The Express Lane to Spam and Centralization: An Empirical Analysis of Arbitrum's Timeboost

arXiv:2509.22143 · cs.CR · Submitted 2025-09-26 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Security Radio. Generated commentary on the latest security and cryptography papers.

Nadia: Today's paper: "The Express Lane to Spam and Centralization".

Elias: The gist: Timeboost fails to deliver on its stated goals of fairness, decentralization, and spam reduction.

Nadia: First, who's behind it and why it matters.

Paper summary: Elias: So looking at this paper again, "The Express Lane to Spam and Centralization: An Empirical Analysis of Arbitrum's Timeboost," the authors are essentially saying that auction-based transaction ordering rights don't automatically deliver the benefits commonly attributed to them >

Nadia: They’re pointing out that instead of broadening participation or reducing strategic behavior, Timeboost actually concentrates access among a small set of sophisticated actors and creates economic conditions that challenge both secondary markets and protocol revenue >

Priya: I think what this really tells us is that the design tension they mentioned is real—while auctions might eliminate direct latency races, they also favor well-capitalized participants rather than spreading the power out >

Elias: And when you look at the numbers, like how three entities secured ninety-nine point seven four percent of auctions, it shows that the mechanism itself isn't inherently fair in practice > <ref:2509.22143#pg1>

Nadia: It seems to be a mechanism that currently allows for collusion and narrows adoption rather than achieving the decentralization they aimed for with Timeboost >

Priya: So what does this mean for the future of rollups? It suggests we need to keep looking at these auction designs because they aren't solving the problems of spam or centralization, even when you give them an auction >

Elias: Yeah. The paper also mentions that they plan to release their datasets and analysis scripts to help with reproducibility for future research on these auction mechanisms >

Nadia: That’s a good point. It shows there’s still work to be done on understanding how these systems behave under real stress, especially regarding spam mitigation >

Priya: So the big picture is that Timeboost, as implemented so far, hasn't quite succeeded in its mission to create a fairer transaction environment for everyone >

Conclusion: Nadia: So we’re wrapping up this look at Timeboost, which is basically the paper’s main title: "The Express Lane to Spam and Centralization."

Elias: Yeah, that title hits exactly what they found—it's about how these auction mechanisms create a high-speed lane that ends up being controlled by a few big players.

Priya: I think it’s important to unpack what they mean when they talk about the express lane itself, because for someone just listening, it sounds like a fancy way to get ahead of the crowd.

Nadia: Right, so Timeboost isn't just about speed; it's about who gets access to that speed and whether that access is fair or if it’s already rigged.

Elias: Exactly, they show this whole thing isn't solving the spam problem as well as people hoped because the control is super concentrated among three entities winning almost all of the auctions.

Priya: That concentration raises a big question for us about decentralization—if only a handful of actors are controlling the priority access, isn't that not what we want in DeFi?

Nadia: It really changes how you look at rollups; it suggests that even well-designed auction systems can still lead to cartel behavior if the bidding structure favors certain kinds of capital.

Elias: And they point out that because of this centralization, secondary markets trying to buy those express lane rights just aren't sustainable anymore because the economics are broken.

Priya: So what does this mean for the day-to-day user who isn't one of those big actors? They’re getting a deterministic delay before they even get to their transaction, which cuts into that promise of fairness.

Nadia: It means Timeboost currently leans more toward internalizing revenue for the system rather than distributing it fairly across the whole ecosystem.

Elias: That leads us right into what they’re suggesting about the limitations of auction-based ordering mechanisms in rollups, and whether this approach is actually helping or hurting long term growth.

Max Planck Institute for Software Systems (MPI-SWS) · INESC-ID / Instituto Superior Técnico (IST), University of Lisbon

cs.CR

Submitted: 2025-09-26

Updated: 2026-10-09

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 79/100

The gist: The gist: Timeboost fails to deliver on its stated goals of fairness, decentralization, and spam reduction.

Key concepts

Timeboost: A transaction sequencing mechanism that uses a sealed-bid second-price auction to grant short-term priority access to a low-latency express lane. This replaces standard first-come, first-served ordering with a market where users bid for faster processing.
Express Lane Control Centralization: The finding that only three entities—Selini Capital, Wintermute, and Kairos—won 99.74% of the auctions controlling express lane access. This extreme concentration suggests the auction mechanism favors sophisticated actors with optimized bidding strategies rather than ensuring fair or decentralized access for all participants.
MEV Mitigation Failure: The analysis found that only about 30% of timeboosted transactions reverted, indicating that Timeboost does not successfully mitigate MEV inefficiencies like spam. This suggests the system is ineffective at reducing transaction spam and reordering by specialized actors.

Terminology

Actor — 1× in this paper · explained in 3 episodes

Summary

The gist: Timeboost fails to deliver on its stated goals of fairness, decentralization, and spam reduction.

How it works

Timeboost is an auction-based transaction sequencing mechanism that replaces the traditional First-Come-First-Served (FCFS) policy with a market for sequencing priority It organizes time into one-minute rounds and allocates access to an express lane through a sealed-bid second-price auction The winning bidder obtains exclusive access to a low-latency submission channel, while regular user transactions incur a deterministic delay of 200ms before reaching the sequencer

Key Empirical Findings

The empirical analysis of Timeboost reveals several critical outcomes regarding its design and impact on the Arbitrum ecosystem

** Express lane control is highly centralized, with three entities winning 99.74% of auctions This dominance suggests that auction-based allocation does not necessarily lead to broad participation The dominant entities include Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds Furthermore, the concentration raises concerns about fairness and long-term accessibility for smaller or new participants The auction mechanism disproportionately favors sophisticated actors with optimized bidding strategies or superior capital allocation The distribution of express lane controllers shows a handful of entities dominating controller access, collectively securing 99.74% of winning rounds This concentration undermines fairness and raises risks of collusion or cartel-like behavior The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.74% of winning rounds The dominance suggests that auction-based allocation does not necessarily lead to broad participation While Timeboost nominally rotates express lane control each round, in practice access is centralized The dominant entities are Selini Capital, Wintermute, and Kairos, which collectively secured 99.

Improvements for AI systems

No direct mitigation of spam through express lane access, as timeboosted transactions often contain a high number of reverts and express lane submission does not guarantee execution. The improved AI system could implement a dynamic threshold for express lane acceptance based on real-time execution reliability metrics to filter out low-probability failed submissions.
Mitigation of controller concentration through participation constraints, as the paper suggests, a handful of entities, primarily Selini Capital and Wintermute, and later Kairos, won more than 99.74% of auctions. The AI system could incorporate a mechanism where a controller could be prevented from winning more than X rounds within a Y-minute interval to prevent persistent privileged lane access.
Improving auction fairness by diversifying allocation methods, as the paper notes that the sealed-bid second-price auction naturally favors highly capitalized actors capable of consistently overbidding competitors. The system could introduce randomized allocation among the top-k bidders rather than deterministically assigning the lane to the single highest bidder to encourage broader participation.
Addressing economic inefficiency by integrating dependency checks into secondary markets, as suggested by the need to reduce risk in resale, the AI system could move resale functionality directly into the protocol by reducing the current round time from one minute to shorter intervals and enforcing dependency checks.
Enhancing MEV strategy awareness by dynamically adjusting priority access based on block positioning, since profitable MEV opportunities cluster at the end of blocks rather than at the beginning. The system could analyze in-block positions and adjust express lane priority rules to favor transactions positioned at the top of blocks, where value is often higher.

Abstract

DeFi applications are vulnerable to MEV, where specialized actors profit by reordering or inserting transactions. To mitigate latency races and internalize MEV revenue, Arbitrum introduced Timeboost, an auction-based transaction sequencing mechanism that grants short-term priority access to an express lane. In this paper, we present the first large-scale empirical study of Timeboost, analyzing over 48.5 million express lane transactions and 494 thousand auctions between January 2025 and April 2026. Our results reveal five main findings. First, express lane control is highly centralized, with three entities winning 99.74% of auctions. Second, while express lane access provides earlier inclusion, it is currently mainly used for CEX-DEX arbitrage, limiting the value of priority access for other types of MEV extraction. Third, approximately 30% of timeboosted transactions revert, indicating that Timeboost does not effectively mitigate MEV inefficiencies such as spam. Fourth, secondary markets for reselling express lane rights struggle to sustain themselves due to poor execution reliability and unsustainable economics. Finally, auction competition declined as secondary markets introduced dedicated CEX-DEX execution lanes, enabling pay-per-transaction instead of pay-per-round billing and leading to steadily reduced revenue for the Arbitrum DAO. Taken together, these findings show that Timeboost fails to deliver on its stated goals of fairness, decentralization, and internalizing long term revenue for the DAO. Instead, it currently allows for collusion and narrows adoption, highlighting the limitations of auction-based ordering as a mechanism for fair transaction sequencing in rollups.