too many connections are a waste of resources, a potential denial-of-service attack on the Arbitrum Sequencer, and only benefit searchers who can successfully compete in the latency game.

Time Acceleration Proposal combines first come, first served with priority auctions. Most transactions can be sorted according to FCFS, but searchers have the opportunity to participate in a priority auction for a sell-back opportunity. Therefore, it eliminates any latency benefits while still allowing users to enjoy the FCFS ordering policy.

3. MEV ethics

All ecosystems need to address the following issues:

**Should we nurture the MEV environment or try to prevent it entirely? **

Surprisingly, there is no straight answer, but many in the tech community have a binary view on the subject.

The term MEV creates a sentiment in many people that we are just throwing users to the wolves, which is always bad

There are two views on the development and prevention of MEV:

• **Anti-MEV camp. **MEV is harmful. This is akin to throwing users to MEV degen wolves and subjecting them to maximum exploitation. We should do everything we can to prevent it from being exploited.

• **Support MEV camp. **MEV is good. It provides a financial incentive for searchers to perform actions that ultimately benefit the user experience and stabilize the market. What's more, the development of MEV is inevitable and we should do everything we can to embrace it.

There are some easily identifiable factors that contribute to the binary views within the community. Often, this perspective is rooted in anecdotal evidence and personal experience in finance.

**Some argue that the prevalence of high-frequency trading in the traditional financial system tends to disadvantage smaller traders in favor of larger trading firms that have the resources (and authority) to execute trades faster. **Additionally, it causes users to trade to get worse deals while allowing large corporations to profit from it.

In contrast, others believe that MEV utilization is inevitable due to the open and permissionless nature of blockchain systems. This is an inherited aspect of the way the system operates, and it can be argued that the stability of a blockchain system depends on our ability to maximize extraction while sharing the profits to all participants.

1) Evaluate how MEV affects the blockchain system

Stolen from mev.day — A big focus of the flashbot event was understanding how MEV affects a blockchain system’s consensus protocols (such as proof of stake).

In order to understand whether MEV is ethically justifiable, we should evaluate how it affects the fair reward assumptions of the first-layer blockchain system and whether it negatively affects the intentions of users to transact.

1) Fair rewards for all proposers

A core property of layer-1 blockchains like Bitcoin and Ethereum is that all proposers receive roughly the same reward for generating blocks on behalf of the network.

The incentive to provide fair rewards to all proposers has two key aspects that underpin the security and reliability of blockchain systems.

• **Keep the proposer set decentralized. **First, it is intended to prevent a single proposer from becoming disproportionately larger than all other proposers over time, potentially allowing them to accumulate enough capital to perform a 51% attack.

• **Financial incentives to follow the longest chain. **Second, it creates an economic incentive for all proposers to continuously extend the longest chain. If the reward of one block significantly exceeds that of the next block, there is a risk that proposers may be incentivized to reorganize the tip of the chain. .

In the Ethereum community, the above insights have led to Proposer-Builder Separation (PBS) as a way to democratize MEV profits. In other words, the focus of embracing MEV is to allow all proposers to share rewards fairly, ultimately ensuring the decentralization and reliability of the network.

The point of Rollup is not to provide fair rewards to hundreds of thousands of participants, but to reward any party willing to step up and keep the system alive.

Conversely, the requirements for providing fair rewards to all proposers are different in a rollup ecosystem, primarily because the underlying trust assumptions are different.

In layer-1 blockchains such as Ethereum, the trust assumption relies on a majority of proposers acting honestly to maintain the integrity of the system. It should be optimized for a large network of different participants and reward them for uptime.

In rollups, the trust requirements are much milder:

• Security, An honest party to maintain the integrity of the system.

• Vitality, any user can submit a transaction using the on-chain forced inclusion mechanism.

Of course, the forced inclusion mechanism should be a last resort option available to users (I'm not a fan of based-rollups)

Almost all users rely on designated proposers to decide the order of transactions and provide soft confirmation on how their transactions will ultimately be executed. Soft confirmations can be supported by a single proposer or by multiple proposers. You can learn more about the different levels of transaction finality in rollups.

**Dispersed sequencing. ** If some rollups want to guarantee uptime for soft confirmations between a committee (or group) of proposers, they might seek stronger trust assumptions, such as an honest majority. This is not a strict requirement for Rollup, and the community is still exploring options.

The important point is that aggregation does not necessarily need to guarantee uptime for hundreds of thousands of participants or maximize participant decentralization. The first priority is to ensure that the system is publicly accessible and that an honest party can step in at the right time to protect it.

Therefore, the need to accept MEV and provide fair rewards to all proposers of the rollup is weaker, especially if there is only one proposer. The question with adopting MEV is not the security of the system, but whether it is in the best interest of the proposer to leave money on the table or make some profits from an additional revenue stream.

This remains an open research question, but empirical evidence suggests that most aggregations today are successfully run using a single Sequencer, without MEV, suggesting this conclusion.

2) Interfering with user transactions

Another aspect to consider when evaluating MEV ethics is understanding the potential impact a searcher's transaction bundling may have on the execution of user transactions, whether positive or negative.

We believe that focusing solely on how it affects user transactions is too limiting of intent.

The assessment should cover a broader perspective, considering the impact on agents within the DeFi protocol and its ability to contribute to the synchronized operation of the DeFi protocol.

Let’s take this opportunity to consider a specific example of deal bundling and whether it can be proven to be an ethical activity.

Be the first to trade

Front-running attacks can force a user's transactions to fail, leading to censorship.

This tactic is often associated with opportunities for searchers to evaluate a user's deal, copy their content, and steal the user.

• Negative — Guaranteed Tx Delivery

Near's Rainbow Bridge utilizes MEV bots to guarantee its fraud-proof transaction delivery. This helps protect the integrity of DeFi protocols and ultimately protects users.

• Negative – Censorship Issues

Front-running transactions can enable censorship when searchers force a user's transaction to fail.

This was witnessed when Vitalik attempted to “dump” SHIBA Tokens and MEV bots interfered with his transactions to prevent him from selling the Tokens.

Vitalik was forced to migrate to CoW Swap and send these transactions directly to the proposers (miners).

• Positive – Unable to recover exposed funds

In the Dark Forest example, MEV bots were able to steal funds while users attempted to recover funds from exposed/corrupted smart contracts.

Sandwiching

This strategy is often associated with searchers changing exchange rates before and after a user trades.

- Negative – Worst exchange rate

In many sandwich situations, users end up with the worst exchange rate when performing an exchange. This is because the searcher moves a price that the user doesn't like, the user's exchange executes, and then the searcher moves the price back.

Searchers benefit by creating arbitrage opportunities and collecting any positive slippage the user may receive. Some argue that sandwich transactions are beneficial to routed transactions, but most discussions focus on the immediate negative experience for users.

-Positive—Best exchange rate (positive)

Just-in-time liquidity (JIT) requires seekers to strategically inject centralized liquidity before users exchange and withdraw it immediately after the exchange.

This allows users to get better exchange rates when making exchanges, while searchers earn a large portion of the fees for facilitating the exchange.

Passive liquidity providers (LPs) may find themselves at a disadvantage because they charge little or no fees because the exchange is executed using the seeker’s liquidity, not theirs.

Unfortunately, it has recently been discovered that the two sandwich strategies can be combined, with the result that both sides suffer. Users will get the worst exchange rate, allowing searchers to get the lion's share of the fees for facilitating the exchange. Therefore, both users and passive LPs suffer losses.

Return

This strategy is often associated with searchers chasing arbitrage-like opportunities.

• Arbitrage-like opportunity (positive)

As far as I know, the most common backdating situation is when the searcher wants to arbitrage the exchange rate after the user has made a large exchange.

This benefits the synchronicity of DeFi protocols as it keeps the token prices in sync and ultimately benefits users as they always pay the market price when exchanging tokens.

I don't recall any counter-running strategies that negatively impacted users or DeFi protocols. If you can think of anything, please leave it in the comments!

3) Objective evaluation and subjective judgment are required

We can shine a light into a dark forest, objectively measure its impact, and make a judgment about whether (and to what extent) we should embrace it.

There is a growing need for an enhanced method to measure the impact of MEV.

The method should include:

• Trade bundling strategies

• App-specific effects

• Profitability of agents

• Potential losses for agents who do not benefit

• occurrence frequency

With the above objective indicators, communities can make value judgments about their morality. For example, if we consider instant liquidity, which provides users with better tail-end asset exchange rates, if it accounts for <1% of all transactions, then it may be a reasonable MEV strategy because the benefits outweigh the risks.

To my knowledge, this type of analysis is completely absent from the MEV discourse. On-chain data is available, but the datasets are not yet easily accessible for the above analysis.

The community is discussing how to strike a balance between supporting MEV activities and maintaining ecosystem fairness, where fairness should be clearly defined.

4) Strengthen subjectivity?

The discussion surrounding the acceptance of morally justifiable forms of MEV raises a fundamental inquiry:

In a blockchain system, who has the responsibility to uphold subjective judgments about what types of MEV should be embraced or restricted?

In the context of a layer-1 blockchain like Ethereum, no central authority has the authority to impose subjective judgments. The responsibility for determining whether certain types of MEV should be excluded rests with the individual proposer or builder. **However, without concerted collective action, this exclusion is often impractical. **

Furthermore, it is highly unlikely to enforce any form of subjective judgment on a network like Ethereum, given the community's commitment to maintaining trustworthy neutrality and adhering to decentralization principles to safeguard the right to transact. Even OFAC sanctions enforcement ultimately failed to gain 100% support.

Now, when we turn our attention to the layer 2 aggregation solution, we encounter a different scenario.

Here, a single entity, the proposer, has the power to exercise subjective judgment over the soft confirmations they may choose to provide for a transaction. For example, in most rollup implementations, trust proposers do not exploit their privileged position to obtain additional rewards by manipulating MEV opportunities. However, it is conceivable that in the future, proposers may choose to restrict certain forms of MEV—at least as much as they can.

This raises the question of the practicality of enhancing subjectivity and leads to an interesting research question:

**Facilitate soft validation of selective MEV. ** Evaluation Sequencer checks the validity of a transaction in real time, determines the deployed MEV policy and decides whether the transaction should be rejected without impacting the overall user experience or increasing the soft final latency of the system.

In other words, if there is an objective way to evaluate the impact of MEV, and a subjective framework to decide which MEV should be tolerated, then how practical is it for Rollup's Sequencer to implement it.

4. Credible neutrality above all else?

By implementing a system that allows proposers to freely exclude certain transactions, we may open the door to further erosion of users’ trading rights.

As we delve deeper into the ethical considerations of MEV and potential justifications for restricting certain forms, broader ethical dilemmas arise—dilemmas surrounding how these judgments may inadvertently promote censorship.

There are legitimate concerns that users’ freedom to transact may be eroded over time as system operators deem certain transactions morally unjust. It may start with transactions that directly harm users, but will eventually lead to scrutiny of other forms of transactions because the technology now exists to enable this.

**I firmly believe that layer 1 blockchains like Ethereum must maintain trustworthy neutrality at all costs. **Protect not only trading rights, but all aggregations built on top of them. This is a prerequisite for ensuring that Ethereum can serve as a root of trust and a platform that protects user funds locked in off-chain systems.

On the other hand, in an aggregation-like system, it is possible for proposers to implement real-time transaction filtering and give up trusted neutrality.

Whatever our views on the importance of trustworthy neutrality, this avenue of research is likely to be pursued. Our community must be actively engaged and aware of the extent to which this is actually being implemented.

Indeed, accepting the practicality of transaction filters may inadvertently lead to a system that erodes users’ freedom to trade.

This is why our community must work in parallel with another research stream that focuses on ordering protocols that tie the hands of proposers, preventing their ability to filter specific transactions, and ultimately protecting users’ rights to transact.

**Cannot be evil by default. **The end game, in my opinion, is to establish an aggregation in which proposers cannot be evil, rather than simply promising that they will not be evil.

Assuming that the community decides to implement a protocol that ties the hands of the proposer, there is reason to worry that MEV must be accepted by default. I don't think that's necessarily the case. For example, include:

• Order on a first come first served basis

• Commission's Ordering Agreement

• Enable reverse runs via micro-auctions

It allows aggregation to employ background-running policies that are generally considered ethically just, while making it difficult to sandwich a user's transactions without direct access to the user.

Without transaction filtering, the trade-off is that the ordering protocol may block entire categories of MEV strategies, but this may be necessary to help protect users' freedom to trade.

Of course, on the other hand, perhaps aggregation shouldn't try to block MEV's opportunities, but embrace MEV entirely. By allowing searchers to participate in an open market, the market can reach some equilibrium around the profits generated by MEV. Anything can happen!

**There is no right answer as to whether to prevent or accept MEV. **

Thankfully, rollups, as a technology stack, gives us the freedom to try all of the above and find a solution that best protects the interests of all parties involved, including trading users, agents in DeFi protocols, and underlying participants in the protocol.