Flashbots · Measurement Guide
How to measure Flashbots bundle execution
Use this reproducible workflow to measure simulation success, relay acceptance, latency, fees, and final on-chain inclusion in your own environment.
Linkable summary
Why link to this page
This page defines the evidence required for a valid Flashbots comparison. It does not publish a universal FRB inclusion or performance result.
The supporting evidence and methodology are documented in the Ecosystem research brief when you reference these KPIs.
For cross-relay comparisons, line up these numbers against the full FRB MEV benchmarks dataset and revisit the Flashbots tutorial when your desk needs step-by-step execution guidance.
Want the raw block builder stats? Open the Flashbots metrics dashboard to inspect relay variance, MEV-Boost relay health, and transaction ordering trends before you green-light new capital.
Flashbots follow-up resources
Make it effortless for readers to keep navigating internal FRB content once they digest the headline numbers.
Execution & learning
- Flashbots tutorial for step-by-step bundle submissions.
- Ethereum MEV guide when teams compare chains.
- Execution hub for escalation macros and governance notes.
Telemetry & support
- Cross-chain metrics hub for broader context.
- MEV gas calculator when budgeting for Flashbots lanes.
- Support & SLA if a relay escalation needs human help.
How to read these Flashbots benchmarks
Measure simulation, relay acceptance, and on-chain inclusion separately. Report p50/p95 latency and total fees with the endpoint, region, builder, sample size, and test window.
- Match the inclusion row to your Ops Pulse widget—variance greater than 8 pts means dive into logs.
- Track latency per POP to decide when to rotate block builders or MEV-Boost relays.
- Use gas/tx to explain why private bundles often outperform public PGAs on transaction fee revenue.
What these numbers mean for MEV trading
Private relay submission can reduce public mempool exposure, but it does not guarantee acceptance, ordering, inclusion, protection, or profitability. Define thresholds from your measured baseline.
Scenario deltas also highlight why structured guardrails beat ad-hoc scripts: budget caps plus relay rotation ensure maximal extractable value does not come at the cost of runaway risk.
Limitations & common pitfalls
- These KPIs assume at least two MEV-Boost relays configured; single-relay setups can skew variance.
- Latency readings exclude public RPC overhead—run the WSS latency test to confirm your region.
- Scenario rows use curated templates; update the dataset whenever you change contract logic or block builder lists.
Always rerun the dataset after new relay releases so long-term performance trends remain trustworthy.
Flashbots MEV-Boost & The Role of Relays
MEV-Boost is an out-of-protocol sidecar that lets validators source blocks from builders through relays. Measure each active route because builder coverage and network conditions change.
Relay Aggregation
Where supported and configured, multiple relay paths can provide fallback options. Record the actual relay list and outcome for every run rather than assuming a performance benefit.
Bundle Merging
Modern builders perform "Bundle Merging," where they combine non-conflicting searcher bundles into a single block. FRB's simulation engine predicts how your bundle will interact with the "Top of Block" (ToB) flow, ensuring your transaction ordering stays valid even in a crowded block.
For a deep dive into relay selection, see our Best RPC Nodes Guide or the Live Relay Status Tool.
Median Inclusion
Record sample size and final inclusion
Avg Latency
Record region and endpoint
Gas per Tx
Private bundles avoid PGA blowups
Scenario Comparison
Backrun (ETH/USDT, 2 tx)
Record retries, relay changes, target blocks, and final inclusion.
Sandwich-risk reduction (stable pair)
Record public-mempool exposure, reverts, and builder outcome.
Protection bundle (wallet)
Record the selected private path, fallback behavior, and confirmation.
Private auction (high congestion)
Record fees, congestion, rejection reasons, and net outcome.
How the data is produced
Run identical bundle templates in a pinned environment. Tag each attempt with relay, builder, region, latency, fee settings, simulation result, target block, and final on-chain status.
- Latency numbers include p50/p95 per region so you can pick the right POP.
- Scenario deltas are relative to a generic bot replay with the same inputs.
- Assign a run ID and retain timestamped raw logs so reviewers can reproduce the analysis.
Need to replicate? Follow the Knowledge Base article "Telemetry harness replay" and compare your hashes.
Run a local measurement
Start with the WSS latency tool, then attach your own relay and on-chain logs.
Open latency testSee it live
Open `/app/dashboard`, pair one node client, and watch the Flashbots gauges update in real time.
Launch dashboardHow to apply the Flashbots metrics
- Establish a baseline from your own sample before defining an inclusion alert.
- Map measured p95 latency to your WSS rotation policy and document the selected threshold.
- Share only comparisons backed by identical inputs, raw evidence, and stated limitations.
Briefing checklist
When presenting Flashbots performance, include the measured baseline, current result, variance, environment, sample size, limitations, and action taken.
- Include screenshots of Ops Pulse gauges with timestamps.
- List any relay escalations or support tickets you opened.
- State whether numbers came from simulation or live capital.
How to use these Flashbots MEV benchmarks
When you submit a change request or investment memo, embed this Flashbots benchmark snapshot. Stakeholders want to see median inclusion, latency, and the delta between FRB and generic bots.
- Quote only the inclusion baseline measured for the environment and date in your report.
- Attach scenario rows that match your strategy (backrun, protection, private auction).
- Attach raw relay and on-chain logs plus screenshots so reviewers can recreate the analysis.
Point readers toward the next click
Help teams stay inside ai-frb.com after they study the Flashbots metrics by giving them an explicit set of follow-up links.
- Install guide for desks that want to reproduce these Flashbots runs on new machines.
- Pricing overview so procurement can review the free performance fee model.
- Docs quick start to standardize verification, SHA-256 checks, and node pairing.
- Support & SLA when relays or block builders need escalation.
FAQ
What inclusion rates can I expect with FRB + Flashbots?
There is no universal rate. Measure simulation success, relay acceptance, and final on-chain inclusion separately for your builder, fee policy, region, and test window.
How do you measure these KPIs?
Replay identical bundles, pin the environment, and record timestamped relay responses plus final transaction status. Keep the raw logs with the report.
Can I plug these numbers into my client pitch?
Only use results measured in your own environment and attach the method, sample size, date, raw evidence, and limitations.
Put these Flashbots MEV benchmarks to work: compare tooling on the MEV Bot Comparison, brief teammates via the Flashbots tutorial, and install FRB wherever you need consistent telemetry.
Close the loop by downloading the installer, sharing the MEV 101 hub or MEV strategies guide with stakeholders, and routing budgets toward Ethereum MEV or other chain landers once the Flashbots data gives you confidence.