Thread by @mohakagr

(1/n) MEV - a strategic way of making money or an accidental one?
What about the fairness of transactions for users?
Are users even aware of what it is in the first place?

All of this, and more in this thread 🧵 👇

(2/n) MEV stands for miner extractable value. It is the value that miners can derive out of transaction validation, specifically by optimizing the order of transactions. They can place profitable transactions first or duplicate users' transactions to make money.

(3/n) When you initiate a transaction, it goes to the mempool, and two actors try to extract value:
- Miners, who observe a pattern of transactions on a particular dApp;
- Trading bots who notice the pattern and start bidding higher fees on the same transactions.

(4/n) In the first instance, miner decides to reap the rewards to themselves.
In the second instance, you notice that there are bots offering higher fees for the same transaction.

Eventually, miners realize the profits they can make themselves by duplicating your transaction.

(5/n) Hence, miners initiate their own tx, prioritize it over everyone else's whilst also claiming all the Priority Gas Auctions (PGAs), even if those transactions don't go through.
Miners mostly "extract" huge amounts of MEV from DEXes - Uniswap being the most popular one.

(6/n) As explained by @phildaian in Flashboys 2.0, miners have several places and opportunities to "extract" value from the process. Primarily:
-Undercutting attack: Miner attempts to fork a high-fee block by retaining some of the tx fees to entice other miners to build on it;

(7/n) Time-bandit attack: this attack utilizes a type of rewinding technique wherein the attacker rewinds to the previous block height and then utilizes the subsequent MEV to forge a 51% attack that mines a fork up to the most recent block height.

(8/n) Another type of attack is sandwich attack, wherein the miner will sandwich your tx bw theirs to generate profit.
Eg: You want to buy token X at price Y. The miner carries out two transactions - T1(buy at Y) & T3(sell at Y+4). You end up paying Y+2 in T2.
@fifikobayashi

(9/n) There has been an astronomical rise in the rise of the total value that has been extracted by miners through MEV. This value is increasing because miners have found different ways of extracting the value.

(10/n) The battle against unfair MEV has been going on for as long as MEV has existed. Some possible solutions include order sequencing auctions, using Flashbots, and the most recent Archer's MEV shield.

(11/n) Imagine this - two entities shoulder the mining process. The first validates transactions. The second simply orders the orders. This separation between the responsibilities is one proposed solution. Let's explore more.

(12/n) Order Sequencing Auctions: Auction the ordering of transactions to a different entity. The only problem here is that this is "centralized" management of order sequencing. It rests the power within a few "elected" sequencers that can front-run any transaction.

(13/n) @chainlink offers Fair Sequencing Services where the ordering of transactions is the responsibility of an oracle network. Since the latter only gets to "order" transactions and not mine them, chances of frontrunning are presumably reduced. Note - oracle is a 3rd party.

(14/n) We can alternatively put our power of research to track the MEV metrics and understand how miners are taking advantage of them. Flashbots does exactly that. Their MEV-Geth sits on the go-Ethereum client and enables a sealed-bid block space auction mechanism.

(15/n) Get a Captain America-like shield to protect from MEV. @Archer_DAO MEV Shield helps protect its users from front-running and sandwich attacks. They have already integrated with @SushiSwap, so users can now stay protected from undue loss of revenue.

(16/n) Did you know that the world black market currently stands at a valuation of $1.8T. It is a necessary evil that is "allowed" to exist because it cannot be removed. Do you think that MEV is a similar inescapable anomaly?

(17/n) There has been an increase in blocks with unconventional ordering. As that blue bar rises, users begin to question the applicability of solutions aimed at protecting users from MEV.

(18/n) While the solutions seem effective, they fail to discuss the main problem of centralizing the power of ordering transactions in a few hands. Once auctioned, miners and order-optimizers can interact off-chain to still manipulate the transaction ordering.

(19/n) The arguments that are made in favor of the solutions often talk about making transactions fair: how does auctioning off order optimization to a third-party network help in achieving that? In fact, it makes things annoyingly more centralized.

(20/n) Moreover, what is to stop the same entity that validates transactions to also order them. Aren't we running in circles with this solution?

(21/n) Front-running-as-a-service is getting popularized as a way of selling/auctioning block-construction services to miners. Ironically, the more we attempt at decentralizing the process, the more it finds a way to get centralized.

(22/n) Archer's MEV Shield offers a workable solution. Other proposed solutions are viable as long as they are effectively tested across a period of time. While finding an absolute solution might be difficult, it's hard to tell if MEV is a problem we can completely do away with.

(23/n) While some solutions are effective, MEV is not giving us the impression of a problem that can be dealt with easily. What do you think?