In cryptocurrencies, MEV typically refers to bots and searchers on chains such as Ethereum that reorder, insert, or censor transactions related to DEX trades and liquidations, extracting value from users in the form of deteriorating prices, failed transactions, and increased costs.
But did you know that Bitcoin also has MEV-like dynamics in its memory pool and policy layer? It’s like its own quiet version of MEV without the front-running swaps by DeFi-style bots. Instead, miners and pools use price signals, memory pool policies, and block templates to determine which transactions are cleared first.
In Bitcoin Core v28, the full paid exchange memory pool policy is turned on by default (mempoolfullrbf=1) and a limited 1-parent-1-child package relay was added. Miners and pools running core or compatible software inherit these defaults, but can choose alternative policies.
However, the public memory pool is only part of the auction that determines which transactions will be settled in the next block, as out-of-band routes to the pool and wallet-level fee controls also play a role.
Within the Bitcoin network, miners and pools are the de facto decision makers. Ultimately, the memory pool used and policy settings determine which consensus-valid transactions are included in the block.
Conclusion: Bitcoin has a soft form, MEV, for everyday users. Small fee changes, package construction (parent + child), and a direct path to the pool can cause a transaction to move ahead of other transactions, even if the transaction is broadcast in the first place.
When miners assemble block templates, transactions are effectively selected in the following rough order:
- Transactions or packages that they have confirmed and consensus validated.
- The package with the highest effective commission rate when combining ancestors and children.
- Exchanges that pay more than competing transactions under BIP125.
- Out-of-band transactions or pool-level policy filters that override pure commission rates.
In reality, this is how miners silently decide which transactions will “win” in the next block.
Bitcoin’s “soft MEV” is quiet compared to Ethereum or DeFi MEV, where searchers run arbitrage, sandwiches, and liquidation bots to extract value from smart contract interactions.
There is no front running of DEX swaps or liquidation auctions. Instead, miners and pools coordinate orders through fee-based incentives, package selection, and occasional off-chain payments. This contrast is why this MEV is difficult to catch the average user’s eye.
How miners choose winners within the Menpool
Recent fees and memory pool dataframes why small order edges matter. According to YCharts, the average on-chain fee is $0.68, which is down from last year.
The October hourly window saw bursts and near-empty gaps in mempool.space’s block fee rate view, resulting in periods where small deltas in absolute fees could move transactions to the top of the template.
According to Hedge With Crypto, fees will fall to around 0.96% of block rewards in June 2025, the lowest share since January 2022. According to BitInfoCharts, the hash rate is hovering around 1.1 zettahashes per second, and the competition for increasing template yields is steady.
With ancestral fee mining and package relay, actual fee auctions are increasingly package-based rather than simple transaction-by-transaction.
Starting with Bitcoin Core’s ancestor fee rate mining (PR #7600), block templates take into account the combined fee rates of ancestor and descendant packages. That’s why in CPFP, a low-fee parent and a high-fee child can win on isolated high-fee transactions.
This is why the child paying parent periodically pulls the stuck parent into the block when the combined package clears the minor threshold.
According to the summary of the v28 release of No Bullshit Bitcoin, the default full RBF means that unconfirmed transactions can be replaced by a higher priced version that pays more than all competitors and the bandwidth increment set by BIP125.
The same release also introduced opportunistic one-parent, one-child package relaying, made TRUC (version 3) transactions and P2A output standard by default, and also added a limited form of package RBF.
Later core versions (v29 and later) keep full RBF as the default memory pool policy and continue to evolve package relay.
Out-of-band toll lanes, policy filters, soft MEV
Out-of-band payment rails widen the gap between public memory pool orders and what is mined. ViaBTC’s accelerator sends transactions directly to the pool. This is a path that allows transactions to be promoted at lower in-band fee rates since the missing fees are paid off-chain.
These arrangements can distort template selection and reduce transparency if they occur frequently, as on-chain fees alone cannot account for inclusion.
Miningpool.observer exposes template and block pairs and highlights missing or redundant transactions and conflicts. This provides public evidence of inclusion selection that is inconsistent with the simple maximum fee view.
Policy filters control the effectiveness of relays, but not the effectiveness of consensus. This is the second means of influencing which transactions reach miners on time. Standardity policies are not consensus rules. Miners can include consensus-enabled transactions even if relay nodes drop them.
Recent changes to OP_RETURN demonstrate how propagation is formed by default. The developers merged changes for the v30 cycle, removing the long-standing default 80-byte maximum limit for OP_RETURN in policies, increasing the default data carrier size, and subsequently tweaking how node operators configure that size.
Soft MEV in Bitcoin’s long-term fee economy
The public episode also shows arbitrary filtering at the pool layer. OCEAN chose to filter data for inscription formats, and Marathon’s 2021 OFAC compliance experiment showed that template selection can deviate from pure top-fee rankings if the pool pursues policy or public relations goals.
The rules governing exchanges and packaging establish practical limits on priorities. BIP125 requires exchangers to pay higher absolute fees than all competitors and also covers minimal incremental transit fees.
However, RBF rules (including BIP125) are menpool policies, not consensus. Miners can always mine the first consensus-valid replacement they find.
Wallets with increasing fees often aim to jump to the next block’s fee bucket with a substantial increase to avoid repeated churn, but this is a heuristic rather than a rule. CPFP is a straightforward way to raise fees when a parent gets stuck, and v28’s 1-parent, 1-child relay increases the likelihood that the child sponsoring the fee will quickly arrive at the peer pool to change the next template.
According to the opt-in RBF FAQ, zero confirmation acceptance remains a risk, and the risk will increase as full RBF is widely deployed. This is because nothing is checked globally first, and due to asynchronous relaying, the replacement may reach the miner before the original reaches the miner’s template builder.
What this means for everyday users
From a wallet user’s perspective, small decisions about how fees are set or how transactions are structured can silently move miners up or down the queue.
Queue jumping over RBF is common, and higher-fee alternatives may overtake previous broadcasts. CPFP allows you to sponsor a stuck parent by paying from your child, increasing the effective rate of the package. Accelerators with direct pool access serve as emergency lanes when public pools are crowded.
In reality, small price differentials and package structures are the “soft MEV” edges that determine who clears first.
Consider two similar transactions. Alice sends the payment with a modest fee, but Bob uses RBF to increase the fee by a few Sat/vB. Even if Alice broadcasts first, Bob’s top substitutes can jump to the next block under BIP125.
Or imagine a stuck parent transaction is rescued by a child. When attaching children at a high price, the combined package is often included faster than a single expensive transaction with no dependencies.
Similarly, if you use a pool accelerator and pay out-of-band fees, you may still win on transactions with lower on-chain fee rates.
Improved template visibility and narrowed the information gap for soft ordering choices. Bitcoin Optech mentioned its work on a cluster menpool heuristic to detect fee increases in block templates, and a proposal for nodes to share templates so miners can compare what to include.
These ideas are intended to make it easier to spot deviations from price maximization, whether due to OOB compensation, policy filters, or simple delays.
The forward path will depend on fee levels and burst frequency, with incentives growing as the block subsidy shrinks below 3.125 BTC in future halvings.
If the average fee stays around $1-2 and the fee share is near the low single digits, most of the soft MEV activity will come from moderate RBF bumps and CPFP near the anchor, and OOB will be used as an emergency lane.
If bursts occur repeatedly around inscriptions, headlines, or a more lenient OP_RETURN policy environment, the average rate can jump into a higher bracket in a short window. Fee shares may reach high single digits on spike days, and out-of-band paths and package bids will become more apparent in template and block diffs.
If a persistent high-fee regime emerges and the fee share is on an upward trend, Carlsten et al.’s theory follows. Bitcoin’s large hashrate and pool structure actually discourage execution, but make time bandit incentives more important.
The mechanism remains simple. Miners use RBF and CPFP as specified in BIP125 to build templates using ancestry-aware scoring, wallets, and service source fees. Package relays were introduced in Core v28 and later, and OOB lanes provide a priority direct channel to the pool.
This is Bitcoin’s quiet MEV. Miners and pools do not front-run swaps, but use fees, packages, and side channels to silently pick winners within the memory pool.
(Tag translation) Bitcoin
