Competition on Robinhood Chain does not begin only when a transaction is submitted. It begins when a system first receives the latest onchain state. For searchers, trading bots, and automated trading systems, receiving a block earlier means updating local state, identifying opportunities, and constructing transactions sooner.
BlockRazor deployed benchmark clients in three AWS US East (Ohio) Availability Zones and connected each client to both BlockRazor Robinhood Chain Sequencer Feed and the official Robinhood Chain Sequencer Feed. For matching blocks, the test compared which feed reached the same client first. BlockRazor achieved first-arrival rates of 90.51%, 69.16%, and 75.08% across the three zones. Its P99 relative latency remained between 3.812 and 5.589 ms, compared with 31.128 to 42.707 ms for the official feed.
The results show that BlockRazor delivered most blocks first while also providing more stable block delivery under tail-latency conditions.
What Is Robinhood Chain Sequencer Feed?
Robinhood Chain Sequencer Feed is a real-time block data stream pushed by the sequencer. Nitro nodes subscribe to the feed through WebSocket, receive new blocks, and continuously update their local view of the chain.
Unlike an RPC client that actively queries historical or latest data, Sequencer Feed pushes new blocks directly to subscribed nodes. It is an important data input for tracking the latest Robinhood Chain state. Feed speed and stability affect how quickly a node catches up with blocks and updates its state. Network jitter, connection interruptions, or high tail latency can temporarily leave a node behind the latest height, causing systems that depend on the node to make decisions from older state.
The difference may be negligible for ordinary data queries, but it is important to searchers, trading bots, and automated trading systems: receive the block earlier → update node state earlier → identify onchain opportunities earlier → simulate and construct transactions earlier → preserve more time for transaction submission.
A low-latency feed cannot guarantee transaction success or strategy profitability. However, when strategy, fee, and execution conditions are comparable, earlier state gives a trading system a larger reaction window, improves its competitiveness in latency races, and increases its opportunity to capture potential revenue.
Benchmark Methodology
Test regions and subjects
We deployed a benchmark client in each of three AWS US East (Ohio) Availability Zones:
use2-az1use2-az2use2-az3
Each client established two WSS connections and received data from:
- BlockRazor Robinhood Chain Sequencer Feed
- Robinhood Chain official Sequencer Feed
Receiving both feeds on the same server gave the compared feeds the same client environment and local network conditions, reducing the effect of deployment differences. Testing three Availability Zones also showed whether the direction of the result remained consistent across different network paths within the same region.
Test procedure
- Connect to BlockRazor Feed and the official feed simultaneously from the same test server in each Availability Zone.
- Receive and parse the block data pushed by both feeds.
- Match identical blocks received from the two feeds.
- Record the arrival timestamp of each matching block from each feed.
- Assign a relative latency of
0 msto the feed that arrives first. - Assign the later feed a relative latency equal to the difference between the two arrival timestamps.
- Calculate first-arrival rates and P50, P90, P95, and P99 relative latency for each feed.
This benchmark measures the relative arrival time for identical blocks delivered by two feeds to the same client. A value of 0 ms therefore means that a feed arrived first for the corresponding sample. It does not mean that the end-to-end latency from the sequencer to the client was zero.
Benchmark Results
Higher first-arrival rate in all three Availability Zones
| AWS AZ | Sequencer Feed | First-arrival rate | P50 | P90 | P95 | P99 |
|---|---|---|---|---|---|---|
| use2-az1 | BlockRazor | 90.51% | 0.000 ms | 0.000 ms | 1.095 ms | 3.812 ms |
| use2-az1 | Robinhood official feed | 9.49% | 6.573 ms | 15.456 ms | 19.721 ms | 42.707 ms |
| use2-az2 | BlockRazor | 69.16% | 0.000 ms | 2.151 ms | 3.230 ms | 5.589 ms |
| use2-az2 | Robinhood official feed | 30.84% | 1.719 ms | 6.754 ms | 9.782 ms | 31.128 ms |
| use2-az3 | BlockRazor | 75.08% | 0.000 ms | 1.764 ms | 2.861 ms | 5.330 ms |
| use2-az3 | Robinhood official feed | 24.92% | 2.524 ms | 12.306 ms | 16.728 ms | 37.805 ms |
BlockRazor achieved a higher first-arrival rate in all three Availability Zones. The advantage was most pronounced in use2-az1, where BlockRazor arrived first for 90.51% of matching blocks. Even in the closest zone, use2-az2, BlockRazor arrived first for 69.16% of matching blocks. Its first-arrival rate in use2-az3 was 75.08%. The consistent direction across all three zones shows that the advantage was not confined to a single AZ or network path.
Earlier state updates and a more stable reaction window
BlockRazor recorded P50 relative latency of 0 ms in all three Availability Zones. In other words, BlockRazor was the first feed to reach the client for at least half of matching samples in every zone. Tail latency shows a larger distinction: BlockRazor P99 remained between 3.812 and 5.589 ms, while the official feed reached 31.128 to 42.707 ms. BlockRazor therefore maintained more stable block delivery when latency fluctuations were more pronounced.
For node operators, lower tail latency can reduce temporary state lag caused by feed jitter and help local state follow the sequencer more closely. For searchers, trading bots, and automated trading systems, earlier state enables earlier detection, opportunity evaluation, simulation, and transaction construction, preserving more time for submission.
Transaction success still depends on strategy validity, submission speed, fee configuration, competitors, and onchain execution. When other conditions are comparable, however, an earlier and more stable Sequencer Feed shortens the total reaction time from state update to transaction submission, improves the probability of winning competitive execution, and substantially increases the opportunity to capture potential revenue.
Getting Started
BlockRazor currently provides Ohio and Tokyo feed endpoints, allowing nodes to connect through an entry point closer to their deployment and avoid unnecessary cross-region transmission. In addition, BlockRazor Sequencer Feed is compatible with the standard feed interface used by Robinhood Nitro nodes. Operators can connect over standard WSS by updating --node.feed.input.url in the startup configuration.
If your team operates a Robinhood Chain node, searcher, trading bot, or other real-time trading system, integrate BlockRazor Sequencer Feed and benchmark it from your own production region. If you encounter any issues during integration, please contact us.