The Express Lane to Spam and Centralization: An Empirical Analysis of Arbitrum's Timeboost
This empirical study of Arbitrum's Timeboost mechanism reveals that despite its goal of fair transaction sequencing, the system has resulted in extreme centralization, failed to mitigate spam or support a viable secondary market, and ultimately reinforced collusion while reducing DAO revenue.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine the internet as a giant, bustling digital marketplace where people trade digital assets like tokens and coins. In this world, there's a sneaky class of traders called "searchers" who act like high-speed racers. They use super-fast computers to spot tiny price differences between different shops (exchanges) and race to buy low and sell high before anyone else can. This is called "Maximal Extractable Value" (MEV). The problem is, these racers often cheat by bribing the traffic controllers (the people who decide the order of transactions) to let them cut in line, pushing regular users to the back and stealing their profits. To fix this, the creators of a popular digital highway called Arbitrum built a new system called "Timeboost." Instead of letting the fastest computers win the race, they decided to hold an auction every single minute. If you want to skip the line, you have to bid money for a "fast pass" ticket. The idea was that this would make things fairer, stop the spammy racers, and give the money to the community instead of the cheaters. But does this new system actually work, or does it just create a new kind of problem?
This paper is like a giant detective report that looked at over 48 million transactions and nearly half a million auctions on the Arbitrum highway between January 1, 2025, and April 13, 2026, to find out. The researchers, who are like digital archaeologists digging through the history of the blockchain, discovered that the "Timeboost" experiment didn't quite turn out the way the designers hoped. It's important to note that the new system only started running in April 2025, so the data covers the period before and after its launch. Instead of a fair, open race where anyone could buy a fast pass, the system became a playground for just three giant players: Selini Capital, Wintermute, and Kairos. These three entities won a staggering 99.74% of all the auctions, effectively locking out everyone else. It's as if a school cafeteria auctioned off the "fast lunch line" ticket, but only three rich students could afford to bid, and they bought it almost every single day.
The report also found that having a fast pass didn't guarantee a smooth ride. About 30% of the transactions sent through this "express lane" actually failed and had to be thrown away (reverted). This means the system didn't stop the spam; it just moved the spam from the regular line to the fast lane. Furthermore, the researchers found that the most profitable trades often happened at the very end of the day, not the beginning. So, paying for a fast pass to get to the front of the line was often a waste of money because the best opportunities were waiting at the back anyway.
Finally, the paper looked at the "secondary market," where the winners of the auctions would resell their fast passes to other traders. This market turned out to be a disaster. The resellers, like a company called Kairos, ended up paying huge fees to the system but making very little profit, and their customers faced high failure rates. As the competition between the big players died down, the money the community collected from the auctions dropped significantly. In short, the paper suggests that while the idea of auctioning off fast lanes sounds fair in theory, in practice, it just concentrated power in the hands of a few wealthy players, failed to stop spam, and made the whole system less profitable for everyone involved. The "Express Lane" turned out to be less of a fair shortcut and more of a private club that didn't even work very well.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.