Live public dashboard · Updated every minute · No signup

See real forex arbitrage opportunities, live

BJF’s detection program runs continuously against a tier-one institutional fast feed and a panel of retail forex brokers (anonymized as Broker A, Broker B). Every event below is a real opportunity window. Free public access. Methodology peer-citable.

Honest disclosure

Brokers are not named. The data reflects measurement against specific retail venues but is published anonymously to focus discussion on execution-quality patterns rather than on individual broker reputation. The detection methodology is documented in an open-access peer-citable research preprint on Zenodo and in a companion working paper on the Munich Personal RePEc Archive.

Quick start by time available

Under 1 minute: KPI cards at the top, hour-by-day heatmap, latest events table.

2 to 5 minutes: Best Window analysis, daily series, broker lag profile.

Power users: Filter by symbol, broker, time range. Export CSV.

Live data

What the dashboard shows in real time

Arbitrage Statistics Analyzer

—

OPPORTUNITIES—detected events

AVG DIFFERENTIAL—mean max-divergence

AVG DURATION—execution window

MAX DIFF—single biggest event

THEORETICAL PIPS—sum of max-divergences

BEST WINDOW—

Opportunities — hour of day × weekday

Differential distribution (pips)

Duration distribution (ms)

Opportunities per day

Trading session breakdown

Top symbols by opportunity count

Slow broker lag profile — avg event duration by hour (ms)

Cumulative theoretical pips

Detected arbitrage events

#	Timestamp (UTC)	Symbol	Direction	Diff (pips)	Duration (ms)	Slow Broker	Session

—

Ready.

Metrics explained

What each KPI measures

Five primary indicators run continuously. Each measures a distinct property of the opportunity flow. Understanding what each one means is the difference between glancing at the dashboard and using it.

1,247

events / day typical

Opportunities Count

Number of detected windows in the selected time range. Cross-time persistence validates that arbitrage opportunities are not transient. High counts during specific hours signal session-overlap concentration.

Count metric

94 ms

EURUSD reference

Median Window Duration

Half of all windows close in less than this. The broker’s quote update lag. Lower medians indicate faster broker price servers; higher medians indicate slower update cycles.

Latency metric

0.38

pips average

Avg Differential

Mean signal-time differential between fast feed and broker quote. Indicates the depth of the opportunity, not just frequency. Higher averages mean larger per-trade theoretical edge.

Depth metric

07 to 10

UTC band

Best Window

The hours of day with the highest opportunity density, identified parameter-free by the Sliding Best-Window algorithm. Allocate detection resources to these hours.

Time-of-day metric

“Public disclosure mechanisms have historically improved market quality in other asset classes. TRACE made US corporate bond pricing transparent. MiFID II RTS 27 reports made EU venue execution quality observable. For retail forex, no equivalent mandate exists. This dashboard is voluntary disclosure, contributing to the same pattern.”

Fesenko (2026), Persistence of Latency Arbitrage Opportunities in Retail Forex Markets, MPRA working paper

Methodology

Why this dashboard is free and public

Retail brokers have direct observation of their own execution quality. Retail traders typically do not. Public disclosure mechanisms close that information gap. The dashboard is BJF’s voluntary contribution to a pattern that other asset classes already have at the regulatory level.

Mechanism	Asset class	Jurisdiction	Status	Update
TRACE	US corporate bonds	FINRA / SEC	Mandatory, 2002	Within 15 min
RTS 27	EU venues	ESMA / MiFID II	Mandatory, 2018	Quarterly
Rule 605	US equities	SEC / FINRA	Mandatory, 2000	Monthly
BJF Arbitrage Stats	Retail forex	Voluntary	Voluntary, 2026	Every minute

Reading the data

How to interpret the dashboard

Five operational steps. Each uses one or more KPIs to answer a specific question about the broker panel.

Check the KPI cards

Start with Events count and Median Duration. Events over 500 with median under 150 ms signals an active broker panel. Events under 100 or median over 300 ms signals quiet conditions or fast broker quotes.

Look at the hour by day heatmap

Bright cells are dense; dim cells are sparse. London open (07 to 10 UTC) and London / New York overlap (12 to 16 UTC) typically light up. Asian session (23 to 02 UTC) is usually moderate. Sundays appear dim because volume is light.

Use the Best Window

The Best Window KPI tells you the highest-density 2 to 4 hour band of the trading day. Schedule attention, software resources, and capital around this window.

Compare against the broker lag profile

The “Slow broker lag profile” chart breaks down average event duration by hour. A broker whose lag is constant has a fundamentally slower update cycle. Lag spikes during specific hours signal active risk-management policy (wider spreads, last-look hold).

Apply to your own broker selection

If your strategy runs at round-trip time T and the panel shows median window duration W, your expected retention ratio is approximately the survival probability P(W greater than T). Lower retention means broker selection matters more. Higher retention means execution latency is less of a binding constraint.

Want this same detection on your broker setup?

SharpTrader Pro is the production-grade software that powers this dashboard. The same engine, the same Best Window algorithm, the same sub-100ms detection, calibrated for your specific broker pair. Includes broker selection assistance, VPS discounts, installation support, and lifetime technical assistance.

View SharpTrader Pro →
Read the methodology paper

FAQ

Common questions about this dashboard

How often is the dashboard updated?

Every minute. The detection program runs continuously against the tier-one fast feed and the broker panel. Events are written to the database in real time and the dashboard reads from that database on page load and on filter changes.

Why are the brokers anonymized?

Two reasons. First, focusing the dashboard on execution-quality patterns is more useful than naming specific brokers, because broker behavior changes over time. Second, anonymization avoids targeting specific firms, which is the right ethical default for a public disclosure mechanism. The methodology paper documents the measurement approach so that anyone can replicate the analysis with their own broker panel.

What is a tier-one fast feed?

An aggregated institutional best-bid-best-offer composite price feed sourced from multiple tier-one liquidity providers. Updates at sub-millisecond resolution. The closest practical approximation to the true interbank price for any given currency pair. Retail broker quotes lag this feed by tens to hundreds of milliseconds.

Can I run this same detection on my own broker?

Yes. SharpTrader Pro runs the same detection engine, the same Best Window algorithm, the same KPI calculations. The difference is that on your own setup, the broker is your own broker (not anonymized) and the detection results are private to your account. Configuration support is included with every SharpTrader purchase.

Is this data peer-reviewed?

The methodology is documented in a peer-citable open-access research preprint on Zenodo and in a companion working paper on the Munich Personal RePEc Archive. Neither paper has been through journal peer review at the time of writing. The reference implementation is open-source for independent verification.

Does this dashboard predict future opportunities?

No. The dashboard shows historical and current opportunity data. It does not predict future windows. The Best Window output identifies time-of-day patterns that have been persistent in the past and are likely to continue, but it is a descriptive statistic, not a forecast.