CryptoQuant

What if we could make the Ethereum transaction journey and block construction process faster, cheaper, more flexible, censorship resistant, & robustness? What if we could do this today? This is what the Blockspace Forum is about. 🧵👇

Ethereum produces the highest-grade blockspace on the planet today. PBS has been criticised a lot, but it’s actually done a great job alleviating centralisation pressure on the core protocol. The job isn’t done yet though, let’s keep pushing the status quo forward 🫡 Post Below 👇

@kakia1989 @AliTslm @EntropyAdvisors @totorovirus volatility clustering would explain this?

@AliTslm @EntropyAdvisors @totorovirus One possible explanation for this is the fact that the correlation between previous minute markouts and current minute bids is significant, suggesting that the previous minute markouts is used to predict the next minute value when bidding.

In a joint paper with Christoph Schlegel (Flashbots), @AliTslm (@EntropyAdvisors), Ko Sunghun and @totorovirus (Matroos), we look into the performance of TimeBoost fast lane bidders, in particular, the correlation between the bids and profit estimations - 5s markouts. Short 🧵:

1,000 days. 10,000 commits. 2,000,000 blocks won. 29,000,000,000 bundles received.

@StackDigest One of the reasons, yes. Kairos can only guarantee your bundle succeeds on its current state but that is never the “true” state. Monopoly of slots is down to the fact the majority of MEV on Arbitrum is CEX/DEX and there are only two major players

@cryptoquantHQ Ooh gotcha, so I think I was completely misunderstanding this, thanks for pointing it out! So you mean the revert rate of Kairos comes from the state changed by the OFA bundles, and the monopoly of Timeboost slots comes from the advantage of low fees for large trading firms?

A few weeks ago, I was looking at the revert rate of L2s at @Dune And found that Arbitrum has the highest tx revert rate among major L2s. I thought Timeboost was working fine, so it was weird to me. Why? A thread about MEV in fast finality rollups 🧵

There seems to be a misunderstanding here. Kairos has always had sub auctions and has never accepted “fixed fee” bundles. There was a bug where bundles in follow up sub auctions could revert as local state was not progressed correctly but since fixed. Private OFA deals cause more reverts. A wallet can send its user flow to an OFA platform and the sequencer at the same time and an OFA searcher can submit a backrun 100ms later where it often ends up in the next block. Searchers can see the user tx in the feed for block N and submit a backrun to Kairos which is simulated on block N state so also valid. However, you could never beat the OFA searchers to the backrun. We actually have more searchers using Kairos now than ever before they definitely didn’t leave due to reverts. We had to scale down auctions won as atomic arb fees < CEX/DEX fees so hard to compete with Wintermute/Selini 😅

(2) Secondary Market Failure Centralization could have been mitigated through a secondary resale market, similar to Ethereum’s MEV-Boost model (or execution ticket mechanism). Titan Builder’s Kairos attempted to build exactly this - a secondary auction platform for reselling TimeBoost slots. However, the attempt failed: Kairos now accounts for only about 0.6 % of all TimeBoost transactions. According to recent paper (arxiv.org/pdf/2509.22143), the reason lies in a product-level design failure. Kairos obtained the slot first, then allowed searchers to submit bundles to it - but instead of running a sub-auction, it simply accepted bundles for a fixed fee and submitted them in bulk. Because these bundles often had dependency conflicts, many reverted. As a result, Kairos’ revert rate soared to nearly 50 %, and searchers quickly abandoned the platform once they realized that paying didn’t guarantee execution success. In short, Kairos failed because it did not re-auction its resold slots, which led to inefficiency and high failure rates.

If anyone wants to do some sleuthing, here are the addresses: 0x2E8DD7074e8fC80A47F0bBfAb3c3b58D54796430 0xd89c31cebC4c153e4942dBBE939d25f1944bF20c 0xE5b25ac4f3eECEeB651fDC90ADC39dc7804e2236 0xE45Dab1963db3604F3A4D1c5b05CA62b45928bD0 0x80689dDDcA621B4aD8Efd3ab6Ad306e050f3dC91 0x3c3c72283527f02c1d2A3BDeBe684260408f50bF 0xff31F52C4363b1daCb25D9DE07dfF862bf1d0e1C 0x35118b21744B188b88075710b31e67BC638C6b4F

As of yesterday, these wallets have burned through the ~60 ETH they were funded with and are no longer active. Now we wait to see if anyone tops them up...

@a1pha_slip They aren't actually trading anything, just making zero ETH transfers to the router so struggle to see which metrics this would inflate

@cryptoquantHQ hyping tx volume/interest for maestro shiX/meme coins?

Someone is burning $100k a month, making zero ETH transfers to the Maestro router, but why? Whilst trawling etherscan, I noticed an unusual amount of high-priority fee empty transfers to Maestro: Router 2. Opening the Maestro router showed these were extremely frequent, often occurring more than actual swaps and almost always paying ~35 Gwei in priority fees.

Using delivered payload and builder bid data scraped from all relays, we can simulate the impact of these suspicious transactions on BuilderNet's market share. Simulations show that without them, BuilderNet’s winrate would have been ~33% over the period vs the observed ~38% winrate.

What’s the point of this? The answer could lie in the block inclusion. If we examine the builder distribution, we can see the suspicious transactions land almost exclusively with BuilderNet, with a small portion also going to beaverbuild. Compare this to regular Maestro flow, which lands across all builders.

@nmb_cryptoenth @zea2D_ coinbase.transfer makes up a significant portion of total searcher bribes. Your charts would look very different if you included them. See libmev dashboard for builder profits since merge libmev.com/builders?prese…

@zea2D_ You're right, I didn’t count it but will add it. You can check these payments at payload.de/data/. They’re usually small and don’t occur in every block. The losses occur when builders pay very large bribes to validators, as I’ve shown in the dashboard’s dispersion charts.

$ETH block production is oligopolized. Only two entities decide where to place your transaction in the block. Research about MEV-boost and Builders in the thread🧵:

@donnoh_eth @ralexstokes @fradamt @bertcmiller We recently made significant improvements to our blob handling pipeline and are now above the baseline on the 7d stats. As mentioned by others, we are also closely working with various relay teams on block merging and Optimistic V3, so expect to see these numbers increase 🫡

how can we properly scale blobs if top builders are censoring, artificially pushing down max throughput and pricing? cc @ralexstokes @fradamt h/t @bertcmiller for the dashboard

Interestingly, this isn't even the largest "fat finger" trade we have refunded. We once had a searcher accidentally bribe 80 ETH due to an error in their pricing feed

I’m seeing most atomic arbs captured outside of Timeboost even when we control the express lane. 1% bribes available on arbs to anyone that can integrate quickly. Probably won’t stay this way much longer and DMs are open if any Arbitrum searchers teams want some more info

@ElvianElvy_NFT @BeatzXBT @Dub0x3A Even for analysing past data it’s massively advantageous to know when you actually received the messages, when they came late, when connections dropped etc

@BeatzXBT @Dub0x3A CCXT is definitely too slow for anything remoted to HFT but for what he is doing, at least from what it seems its past data not real-time. If its real-time definitely a different api.

Been building out market data collection infra for the MM transition (pricing and models) Realized that juicing out network alpha was essential in accurate pricing at the frequency I want to compete in, finally finished it up and it's lookin quite nice (even in high vol!)

We are thrilled to announce the Ethereum mainnet launch of the Titan Relay! A Rust-based MEV-Boost relay designed for high performance, global distribution and robustness. Check out our full announcement here: titanbuilder.substack.com/p/titan-relay-…