The inevitability of MEV: On-chain game you can't see

Author: Sumanth Neppalli Source: decentralised.co Translation: Shan Oppa, Golden Finance

This article continues the topic we have repeatedly discussed: income and capital flows . A thorough analysis of the source of miner extractable value (MEV) and the current new methods to mitigate MEVs.

You may have noticed that almost all of our articles in recent times revolve around one common point: the primitives of frequent funds and large-scale flows . Our core view is still: blockchain is the new financial "track", and products that can drive capital flows at high frequency will become winners in the next stage of crypto evolution. This logic also guides us on how to view the M&A market.

Now, it's time to understand how the value in each block is extracted .

Welcome back to 2010. That year, Google launched the " Binary Advertising Bidding System ". The principle of running is simple: the highest bidder wins the ad slot, but only pays the second highest price.

This is perfect in the eyes of economists: advertisers don’t need to worry about “it’s expensive to buy at a high price.” However, Google secretly manipulated it behind the scenes and quietly pulled millions of dollars from it .

For example, the highest bid is $20, the second highest bid is $16, and the advertiser only needs to pay $16. But Google paid website publishers the third highest bid (such as $10), which means Google quietly swallowed $6 profits. They put the hidden profits into an internal fund called " Bernanke Pool " to achieve quarterly profit targets and please Wall Street. It was not until an antitrust lawsuit in 2016 that the truth was revealed.

Although Google turned to a " one-price auction system " in 2019, that is, advertisers pay as much as they pay (minus platform handling fees), this historical statement: Even the "fair" mechanism may be distorted when auctioneers control the underlying structure .

Interestingly, this kind of history is repeating itself in the blockchain world . The maximum extractable value (MEV) phenomenon is essentially an extension of a secret auction , which allows miners/verifiers to extract profits by reordering, inserting or excluding transactions in blocks.

For ordinary users, this kind of operation is mostly invisible and intangible, but it affects every on-chain behavior participant - this is simply an invisible tax .

Then will MEV move towards secrecy and centralization like Google? Or can we build a system that is open, transparent and rewards to users ? Can we design a system that makes this inevitable value extraction beneficial to the entire ecosystem, rather than letting a few people get rich ?

Physics of Delay

Blockchain is a distributed network of thousands of computer nodes (miners or validators) that together receive and process transactions.

The validator has two roles:

1. Communication node (receive and broadcast transactions);

2. Compute nodes (execute and verify transactions).

Since these nodes are distributed around the world, communication inevitably has delays - limited by the speed of light.

In order to ensure the consistent transaction order, each chain has set a "block time" to synchronize all nodes. Each round is rotated to select a validator to package the next set of transactions, and this mechanism is maintained by consensus protocols (such as PoS).

• Bitcoin: One block is released every 10 minutes

• Ethereum: 12 seconds

• Solana: about 400 milliseconds

• Second-layer chains (such as Arbitrum and MegaETH): 10ms to 250ms

Each time window, regardless of length, creates opportunities for validators to reschedule transactions within the block to maximize their own profits rather than prioritizing user fairness. Ideally, it should be first come first served. But since nodes are distributed around the world, this is difficult to achieve. When a user initiates a transaction, due to network delay, it is guaranteed that not all nodes can receive the transaction at the same time. This means that sometimes unfair transaction order (the end user pays an additional fee, and the MEV users make money) can be packaged into the block without violating any block building rules.

MEV is a big business

If Joel bought ETH at $1800 on a decentralized exchange (DEX) like Uniswap and set a 10% slippage tolerance, that is, he can accept that the price will rise to up to $1980 when it is closed.

Joel's transactions enter the memory pool (the transaction memory pool, which is the "waiting room" on the chain), where they wait for it to be packaged into the block by a certain validator. At this time, a robot discovered the transaction, copied the same payout order , and submitted it with a higher gas fee, so its transaction will be executed first .

The higher gas fee is essentially a bribe to validators , allowing them to rank the robot's transactions ahead of Joel. The robot's payout will push up the price of ETH on DEX, for example to $1900. Joel's transaction is executed at this time and is traded at this price after the rise. The robot then immediately sells its ETH for $1900, thus earning the spread (net profit after deducting gas fees).

Joel did get ETH but paid $100 more. And the robot made the $100 profit. This kind of thing happens thousands of times a day in the crypto market.

More extreme cases also exist. For example, a poor trader forgot to set slippage, and the robot swallowed $200,000 in profits in one bite. This time, the "raider" is the famous jaredfromsubway.eth , a robot address, which has long dominated the Ethereum gas consumption list with extremely high gas consumption, and is specially used to insert the transactions he wants to attack. It is estimated that Jared made more than $10 million in profits through MEV attacks.

MEV mainly manifests in three forms:

1. Arbitrage : Find out that the prices between exchanges are not matched, buy low and sell high in the same block. For example, ETH trades for $2,500 on Uniswap, but trades for $2,510 on Sushiswap. Robots can buy ETH on Uniswap within the same block and sell on Sushiswap, thereby locking in profits of $10 per ETH without taking market risks. It is worth noting that this is good for the market, as it will eventually make asset prices consistent across different platforms.

2. Sandwich Attack : Check out Alice's large payout orders in the memory pool, grab her before she buys (pushing up the price), get her to pay a higher price, and sell it immediately. Robots earn the difference, while Alice bears the slippage. The example mentioned above is a sandwich attack. Joel paid an extra $100, which is actually profit from the MEV value chain. This is not advisable because the user ends up paying more than the necessary amount.

3. Liquidation : In a lending agreement, when a position meets the liquidation conditions, the MEV withdrawer will compete for the first liquidation and receive a liquidation reward. Suppose Saurabh borrowed $15,000 worth of ETH and $10,000 worth of USDC on Aave. After the price fell, his ETH collateral fell to $11,000, breaking the liquidation threshold. A robot competes to liquidate 50% of his loan. It repaid $5,000 in USDC and in return it received $5,500 worth of ETH (benefited from a 10% liquidation bonus). This is $500 in profit, you can get it with the first action. This can be good or bad, depending on the situation. The incentives provided in the form of liquidation are beneficial to the overall health of DeFi. But most of this revenue ends up going to validators.

From US$550 million in 2021 to double to US$1.1 billion in 2024, the scale of MEV extraction has shown explosive growth. Due to Ethereum's open memory pool and deep DeFi liquidity, it remains the "episode" of MEV, with more than 100 active robots active here. To date, about 75% of MEVs have occurred on Ethereum.

In the past 30 days, 66% of MEV transactions on Ethereum were sandwich attacks , 33% were arbitrage , and clearing exchanges accounted for less than 1%.

As on-chain transactions expand to other public chains, MEVs also spread. Solana, BNB Chain and various Ethereum Rollup (L2) have all become the "hunting grounds" for robots to pursue profits. Even Binance founder CZ was attacked by sandwich when exchanging coins.

The sandwich robot on Solana has earned over $4 million (about 24,000 SOL) in the past 30 days , which is almost 50 times the revenue of Ethereum robots during the same period (about 80,000).

Cross-chain bridges have also transformed the competition of MEVs into a cross-link race, with robots shuttle back and forth between different ecology, trying to capture every opportunity. In December 2024 alone, MEV activity on Solana exceeded $100 million , mainly affected by market volatility caused by Trump’s re-election campaign.

In 2024, the total transaction volume on DEX reached US$1.5 trillion , while the cost of MEV accounts for about 0.1% of trading activity. Frontier Labs estimates that this number may rise to 1% , which is expensive for large-scale transactions.

It’s easy for us to see MEV as “bad, evil.” But the fact is: in any financial market, value loss is an inevitable phenomenon . The really important question is whether we can reduce this churn or distribute it more equitably to all market participants .

MEV 's supply chain

Early blockchains gave validators two "super permissions":

1. Decide which transactions are included in the next block;

2. Determines the sort of these transactions.

This raises a serious problem, like the Dark Pool described in Flash Boys . Just as stock exchanges provide "privileged channels" for high-frequency traders, validators in blockchain can also privately reach cooperation agreements with robots to ensure that transactions of these robots are processed before ordinary users trade.

This "pay priority" mechanism allows insiders to always get the best price, while ordinary users can only eat "sweet soup".

To address this centralized risk, the Ethereum ecosystem is advancing an architecture called Proposer-Builder Separation (PBS) . It separates the process of block "building" and "oning":

1. User submits a transaction or advanced intent (such as "swap token A with the best price");

2. The wallet processes these transactions and sends them through nodes to the searcher/builder/memory pool;

3. Searchers scan memory pools to find arbitrage opportunities and package related transactions;

4. The builder builds the complete blocks from these packaged transactions and auctions them out;

5. The validator (or proposer) selects the optimal block from it, checks its legitimacy, and adds it to the chain.

This mechanism limits the power of validators - they can only choose from the sorted blocks submitted by the builder, thus allocating the opportunities for MEVs more widely to market participants , forming a more fairer block building competitive market.

The most widely used PBS is MEV-Boost launched by Flashbots , which has been adopted by more than 90% of Ethereum validators as of the beginning of 2025.

From "miner" to "maximum" value extractor

MEV, miner extractable value was first dominated by miners, but has now evolved into the maximum extractable value . Extractors are no longer just miners, but rather a huge ecosystem of multiple roles .

When you click on "Swap" at Uniswap (DEX based on automatic market makers) or Raydium (DEX based on order book), it is almost impossible for your opponent to be another average person .

What you are facing is a professional market maker like Wintermute . They make profits by capturing the bid/ask spread (i.e. bid/ask spread). In the AMM model, they provide liquidity by depositing funds into liquidity pools, earning handling fees from user transactions.

It is unrealistic to "eliminate" MEV, because it is **deeply embedded in

the block time economic structure** .

• On the positive side: arbitrage helps price consistency between CEX and DEX markets, while subsidizing cybersecurity through MEV commissions ;

• Negative side: sandwich attacks and gas bidding wars make ordinary users pay higher costs.

MEV is an inevitable product of an efficient market - wherever there is profit, there will be people competing for it .

In the current ecosystem, professional searchers, builders and market maker robots are the biggest beneficiaries . The ones that bear the costs are ordinary traders: they are snatched up, suffer additional slippage, or trade opaque because liquidity moves to an invisible "dark pool".

The robot frantically sent "bullet transactions" to compete for the opportunity of MEV, completing the snatch in a few milliseconds , causing memory pool blockage, bandwidth waste and transaction fees to rise. This is what we call " invisible tax" .

The rules of the game for MEV are not “elimination” but rather to be clear – who should enjoy the benefits and under what rules to allocate them .

Strategies to reduce MEV

The ecosystem is currently trying four main ways to deal with MEV problems: hiding , utilizing , minimizing , and redirecting . Each approach has different tradeoffs between efficiency, fairness, and technical complexity.

MEV Hide

The easiest strategy is to not disclose the transaction content before the transaction is packaged into the block. Such tools include Flashbots Protect and Cowswap's MEV Blocker.

These services are used directly: the user's transactions are not sent to the public memory pool, but are submitted privately to the block builders . This way, the MEV robot cannot see the transaction before it is processed.

The disadvantage is: you have to wait until a validator using these services becomes a block producer . On Flashbots Protect, this wait can last up to 6 minutes . However, during this period you can cancel the transaction at any time because the transaction has not been sent to the memory pool yet.

Market makers and large-cap traders usually use this type of service to avoid premature exposure of trading intentions . So far, more than $43 billion in transactions have been sent through Flashbots Protect.

But I am reserved for this centralized privacy solution because it reminds me of the "dark pool exchange" in traditional finance. These platforms were initially designed to protect users, but often evolved into internal privileged systems like Robinhood .

Flashbots and Beaverbuild are exploring the use of trusted execution environments to encrypt and prove their behavior is honest and reliable. This direction is expected, but it has not yet been verified to be effective on a large scale .

Some communities have also begun to take the initiative. For example, the BNB community voted to pass the "Good Will Alliance" and required validators to accept only block proposals from compliant MEV builders. These builders filter out malicious MEV transactions and will be penalized if validators do not use compliance infrastructure .

MEV Utilization

Some protocols are not about eliminating MEV, but about "fighting poison with poison" and weaponizing competition among robots through a private bidding mechanism.

For example: Suppose Joel wants to exchange 100 ETH for USDC. In traditional AMM mode, this transaction will appear publicly in the memory pool and is prone to sandwich attacks. However, if you use RFQ (Request-for-Quote) mode, Joel will send a coin exchange request to a group of market makers networks.

For example, Wintermute quotes USD 2,000/ETH, while DWF Labs quotes USD 2,010/ETH. Joel chose the latter and completed the transaction at a good price in 2010, without slippage and avoided a snatch .

Behind the scenes, every market maker is evaluating how much they can make from the deal. They will extract liquidity from different sources while ensuring profits to give Joel the best execution price while surpassing its competitors.

However, RFQ systems also come with their costs. It relies on a stable, all- weather online, liquid market makers network to respond in real time. If there are insufficient participants, the system will respond slowly - the user can only wait, and the price is still fluctuating.

The RFQ model is more used in under-liquid markets such as bonds , because the order book in these markets is usually too shallow. If the institutions involved in market making are not credible enough or lack decentralization, RFQ may degenerate into another "insider club" .

To solve these problems, Python built Express Relay on Solana , an off- chain market maker market . Any DeFi protocol can be accessed through it to a competitive market maker pool without having to develop separate interfaces for each market maker , which not only simplifies the integration process but also reduces MEV losses.

Jito adopted a different strategy and eventually became the dominant validator client on Solana , currently controlling more than 90% of the staked SOL . Jito once tried to introduce a Solana memory pool, but was quickly exploited by attackers. Someone was willing to spend more than $300,000 to seize block priority, and this attempt was eventually abandoned.

Jito now runs an off-chain auction every 200 milliseconds , selecting the most profitable transaction to package into the next block. If the user wants to execute the transaction first, add a "tip" to the transaction to prevent MEV attacks. The highest bidders preferentially execute transactions , and these tips translate into validator revenue—which currently accounts for more than half of Solana’s validator revenue.

MEV Minimization

This approach is based on the order flow auction mechanism , with the goal of designing smarter auction mechanisms to reduce the total amount of MEV that can be extracted .

When transactions are processed one by one, the robots are created to view each transaction and quickly insert profitable arbitrage operations to make profits. If multiple orders are packaged and executed uniformly at the same price , the MEV brought about by the ordering and time difference will be smoothed out.

CoWSwap is the pioneer of bulk trading. Its core concept is very simple: when one user wants to replace ETH with DAI, and the other person wants to replace DAI with ETH, they can actually switch directly and do not need to go through traditional exchanges at all. CoWSwap will collect user's trading intentions in a short time window and prioritizes matching these "natural rival plates". It will only seek the liquidity pool on the chain if no match is found.

What's even better is that users do not need to have the expertise in crypto market structure . When trading on CoWSwap, there is no need to manually set slippage tolerance, nor do you need to set up routing policies for different pools. The user only needs to submit the content they want to trade, and the system will complete the execution through a group of special market makers called "solvers".

These solvers will compete in fair bidding to get the best price for you . In each round of batch trading, the transaction prices of all assets are unified, which further prevents jumping.

The actual results say it all: CoWSwap decentralized exchange has processed about $100 billion in transaction volume so far. One of the leading solvers, Barter, took up about 15% of the market share , and handled more than $11 billion in transactions through the agreement. Barter's continued growth further verifies how CoWSwap's bulk auction mechanism reduces MEV through fair pricing rather than time advantages.

This approach is also highly consistent with the research views of Eric Budish , a professor of economics at the University of Chicago Booth School of Business. He believes:

“Processing orders every second in high-frequency auctions eliminates meaningless competition for speed.”

He pointed out that this batch processing mechanism can solve the "prisoner's dilemma" problem in the continuous order book market (such as centralization or the limit order mode in some DEXs), and can also truly allocate the transaction costs saved to ordinary investors.

In the cryptocurrency market, the continuous limit order book rewards those who are "fastest" - which leads traders to invest in better hardware, faster robots, or direct-connected nodes, which are not good for ordinary users and are just the price to survive. The batch auction system similar to CoWSwap subverts this logic: all transactions are processed at the same price within a fixed time window, and the speed is no longer important, and the focus is on price discovery and user value.

MEV Recycling

Some innovators have taken a more pragmatic approach : Since MEV cannot be eliminated, why not capture it and return it to the community ?

Arbitrum 's TimeBoost is an example of this idea. It sells a 200 millisecond "fast lane" every minute through a closed binary auction - just like the Grocery Store's VIP fast checkout channel, where the highest bidder can intervene in a queue for transactions within this short time .

Searchers who want to prioritize their transactions can bid by predicting the potential of MEVs in the next 60 seconds rather than through the fierce Gas war.

This mechanism changes the game for an attacker : any transaction executed through a fast lane will not be attacked by a forward run or a sandwich. Moreover, since auctions are rotated every minute and are open to all searchers, it is almost impossible to form an exclusive MEV monopoly alliance .

The end result is: MEV is no longer a mechanism for secretly increasing taxes, but has become a " public income tool ." Currently, 97% of TimeBoost 's revenue is returned to the ARB DAO vault , which theoretically generates up to $30 million in annual revenue.

Even Jito , which we mentioned earlier, adopts a hybrid strategy : 3% of the tips users pay for priority transactions are reassigned to Jito DAO and JitoSOL holders .

Choose the right auction mechanism

Among the five main auction mechanisms that dominate the MEV field, we can use a simple example to illustrate their differences: suppose there are three bidders (searcher/solver/block builder) who bid $100, $80, and $60 for one block location.

Which auction mechanism is "best", there is no standard answer , depending on the specific objectives of your agreement:

• If you are pursuing your current considerable income , continue to use closed-price auctions

• If you pursue community recognition and long-term sticky users , you can use a mechanism similar to EIP-1559 - such as an intention order system with basic fees plus a unified price.

• If you want to break the delayed arbitrage alliance , use frequent batch auctions to let "price" rather than "speed" determine who can enter the block.

• If your scenario requires extremely high transaction speed (such as DEX), then private order flow is the best choice.

Where is the next step?

Does the strategy of hiding order flows and auctioning them to private market makers sound familiar? The story of Wall Street 's "Dark Pool" is repeating itself on the chain . This trend is expected to accelerate as the crypto market becomes increasingly institutionalized and intersects with traditional auction mechanisms.

Currently, only a very small number of the most professional teams can compete in the role of searcher, builder or solver. Their complexity advantages will gradually accumulate over time : such large institutions are equipped with top-level infrastructure and have engineering teams developing exclusive ML algorithms, which is likely to form an overwhelming advantage. We may even see traditional giants like Morgan Stanley or Goldman Sachs entering the field.

Blockchains have also begun to form ideological stances on the MEV issue: Solana is obsessed with ultra-low latency and naturally tends to follow the NASDAQ speed advantage in its privacy order flow mechanism. In contrast, Ethereum uses PBS and MEV-Boost to democratize access. Other blockchains explore new directions based on their own architectural priorities.

But the most noteworthy innovation may appear in Layer 2 . These new chains have the opportunity to design MEV-resistant architectures from scratch. For example, Arbitrum 's TimeBoost demonstrates that L2 can be more freely tested in value allocation and auction mechanisms.

The composability and permissionless nature of DeFi make the field of encryption an excellent laboratory for experimental auction mechanisms. In traditional finance, "frequent batch auctions" have attracted attention since 2015, but it has been difficult to advance due to regulatory resistance. And on the chain, we can quickly iterate and implement it in a few months, just like Sei does.

Another direction is to build a decentralized market maker network . In the future, we may see the emergence of a reputation market for block builders , and participants prove honesty by pledging tokens. Combined with re-staking protocols like EigenLayer , market participants can establish transparent reputation scores for upper-level protocol calls.

If you think these sound too crazy, think back: Twenty years ago no one could have imagined that microwave signal towers would reshape the S &P order book . Incentive mechanisms will always find new ways out.