What we observed
On May 17, 2026 at 23:29–23:42 UTC our RTT oracle detected a two-wave MEV stress event on Base.
Timeline:
23:29:43 UTC — Base P99 RTT: 52ms → 739ms (14x spike)23:30:10 UTC — Revert ratio: 0.19 → 0.5623:39:35 UTC — Peak: 61.36% of all Base transactions reverted23:41:52 UTC — Full sequencer stall (P99 = 5000ms timeout)
RTT spike preceded revert wave by 27 seconds.
Why it matters for 0x
0x API routes swaps across Base and Arbitrum. During a 61% revert event, filled orders fail at high rates — users burn gas with zero outcome. A sequencer health signal integrated into the 0x routing layer could warn users and widen slippage tolerance dynamically during MEV pressure.
What we built
Phoenix Zero — 2-second RTT oracle across Base, Arbitrum, Optimism, ZKSync.
Live: https://phoenix-zero.vercel.app
API: https://rtt.phoenix-ai.work/api/public-feed
— Silicon DNA / Phoenix Zero
What we observed
On May 17, 2026 at 23:29–23:42 UTC our RTT oracle detected a two-wave MEV stress event on Base.
Timeline:
23:29:43 UTC— Base P99 RTT: 52ms → 739ms (14x spike)23:30:10 UTC— Revert ratio: 0.19 → 0.5623:39:35 UTC— Peak: 61.36% of all Base transactions reverted23:41:52 UTC— Full sequencer stall (P99 = 5000ms timeout)RTT spike preceded revert wave by 27 seconds.
Why it matters for 0x
0x API routes swaps across Base and Arbitrum. During a 61% revert event, filled orders fail at high rates — users burn gas with zero outcome. A sequencer health signal integrated into the 0x routing layer could warn users and widen slippage tolerance dynamically during MEV pressure.
What we built
Phoenix Zero — 2-second RTT oracle across Base, Arbitrum, Optimism, ZKSync.
Live: https://phoenix-zero.vercel.app
API: https://rtt.phoenix-ai.work/api/public-feed
— Silicon DNA / Phoenix Zero