TL;DR

• MEV (Maximal Extractable Value) is an old topic every blockchain deals with. It’s a complex game involving users, DeFi protocols, public chain foundations, validators, searchers, and more. New twists keep popping up, sparking interesting research questions.
• How should the MEV ecosystem of a public blockchain be designed? This is a typical multi-objective optimization problem with no absolutely correct answer. Therefore, by examining the current status and future development of the MEV ecosystems across various Layer 1 blockchains, one can reveal their value propositions and assess their priorities within the multi-objective optimization framework.
• The forms of MEV are diverse.There is no standardized definition of malicious MEV. However, “sandwich attacks,” also known as “sandwiches,” demonstrably undermine the interests of ordinary users. For an automated market maker (AMM) transaction to be sandwiched, two conditions must be satisfied: 1) the transaction is visible to the attacker, and 2) the user has set a high slippage tolerance during the AMM transaction, creating an opportunity for successful arbitrage. To avoid being sandwiched, users must either 1) enhance the privacy of their transactions or 2) reduce slippage tolerance, thereby diminishing the attacker’s arbitrage potential.
• This article analyzes Ethereum, BSC, and Solana as examples, evaluating the current status and future plans on MEV while exploring the value propositions they reflect：

• Ethereum prioritizes decentralization;

• BSC focuses on protecting users’ transaction experiences; and

• Solana emphasizes transaction efficiency and market competition.

ACK:

We express our gratitude to the Helius team (@heliuslabs) for their comprehensive report on Solana MEV and detailed sharing from the BNB Chain (@BNBCHAIN) Core Dev team. We also want to acknowledge the efforts of the #Flashbots team, the Jito team (@jito_labs), and the Eigenphi (@EigenPhi) team in promoting transparency regarding MEV, as well as the data contributors on Dune (@Dune), whose meticulous work has been cited in this article. Finally, we appreciate the contributions of teams such as TrustWallet (@TrustWallet), Pancake (@PancakeSwap), and GMGN (@gmgnai) for their efforts in user education and MEV protection.

MEV: An Enduring Topic with Continual Developments

On March 10, 2025, the Jito dashboard indicated a drop in the volume of bundle transactions and Tips on the Solana. Concurrently, some users reported an increase in the number of sandwich attacks on SOL, bringing the issue of MEV back into discussion as an anomaly.

Simultaneously, in recent months, the Binance Smart Chain (BSC) community has voiced frustrations regarding sandwich attacks (commonly known as sandwich bots) due to increased activity in memecoin trading. To tackle this issue and enhance user experience, BSC is actively optimizing its infrastructure by significantly reducing block times and adjusting consensus mechanisms to mitigate harmful MEV activities.

On Ethereum, the MEV issue remains persistent as well. In March 2025, an anonymous researcher named Malik672 proposed a “Decentralized Random Block Proposal” system, aiming to eliminate MEV by randomizing the block selection process. This system employs a shared randomness algorithm combined with Byzantine Fault Tolerance (BFT) mechanisms, ensuring all Ethereum clients (such as Geth and Nethermind) can participate in block construction, rather than limiting this capability to just a few large builders.

MEV is fundamentally a complex game involving multiple participants without perfect solutions. In decentralized blockchain environments, users, validators, searchers, and infrastructure providers each pursue their own objectives, often resulting in conflicting goals. The strategies each blockchain ecosystem employs reflect their unique priorities and value orientations.

Fig 1a. Jito Bundles and Tips data(2025.3.10)

Fig 1b. Jito data overview(2025.3.10)

Data from Jito’s dashboard indicates that prior to March 9, 2025, Jito handled around 13-20 million transaction bundles daily, generating roughly 10,000-15,000 SOL per day in tips. After March 9, daily tip revenue dropped to around 8,000 SOL. Despite this decline, Jito’s daily revenue remains substantial.

This recent incident raises fundamental questions: Why do users fall victim to sandwich attacks, and why are they willing to pay extra fees for services like Jito? What is the current state of MEV on various public blockchains?What might the future of MEV look like for BSC? Let’s move forward with these questions in focus.

Why Does MEV Happen on Every Blockchain?

MEV is a universal problem because most blockchains share a similar fundamental design. Let’s briefly revisit how typical blockchain transactions are processed:

• Users initiate transactions via wallets, which then forward these transactions to RPC nodes.
• RPC nodes subsequently relay transactions to validators (or nodes that themselves act as validators).
• Validators temporarily store pending transactions in an internal cache called a transaction pool (often termed Mempool, TxPool, or Tx Queue, etc).
• Once validators gain the right to produce a new block, they select transactions from the mempool based on specific criteria (usually transaction fees), then pack and broadcast them to the network.

Initially, validators prioritized transactions solely based on fees (Gas or Transaction fees). However, with the rise of decentralized finance (DeFi)—particularly Automated Market Makers (AMMs)—additional arbitrage opportunities emerged. These opportunities, captured by strategically ordering or inserting transactions, gave rise to the concept of Maximal Extractable Value (MEV).

One prevalent MEV tactic is the Sandwich Attack, primarily occurring on decentralized exchanges (DEXs) utilizing AMMs. In a sandwich attack scenario, an attacker places two transactions around a victim’s transaction: one just before the victim’s trade to manipulate the asset price unfavorably, and another immediately afterward to profit from the resulting price shift. Crucially, the profitability of sandwich attacks depends on the user’s chosen slippage tolerance—the maximum permissible deviation between the requested and executed price. Higher slippage settings increase vulnerability to these attacks.

A successful sandwich attack typically requires three conditions:

• Transaction visibility: User transactions are exposed within a transaction pool visible to attackers, enabling attackers to identify target transactions in advance.
• The attacker’s front-running and back-running transactions are successfully included by the validator in the same block (in most cases) or across consecutive blocks.
• Once the final block is confirmed and added to the blockchain, the price of the user’s transaction is significantly impacted by the attacker’s transactions, a phenomenon commonly referred to as “sandwiching.”

From the user perspective, two primary concerns arise:

• How can users prevent their transactions from being exposed to potential attackers (i.e., addressing the issue of transaction “visibility”)?
• How can users ensure that their transactions are packaged into blocks by validators more quickly and reliably, thereby reducing latency and uncertainty?

From validators’ standpoints, the focus shifts towards maximizing profits. This is reflected in their efforts to more effectively filter high-value transactions, and capture additional profits generated by MEV.

Then, two specialized roles—Searcher and Builder—have gradually emerged within different blockchain ecosystems, each tailored to optimize MEV extraction.

• Searchers actively scan transaction pools for MEV opportunities. They package profitable transactions (such as sandwich attacks, arbitrage, and liquidations) into bundles and offer additional tips to encourage prioritization.
• Builders(on PBS Blockchains) are responsible for filtering, ordering, and optimizing the transaction bundles submitted by Searchers. Their goal is to construct block structures that maximize value and are more readily accepted by Validators, thereby enabling Validators to capture MEV profits quickly and efficiently.

Validators typically do not refuse to collaborate with Builders and Searchers, as this model provides them with stable economic returns and establishes an effective profit-sharing chain. In fact, this multi-party game has gradually become a key component of the MEV ecosystem, continuously driving the optimization and competition of MEV infrastructure.

Each blockchain ecosystem develops distinct MEV solutions based on its unique characteristics. For instance, Jito, widely adopted on Solana, is an infrastructure specifically designed to address MEV issues by finding an optimal balance between Validators, Searchers, and users.

Solana MEV Ecosystem: Trading Efficiency and the Role of Jito

Jito as the Main MEV Mechanism: Faster transactions, users voluntarily pay for priority

Jito is one of the most primary MEV tools used on the Solana. Its core feature is a specialized private mempool where user transactions are temporarily stored in a private environment rather than immediately exposed publicly. This design effectively prevents attackers from seeing and front-running transactions, reducing the possibility of sandwich attacks. Jito also introduces an economic incentive system, allowing users to pay additional “tips” to validators to prioritize their transactions, thus improving both transaction speed and security.

In practice, Jito operates by enabling searchers to submit transaction bundles (Jito Bundles) accompanied by extra tips. Validators then prioritize these bundles during block construction. Over the past year, Jito processed more than 3 billion transaction bundles and generated over 3.75 million SOL in tips, significantly surpassing similar MEV solutions on Ethereum.

Due to Solana’s high throughput and short block times—combined with heavy Memecoin trading activity—MEV transactions typically appear in high-frequency, small-value formats. As a result, individual arbitrage or sandwich attack profits are relatively low per transaction but occur at massive volumes. For example, the average profit per sandwich attack on Solana is about 0.0425 SOL ($8.7 USD), far lower than Ethereum, yet the total transaction volume is extremely large.

In summary, MEV on Solana exhibits these key characteristics:

• High-frequency, low-profit trades: In 2024, Solana MEV bots executed around 90,445,905 arbitrage trades, averaging just $1.58 per trade in profit.
• Users willing to pay for speed: Jito users regularly pay additional tips to ensure their transactions receive priority. For instance, in November 2024, daily tips on Jito peaked at 60,801 SOL, showing that users accept higher MEV-related costs during active market conditions.
• High slippage due to front-running: Frequent traders, especially those using automated Telegram bots to trade Memecoins, typically set higher slippage tolerances to ensure their trades go through. This inevitably leads them to trade near the maximum slippage, voluntarily giving up some value to MEV bots. Data shows the average sandwich attack on Solana yields about 0.0425 SOL (~$8.7 USD).

Alternative MEV Solutions beyond Jito: Private Mempools

Although Jito is dominant, it doesn’t handle all MEV activities on Solana. Private mempools operated by validators also compete strongly for MEV opportunities:

• DeezNode Private Mempool: Some validators (such as DeezNode) run their own private mempools, allowing searchers to bypass Jito and pay validators directly for priority inclusion. In the past 30 days, this mechanism processed around 1.55 million sandwich attacks, generating total profits of 65,880 SOL (~$13.43M), averaging 0.0425 SOL per attack.
• Paladin-Solana anti-sandwich mechanism: About 6% of validators on Solana use Paladin, an alternative that proactively rejects sandwich transactions. Paladin also incentivizes validators with the PAL token to maintain fairness and reduce sandwich attacks.

From the user’s perspective, the primary concern is transaction speed rather than individual transaction cost details. Due to the frequent emergence of memecoins on Solana, many users rely heavily on automated bots for fast, timely trading. Users rarely focus on whether these bots integrate Jito or if their transactions are subjected to sandwich attacks—they simply want quick results.

Therefore, MEV activities remain prevalent and active on Solana, shaped largely by high-frequency trading and extensive bot usage. Rather than completely eliminating MEV, the ecosystem reflects a dynamic balance between users, bots, validators, and searchers. The existence of mechanisms like Jito demonstrates market-driven demand, representing an equilibrium among various participants seeking optimal returns within the MEV landscape.

Ethereum’s MEV: Solutions under Decentralization, Shifting Ecosystem Structure

Ethereum has consistently been a primary focus of MEV challenges. To mitigate on-chain issues such as sandwich attacks, researchers from the Ethereum Foundation have proposed the Proposer-Builder Separation (PBS) model and pursued ongoing research and development to safeguard the Ethereum ecosystem from the adverse effects of MEV. In collaboration with the ecosystem infrastructure project Flashbots, the Foundation employs the PBS model to address the needs of various participants. Flashbots has implemented a transparent and permissionless auction mechanism to standardize MEV extraction, aiming to increase transparency while promoting a more equitable distribution of profits among stakeholders, including validators and users.

According to Flashbots data, MEV revenue on the Ethereum mainnet averaged over $500,000 per day in 2023. By 2024, the rapid expansion of the Ethereum Layer 2 ecosystem redirected some MEV trading opportunities, stabilizing Layer 1 MEV revenue at approximately $300,000 daily as of the latest reports.

However, since entering 2025, the MEV ecosystem on Ethereum, while remaining active, has exhibited a marked decline in overall profitability. Data recorded on March 4, 2025, reveals that sandwich attacks constituted $289.76 million, or 51.56% of the total MEV transaction volume of $561.92 million. Despite this significant volume, the profit generated was only $6,320, accounting for just 4.11% of total MEV profits. This figure underscores a substantial decrease in the per-transaction profitability of sandwich attack strategies. Over the same period, total MEV costs on Ethereum rose by 28.36% to $358,850, while total revenue increased by only 6.90% to $512,660. Consequently, net profit contracted significantly to $153,810. These findings suggest that, despite the persistent frequency of MEV transactions, escalating competition, rising costs, and enhanced infrastructure to counter sandwich attacks are collectively reducing the overall profit margin.

MEV on Ethereum Dominated by Institutions and Whales

Due to high gas fees on Ethereum L1, retail users prefer to trade smaller amounts on Layer 2 networks (like Base, Arbitrum) or other low-cost blockchains. As a result, Ethereum’s primary MEV participants are now institutional players, large-scale traders (whales), and professional market makers. These large MEV trades underline Ethereum’s position as a key liquidity center in DeFi, but they also lead to significant slippage, creating ongoing arbitrage opportunities for MEV bots.

The reduced profitability of sandwich attacks in 2025 can be attributed to:

• Increased competition: More MEV bots competing over limited arbitrage opportunities have significantly compressed simple strategy profits.
• Improved trading strategies by institutions: Institutions widely adopted strategies like TWAP (Time-Weighted Average Price) and DCA (Dollar-Cost Averaging), breaking down large trades to minimize MEV risks.
• Widespread adoption of anti-MEV solutions: Private transactions, batch auctions, and Order Flow Auctions (OFA) reduce attackers’ ability to capture slippage and perform sandwich attacks.

Ethereum’s MEV Future: Complex Strategies Become the Norm

With the rapid growth of Ethereum’s L2 ecosystem, MEV opportunities are shifting from L1 to L2. However, Ethereum L1 remains the core venue for large-scale institutional DeFi activities, so MEV will continue evolving into more complex strategies. Emerging MEV forms include:

• Cross-chain Arbitrage between L1 and L2 networks.
• Liquidations: Increasingly critical as DeFi lending grows.
• New mechanisms like Order Flow Auctions (OFA), shifting MEV profitability toward collaboration with liquidity providers rather than exploiting user slippage alone.

Overall, the MEV ecosystem on Ethereum is undergoing structural transformation. Since the identification of the MEV issue, Ethereum has consistently explored various solutions, including architectural proposals such as Proposer-Builder Separation (PBS). Although the profit margins of straightforward strategies, such as sandwich attacks and front-running arbitrage, have been substantially reduced, more intricate and specialized MEV strategies continue to emerge and evolve. This suggests that the MEV issue will persist on the Ethereum mainnet over the long term. Moreover, the ongoing interplay of interests among searchers, block builders, validators, users, and MEV-related infrastructure projects (e.g., Flashbots) is expected to endure. Competition centered on transaction ordering, value extraction, and fairness will drive the continuous evolution of the MEV ecosystem.

BSC’s MEV Ecosystem: Rapid Growth Brings New Requirements for Transaction Experience

Although MEV issues on BSC frequently attract community attention, what is the actual situation?

According to data from Dune, since the second half of 2024, the proportion of sandwich attacks in all DEX transactions on BSC gradually increased and surpassed Ethereum for the first time in December 2024. However, over the long term, the proportion of sandwich-attacked DEX transactions on both BSC and Ethereum has consistently stayed below 8%. After reaching a high in February 2025, the proportion on BSC has decreased back to around 4%.

Fig 2. ETH Sandwiched DEX Transaction vs BSC ETH SandWiched DEX Transaction

The data shows BSC’s sandwich attack proportion isn’t significantly higher than Ethereum’s, maintaining a similar level overall. So why do many BSC users frequently feel targeted by sandwich attacks? The main reason is the recent increase in trending tokens on BSC, leading to higher user transaction activity and heightened awareness of MEV.

To better understand this, let’s briefly review why sandwich attacks happen:

• Transaction pool visibility: Once a user transaction appears in a public mempool, attackers can identify it and execute sandwich attacks.
• High slippage settings: Users set high slippage tolerance in AMM trades to ensure transactions go through, inadvertently creating arbitrage opportunities.

During periods of highly on-chain trading activity for trending tokens, users primarily aim to complete their transactions as quickly as possible. Consequently, they often set higher slippage tolerance, indirectly providing more room for sandwich attacks. While many wallets and nodes offer MEV protection (like private mempools or private RPC endpoints), not all users choose these private transaction options. Aside from users overlooking MEV protection settings, private transactions do not necessarily speed up transaction confirmation. Conversely, during peak traffic, transactions in the public pool can actually be picked up faster if users increase their transaction fees slightly. Regular stablecoin transfers or BNB transfers usually don’t require additional MEV protection and might even perform better in public pools.

From a technical perspective, the current MEV infrastructure on BSC resembles Ethereum’s PBS model, where transaction ordering is created by builders and submitted to validators. Some builders offer private mempool services, but transactions in public pools can still be targeted by searchers who construct attack bundles. Unlike Solana’s private pools, BSC’s private transaction pools primarily protect transaction privacy without significantly improving transaction confirmation speed or priority.

Therefore, effectively addressing MEV requires two key optimizations:

1. Improving transaction privacy: Reduce the likelihood of transactions being visible in public pools to lower sandwich attack risks.
2. Enhancing blockchain performance: Shorten block intervals and increase throughput, allowing quicker transaction inclusion and reducing exposure to MEV.

How is BSC Addressing MEV?

In the long run, BSC aims to reduce negative MEV impacts through technical upgrades and performance optimizations. According to the latest roadmap, BSC plans to shorten block intervals to under 750ms, officially entering the sub-second block era. This improvement will directly:

• Enhance user experience: Transactions confirm faster, reducing the time they remain exposed in transaction pools and lowering MEV attack risks.
• Improve transaction security: Faster confirmations reduce the chance of failed transactions due to price volatility. Users won’t need to set excessively high slippage, indirectly reducing arbitrage space for MEV bots.

Thus, while MEV activity on BSC continues, its scale isn’t notably worse than other chains. BSC clearly plans to tackle MEV by moving towards sub-second blocks (750ms) to improve both blockchain performance and user experience.

Besides shortening block intervals, BSC actively explores more comprehensive private mempool solutions, like leveraging TEE technologies, to balance the interests between users, searchers, and validators.

Summary

MEV is a universal, complex problem faced by every blockchain. Currently, each ecosystem adopts different strategies to balance participant interests:

• Solana uses Jito and other private mempools to reduce visible transactions, introducing tip mechanisms to speed up confirmations.
• Ethereum implements PBS (Proposer-Builder Separation) to make MEV competition market-driven and transparent.
• BSC improves blockchain processing power and reduces block times, enhancing transaction experience and reducing MEV exposure.

For ordinary users, regardless of the blockchain used, it’s advisable to connect to private transaction pools or private RPC nodes when performing AMM trades. This helps minimize exposure to public transaction pools and reduces the likelihood of experiencing MEV attacks.

Disclaimer:

1. This article is reprinted from [YZi Labs]. All copyrights belong to the original author [Siyuan H (@cyodyssey), Dana H (@danabuidl)]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. The Gate Learn team does translations of the article into other languages. Copying, distributing, or plagiarizing the translated articles is prohibited unless mentioned.