Based Rollups: Why L1 Sequencing Creates Trustless L2s
Based rollup L1 sequencing is the paradigm where Ethereum's own block proposers (validators) are responsible for ordering transactions on a Layer 2 rollup, rather than a dedicated sequencer set. This design, first articulated by Justin Drake in 2023, redefines trust minimization by aligning the L2's sequencing with the L1's consensus. Instead of relying on a separate permissioned or even decentralized sequencer that users must trust for liveness and censorship resistance, a based rollup outsources sequencing to the Ethereum base layer. This article provides an advanced analysis of based rollup architecture, focusing on the economic security guarantees provided by the L1 sequencer set and the unique composability benefits that emerge from this design choice.
By inheriting the full security properties of Ethereum's proposer-builder separation (PBS) and its robust validator set, based rollups offer a path to truly trustless L2s. However, this architecture introduces novel challenges around transaction ordering, MEV dynamics, and inclusion latency. We examine real implementations like Taiko, as well as research from Espresso and other rollup teams, to dissect how L1 sequencing transforms the rollup landscape. Whether you are a rollup developer, DeFi power user, or protocol researcher, understanding based rollup L1 sequencing is critical for evaluating the next generation of Ethereum scaling solutions.
- Based rollups outsource sequencing to Ethereum L1 proposers, inheriting L1's security and decentralization.
- L1 sequencing provides atomic composability across multiple based rollups and with L1 itself.
- The economic security of based rollups is backed by the entire L1 staked ETH, not a smaller rollup-specific bond.
- Based rollups eliminate trust in a separate sequencer set, reducing the attack surface.
- Latency (12+ seconds) is the main trade-off; pre-confirmation techniques are being researched to mitigate.
- Real-world implementations like Taiko are already live; based rollups are becoming a standard architecture.
What Is a Based Rollup?
A based rollup is a Layer 2 scaling solution where the sequencing of L2 blocks is performed by the L1 proposers (validators) of the base blockchain, typically Ethereum. Unlike traditional optimistic or ZK rollups that operate their own sequencer (often a single entity, a federation, or a decentralized set of operators), a based rollup reuses the L1's sequencing mechanism. The term 'based' was coined by Ethereum researcher Justin Drake. In a based rollup, any L1 proposer can include a 'rollup block' in their L1 block, effectively ordering L2 transactions. The rollup contract on L1 validates the state transition (via fraud proof or validity proof) and ensures that the L2 state evolves correctly. This means the rollup does not need to run its own consensus or pay sequencer rewards; it piggybacks on Ethereum's security.
How L1 Sequencing Works in Practice
In a based rollup, sequencing occurs at the time of L1 block construction. An L1 proposer (validator) gathers L2 transactions either from a public mempool or through a designated inbox contract. The proposer packages these transactions into a rollup block and includes its hash within the L1 block. The rollup's smart contract on L1 then verifies the correct execution (via a validity proof for ZK-based based rollups, or via a fraud proof window for optimistic ones). For example, Taiko uses a based design where L1 proposers can submit L2 blocks; Taiko's contract enforces constraints like maximum gas and allowed sequencers. The key insight is that sequencing power is distributed across the entire set of Ethereum validators (hundreds of thousands), making it permissionless and censorship-resistant. This contrasts with traditional rollups like Arbitrum or Optimism, where a centralized sequencer (even if eventually decentralized) exists.
Economic Security of the L1 Sequencer Set
The economic security of a based rollup's sequencing derives directly from Ethereum's consensus. L1 proposers have an economic incentive to include valid rollup blocks because they earn priority fees from L2 users. If a proposer includes an invalid block, it will be rejected by the rollup contract, and the proposer loses the fee (and in some designs, gets slashed). This aligns the proposer's profit motive with honest sequencing. The total economic security is the L1's security budget: the market value of ETH staked, which is in the tens of billions. Compare this with a separate rollup sequencer set, which might require millions of dollars in bonded tokens to achieve similar safety. Furthermore, because the L1 validator set is already well-distributed and tested, there is no additional trust assumption for sequencer honesty, liveness, or decentralization. The based rollup simply inherits these properties.
Composability: Atomic Cross-Rollup Transactions and Beyond
Based rollups unlock unique composability properties because all L2 transactions are included in L1 blocks. This allows for atomic cross-rollup transactions: a user can include actions on multiple based rollups in the same L1 block, as long as they are processed sequentially. For instance, a DeFi arbitrage could swap on Based Rollup A and then deposit to Based Rollup B in the same block, with the guarantee that either both happen or neither does. This is similar to L1 atomic composability but scaled. The Espresso Sequencer project explores this concept with Ethereum-based sequencing to enable cross-rollup composability. Additionally, standard L1 composability means that an L1 contract can synchronously call a based rollup's inbox and receive a response, enabling more integrated DeFi. This is not possible with traditional rollups due to block time and sequencing delays.
MEV Dynamics in Based Rollups
In traditional rollups, the sequencer has exclusive access to the order flow and can extract MEV (Maximal Extractable Value) through reordering, insertion, or censorship. This creates a centralization vector and a value capture opportunity for the sequencer. In based rollups, MEV extraction is democratized: any L1 proposer can include L2 transactions, and they compete in the L1 MEV market. The proposer can either run its own L2 searcher or outsource to builders via L1 PBS. Because L2 transactions are just part of the L1 block, the existing MEV-boost infrastructure can be used. However, there is a subtlety: L2 MEV is often dependent on L2 state that is not fully visible to L1 searchers. This leads to the concept of 'based sequencing MEV' where L2-specific searchers need to submit bundles to L1 proposers. Taiko's design includes a 'proposer auction' where L2 block proposers pay L1 proposers for inclusion, similar to how builder payments work.
Liveness and Latency Trade-offs
A key trade-off of based rollups is latency. L1 block time is 12 seconds (on Ethereum), whereas traditional centralized sequencers push blocks every 0.5-2 seconds. This makes based rollups unsuitable for low-latency applications like high-frequency trading. However, included L2 transactions are final once the L1 block is confirmed, offering strong finality. Some projects are exploring 'pre-confirmations' where the L1 proposer promises to include a transaction, giving users fast soft confirmation. Espresso is working on a pre-confirmation layer that could work with based rollups. Another liveness concern: if no L1 proposer includes L2 transactions for a few blocks, the rollup stalls. However, because inclusion is permissionless, a rational proposer will include profitable transactions. If fees are low, some epochs may have gaps. This can be mitigated by a 'sequencer tax' that incentivizes inclusion even for low-fee periods, but that adds complexity.
The Role of Validity Proofs in Based Rollups
Based rollups can be either optimistic (using fraud proofs) or zero-knowledge (using validity proofs). For based ZK-rollups, the L2 block submitted by the L1 proposer must include a validity proof that the rollup contract verifies on L1. This ensures immediate finality. For based optimistic rollups, the block is accepted with a challenge period. In both cases, sequencing is separated from proving. Note that the L1 proposer does not need to generate the validity proof; they can outsource that to specialized provers. For example, a based zkRollup could have a prover market where provers compete to generate proofs for pending L2 blocks, similar to how builders compete in PBS. This separation allows the rollup to benefit from L1's sequencing security while still achieving fast proving (via SNARKs). The Taiko protocol uses a based rollup design that relies on validity (ZK) proofs.
Comparison with Other Rollup Types
To fully appreciate based rollup L1 sequencing, it helps to contrast it with other architectures:
- Traditional Centralized Sequencer: Used by Arbitrum, Optimism (currently). Single entity orders transactions. Pros: low latency, cheap. Cons: trust for liveness, censorship. Some have escape hatches but not user-friendly.
- Decentralized Sequencer (e.g., Espresso, Astria): Separate PoS chain, often using shared sequencing. Pros: decentralized, but introduces new trust assumptions and token economics. Composability only within that sequencer set.
- Based Rollup: Uses L1 validators. Pros: trustless, inherits L1 security, atomic composability with other based rollups and L1. Cons: higher latency, more complex MEV.
Challenges and Open Research
Despite the advantages, based rollup L1 sequencing faces hurdles. One is the issue of 'sequencer fee divergence': L1 proposers may not include L2 transactions if fees are low relative to L1 congestion, leading to rollup liveness issues. Solutions include a minimum base fee enforced by the rollup contract, or a cross-block inclusion incentive. Another challenge is 'block space competition' between L1 and L2 transactions. With high L1 demand, L2 transactions may be crowded out. Furthermore, the integration with L1 PBS requires careful design to prevent MEV leakage between layers. Research into 'based sequencing with pre-confirmations' by Espresso and others looks promising. Lastly, there is the question of how to handle forced inclusion (escape hatch) when L1 proposers collude to censor. Since any L1 proposer can include, collusion would need a majority of proposers, which is unlikely given Ethereum's decentralized nature but not impossible. Economic penalties similar to slashing could be added for proposers that ignore forced inclusion requests.
Future Outlook: Based Rollups as the Standard
Based rollups represent a natural evolution in the rollup design space. As Ethereum moves toward proposer-builder separation (PBS) and ePBS, the infrastructure to support based sequencing becomes more efficient. I expect that many new rollups will adopt based designs to minimize trust assumptions and leverage existing L1 security. The atomic composability feature could enable a 'rollup ecosystem' where L1 is the central nexus for ordering, enabling seamless interactions across different L2s without bridges or relays. Additionally, the rise of based rollups could reduce the need for rollup-specific tokens used for sequencing, as there is no need to incentivize an external set of sequencers. Projects like Taiko are mainnet-ready, and others are in development. For advanced users and developers, understanding based rollup L1 sequencing is essential to navigate the future of Ethereum scaling.
Common mistakes to avoid
- Assuming based rollups require their own validator set: they reuse L1 validators. No additional staking needed.
- Confusing based rollups with 'optimistic' or 'ZK' – based is a sequencing model, not a validity model.
- Thinking based rollups have instant finality: they inherit L1 finality (12 sec for pre-confirm, but actual finality can be longer for optimistic).
- Assuming based rollups are always cheaper: competition for L1 block space may make them more expensive than centralized sequencer rollups.
- Believing based rollups cannot have MEV protection: they can use L1 PBS and inclusion lists to protect users.
- Overlooking the need for a prover market in based ZK-rollups: sequencing and proving are separate.
Frequently asked questions
How does a based rollup differ from a traditional rollup?
A based rollup uses L1 validators to order L2 blocks, while traditional rollups run their own sequencer. This gives based rollups stronger security, censorship resistance, and composability, but higher latency.
Can a based rollup be both optimistic and ZK?
Yes, the 'based' attribute only refers to sequencing. Based optimistic rollups use fraud proofs; based ZK-rollups (e.g., Taiko) use validity proofs. Both rely on L1 sequencing.
What are the benefits of L1 sequencing for DeFi?
Atomic composability with L1 and other based rollups in the same block enables trustless cross-L2 transactions without bridges, improving capital efficiency and simplifying MEV extraction.
Is based rollup L1 sequencing suitable for high-frequency trading?
Not currently, due to 12-second block times. Pre-confirmations could reduce this to ~1-2 seconds, but not to sub-second levels typical of centralized sequencers.
What protocols implement based rollup L1 sequencing?
Taiko is the leading example. Other projects like Espresso and Sovereign Labs explore based sequencing with shared pre-confirmations.
Related reading
Track the entities behind the concepts
DeFi Intel maps 11,000+ protocols, tokens and companies to a typed knowledge graph — with live data, incidents and regulation.