Maximal extractable value (MEV) refers to the maximum value that can be extracted from block production in excess of the standard block reward and gas fees by including, excluding, and changing the order of transactions in a block.
a) MEV is always
The transaction ordering of Ethereum blocks is not constrained by certain rules from beginning to end, but can be freely modified according to the preferences of miners. In the source code of the Geth client, three examples are provided:
- The first: sort by Nonce only
- Second: sort by gas fee and Nonce (default option)
- The third type: after sorting according to owner, sorting according to gas fee and Nonce
In fact, the sorting is completely an open question, and miners can adjust freely to capture greater profits, which is the protagonist of our discussion, MEV.
Before MEV was called MEV, its Alpha was continuously mined by miners. flashfish0x tells the story of a MEV.
Before August 2020, Compound’s price feed was updated by a special oracle tx, so there would be miners monitoring the Mempool, looking for accounts that could be liquidated, and arbitrage using flash loans and other means. Since the arbitrage tx is executed just after the price update tx can the arbitrage succeed, and the ordinary miners (at that time) will randomly sort the arbitrage tx with the same gas fee, so the arbitrageur will send the transaction at the same gas fee as the price update tx of the oracle machine, Win a random probability with other arbitrageurs.
After August 11, 2020, Geth released a new version, and the first tx that arrives on the network after the price of the oracle machine is updated to tx will be arbitraged successfully, so the speed of obtaining tx information is also taken into account in the consideration of competition.
The good times did not last long. On August 17, 2020, Compound switched to a more decentralized oracle. So the MEV catcher competition turned into a pure gas war. They try to pay higher gas prices than everyone else until the liquidation doesn’t capture any profit, or someone wins.
The blockchain network has been such a dark forest with undercurrents from the very beginning, where robots compete with invisible opponents in an unknown environment for opportunities to extract potential profits.
b) MEVs are everywhere
Before formally interpreting the definition of MEV, we can look at some data about MEV, so that we can intuitively feel its existence and impact.
In May 2022, more than half of the trading volume on Uniswap was contributed by MEV bots, while less than 20% of the trading volume came from the Uniswap front-end page.
The data from dex.guru also shows that a large part of the transaction volume of AMM on Ethereum is contributed by MEV robots.
MEV bots are ubiquitous in the dark forest of blockchain, and always dominate.
c) MEV supply chain
So how does an MEV activity come about?
The life cycle of a tx in a crisis-ridden blockchain is as follows:
- Health: Ordinary users interact with the application according to their intentions. The wallet and the application interpret the intention as tx and send it to the public tx pool and other systems for processing.
- Old: MEV seekers (probably miners) pick out all lucrative opportunities and try to capture MEV by sending transactions or deliberately ordering transactions.
- Disease: L1 miners (mostly not MEV searchers) or L2 Sequencer aggregate transactions and produce blocks.
- Dead: A validator (currently a miner) validates the block and the transaction is officially confirmed.
MEV robots carry out on-chain activities according to the will of their owners, packaging tx and sending it to unwitting miners to produce blocks. From a good point of view, they are an important role to ensure market stability and DApp activity; from a bad point of view, they use their inherent advantages (the ability to monitor the entire Mempool) to unequal “ordinary users” exploitation.
d) Specific definition of MEV
The word MEV has two official interpretations, and its full name can be:
- Miner Extractable Value (Miner Extractable Value)
- Maximum Extractable Value
When we combine the two explanations, we get a concept that is easiest to understand: MEV is the maximum value a miner can extract, and is the incentive for inserting tx at a specific location in a block.
Personally, a more comprehensive definition of MEV refers to miners/validators/robots/oracles/roles other than ordinary users, using miners’ asymmetric information to extract the maximum value. The role of non-ordinary users (understand the code, understand the network mechanism, understand the use of MEV), extract additional value through their own “privilege”, and thus make profits, in fact, can be regarded as a generalized MEV:
- Take the miner who produces the block as an example: after seeing the transaction in the Mempool, when it is the turn to produce the block (very small probability), it can completely reorder the transactions in the block according to its own wishes (rather than normal nonce and gas fee ordering), and send new transactions to obtain larger additional profits in addition to rewards.
- Take users who understand the concept of using MEV as an example: use Flashbots Protect as a tx pool to avoid being cut by MEV robots and tx can get priority execution and other advantages; through flashside or the Compound liquidation example above, monitor Mempool to monitor NFT and on-chain activities Arbitrage to get unwitting miners to capture MEV for them at just the right gas fee; via mint. fun monitors Mempool to mint the latest NFT projects.
- Take the most exploited ordinary user as an example: get up in the morning to buy ETH, but the first tx fails due to insufficient gas; the second tx at noon succeeds with great difficulty but is cut by a clip; stays up late at night, and wants to see it Free mint NFT for a long time, but the transaction was monitored by others as soon as it was sent, and finally it was not mint.
The extra profit of MEV extractors comes from other transactions in the block (almost from ordinary users), and squeezes a part of the space of each block (despite the lack of elastic demand for gas fees, their transactions are usually in-block The first one, doesn’t directly affect users tx), so most MEVs are usually seen as a hidden tax on users.
1. Classification of MEV transactions
Here is a clear illustration of 0xminion’s MEV activity:
According to the purpose, MEV operations can be divided into: various arbitrage, simple front-running trading, clip attack, etc.
Most of these operations will be arbitrage operations, which are basically initiated by robots. These robots have done a lot of work for gas optimization. For example, they usually have countless tokens in their wallets, each with only a little bit. quantity. This is mainly due to the use of EIP-2200, when users receive ERC-20 tokens, if there is this token in the wallet, it will save more gas than without. Another is to make the address have more zeros. Of course, the above arbitrage optimization operations have little to do with MEV itself.
MEV itself is mainly divided into the following six types:
a) Front Running
Front Running is where the MEV robot pays a slightly higher gas fee to preemptively execute a transaction before a transaction in the Mempool, such as Swap tokens at a lower price.
When this type of MEV activity does not cause subsequent “casualties”, it can be regarded as neutral (not benign because it is mainly still jumping in the queue); but many times it usually causes subsequent transactions to fail or perform worse, Therefore it is considered malignant.
There are currently many services that focus on improving the capture of such MEVs, such as bioXroute (similar to the Mafia collecting protection fees to provide faster and more private Mempool, thereby speeding up tx confirmation efficiency and defense against external Front Running).
The core of the problem is that this type of MEV directly reflects the information asymmetry between MEV capturers and ordinary users. However, it is impossible for everyone to reach a consensus on the method and algorithm of transaction ordering in a completely open, transparent, and uniform manner.
b) Back Running
Back Running is where the bot tries a different arbitrage, liquidation, or trade after a trade has caused the price to move significantly.
The most typical example is the liquidation arbitrage operation we mentioned earlier. Such MEV activities can be considered benign as a whole and can ensure the stability and continuous operation of the market.
This type of MEV activity is especially well done on a network with a multi-chain architecture like Cosmos, because there are so many networks that such activity is needed to keep DeFi applications running and efficient.
c) Sandwich Attack
Sandwich Attack is a combination of the first two attacks, which is a front and back attack on the transaction. Often called a clip. For example, the MEV robot places a buy order before the trade and a sell order after the trade, allowing the user’s trade to be executed at a worse price.
This kind of attack can sometimes be very complex and the damage can be very large. The following picture is a user who has a robot clipping Curve through 100 million DAI:
It can be generally regarded as vicious, but in a way like a highway speedometer, it spurs users to use more reasonable slippage, thereby improving the overall interests of the network.
d) Time-Bandit Attack
The time thief attack is that after the block is generated, the miner re-mines the block to cause the blockchain to reorg, and at the same time extracts the value in the block they mined. This is the most vicious MEV that will invalidate user transactions. At the same time, this is rare, but it is possible.
In the blockchain network under the Nakamoto consensus, due to the longest chain principle, block reorg occurs very frequently (with large delay), so this type of MEV is likely to be captured. In Tendermint or PoS’s Ethereum, the occurrence of such MEVs will be greatly reduced.
The cool thing about the Time Thief MEV is that future events can affect the past (including “future” transactions into “past” blocks).
But under simple quantification, the expected rate of return for such an attack is not high, and at the same time it will be condemned for deliberately destabilizing the blockchain network (such as Reorg-as-a-Service), so it rarely happens.
The validity of any NFT, asset, and Rollup transaction is actually guaranteed by purchasing a certain space in a certain block on the blockchain network. The following is the supply and demand of the blockchain blockchain space market:
Clogging is to win games such as Fomo3D by continuously monopolizing the entire space of the block.
Remember the Geth source code at the very beginning? Clogging occurs because about 70% of Ethereum miners sort transactions according to the default gas fee ordering.
But at present, because of the complete infrastructure, such as Oracle and Rollup, the transaction submission will be carried out without hindrance. MEV robots will carry out such activities during the network active period, and the cost will be relatively high.
f) Non-Broadcast Transactions
The initiators of the previous activities are basically MEV robots, not the miners themselves.
Non-Broadcast Transactions are tx secretly submitted by block miners, that is, when miners know that they can generate blocks, they immediately insert some transactions into the block to capture MEV.
About 2% of transactions on Ethereum are generated in this form.
2. Is MEV good or bad?
In the previous analysis we explored different kinds of MEVs, interspersed with our evaluation of each and our judgment of whether it was good or bad. So overall, is MEV good or bad? How do we deal with it? How do different networks deal with it in different ways?
The good and the bad of MEV:
- Good MEV: used for arbitrage clearing (Back Running), open and transparent market (Flashbot, MEV Auction), and does not affect the network or other users.
- Bad MEV: used to extract value from other users (Sandwich Attack), sent privately without broadcasting (Time-Bandit Attack, Non-Broadcast Transactions), affecting the normal operation of the network (Clogging).
From the perspective of the number of classifications, MEV seems to be a neutral role that is neither good nor bad, but when considering the essence of MEV (incentive to insert tx into block), this incentive is too huge, which will cause the damage caused by chasing bad MEVs to be far greater than that of good ones. Benefits of MEVs. In addition, the generation of MEV also stems from the asymmetry of information and knowledge. In the blockchain network, there is only equality and no equity.
To sum up, the black market principle of MEV + huge profit drive + complex situation and classification = negative externalities of the blockchain network.
MEV is bad and needs to be cut.
3. In-depth thinking and handling of MEV
Before classifying and discussing strategies for MEV, we need to make one point clear, that is, information asymmetry and externalities in any situation cannot be completely eliminated, and we can only deal with them better.
a) Solve the MEV black market and its huge incentive
MEV is a black market (or dark forest) created by opaque information and knowledge, and this black market has created huge benefits. We will consider strategies for MEV through this lens.
The entire world’s black market is valued at around $180 million and is an inevitable evil that cannot be eliminated. The market for MEVs is also much larger than we can observe.
Since this black market is inevitable, why not let it be exposed to the sun, so that everyone can freely capture and protect their own MEVs, so that ordinary users can form a joint force, and can weaken the confrontation of MEV capturers against users.
Flashbots propose the following five market scenarios to limit the extraction of MEVs, thereby reducing the overall negative externalities of MEVs:
These MEV reduction schemes are actually similar, but the same disadvantage is that it may lead to wasted block space and centralization of block production.
b) Complications of MEV: Scenario Differences
- Single Chain Scenario: Most of this article discusses the MEV situation in a single blockchain network, which is an overly idealized analysis in reality.
- Cross-chain scenario: a blockchain network = a trust zone. In practice, there will be countless blockchain networks with different mechanisms and different types of MEVs, and these countless trust zones allow arbitrage between networks to exist everywhere. For example, even if Ethereum does not have MEV at all, if there is an arbitrage opportunity on Cosmos, the bridge validator or cross-chain application robot can capture MEV on Cosmos, and eventually Ethereum also becomes the victim of this MEV.
- Multi-chain scenario: Examples of cross-chain scenarios can be seamlessly applied to the multi-chain architecture of Cosmos or Polkadot, and the existence of MEV is contagious.
b) The complexity of MEV: network architecture differences
In our article exploring the Rollup network , we learned the difference between a monolithic Layer1 and a Layer2 Rollup built on top of it.
As an “enhanced smart contract” and a “centralized trusted blockchain network”, Rollup currently has a centralized block generation and sorting mechanism, so it has different MEV performances. The current centralized Sequencer (that is, only one node) results in that all MEVs can be captured by Rollup. As more and more liquidity and on-chain activity move to L2, the revenue that L1 miners can earn (including MEV) is actually transferred to L2’s Sequencer.
Of course, the Rollup network itself has realized and solved this problem, and their fair ordering of tx reflects the processing of MEV. Different Rollups have different strategies for MEV, and have completely different meanings to network participants:
When L2 steps into decentralization and the implementation of modular blockchain becomes more prevalent, the issue of MEV requires more research and thinking. For example, the modular layout of Ethereum and Celestia will be different. At present, all Rollups on Ethereum are Secured Rollups. It is only necessary to build n two-way credible cross-chain bridges (their job is to build such bridges) and settle them together on Ethereum. The Sovereign Rollup encouraged by Celestia Separately, DA is likely to be different, and it may take n^2 bridges or a protocol like IBC or XCM to get through. This creates a completely different MEV market and capture opportunity.
From the question of L2 MEV, we can also extend the thinking on the value relationship between L1 and L2, which will not be expanded in this paper.
2. A more centralized blockchain
In order to be faster, Solana does not have a public Mempool, but uses a Gulf Stream like a Gulf Stream that is only visible to miners who are expected to produce blocks (this is also the reason why it is down…). This means that MEV Bot has no way to capture MEV. Only miners who run a Validator and are about to produce blocks theoretically have the opportunity to sort transactions or send new transactions to capture MEV. But combined with the low profit of Solana’s miners, this design actually leads to lower income for miners. So now there is Jito Labs as a client of mev-geth similar to Flashbot, to let miners auction their own block space, and let MEV form an open free market. At the same time, Solana also has solutions for QUIC and per account fee market.
For the topic of MEV, this article is only a superficial analysis. There are a lot of interesting things that have not been mentioned due to space problems (SGX, PBS, etc.). It is recommended that you read more than 50 related links at the end of the article.
In an ideal network:
- Anyone can send transactions (no censorship)
- no spam
- Very low rates (low fees)
And the existence of MEV makes it impossible to have both:
- no spam + low fees: Censorship like Web2 must exist.
- no censorship + low fees: MEV Searcher will send spam messages to affect normal transactions.
- no censorship + no spam: you need a fee market to sort transactions, which requires expensive gas to prioritize transactions.
But don’t worry too much, Web1 or Web2 or the real world has not completely solved the negative externality of “MEV”, and the various solutions of Web3 have done very well. MEV has always been an open problem with no fully correct solution.
So what should we do with Web3?
- L1/L2: MEV is given priority in design.
- Miners: Feel free to capture value without compromising ecology, as this is part of network game theory.
- DApp developers: Design applications with MEV in mind, and whether the mechanism will hurt ordinary users.
- User: Understand MEV, and try not to use MEV mechanisms for poorly designed networks or applications.
- Others: Encourage MEV research, participate in MEV discussions, Front run the crisis.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/foresight-ventures-depicting-taxonomy-dominating-mev/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.