Two prices for the same ounce of gold.
Public 21-day trial. Every 5 seconds we record XAU/USD, XAU/AUD, and AUD/USD from OANDA. We track each market's move from the UTC session open and test whether AUD gold's move matches USD gold after FX conversion. If the gap persists, that may indicate a tradable dislocation; if it closes quickly, the thesis fails.
Open research trial · outcomes unknown · not financial advice · read the fine print
You're arriving on day one. Every number below will either move or refuse to move. Either way, it gets logged. You're paying to watch it unfold, not to be sold a conclusion.
Watch the gap move right now.
Synthetic replay of the session-open divergence engine: USD gold moves, AUD gold lags, and the dollar gap widens until it reverts. Same math as the live feed.
Waiting for a signal to analyze…
No trades yet. Press Start.
USD gold moves. AUD gold sometimes lags.
At 00:00 UTC each day we pin the first XAU/USD and XAU/AUD prints from OANDA. Every 5 seconds after that we measure how far each leg has moved from its open, convert the AUD move to USD at the live AUD/USD rate, and log the gap. When the gap holds above 0.3% for 90 seconds, a signal fires.
The data
Live OANDA platform ticks for XAU/USD, XAU/AUD, and AUD/USD every 5 seconds. The first tick of each UTC day is stored as the session open for all three. Every subsequent tick is written to MySQL with the USD move, AUD move in USD, gap in dollars, and gap as a percent of the session open.
The question
USD gold and AUD gold are the same ounce of metal priced in two currencies. Once FX is factored in, their moves should match. When they don't — when AUD gold lags USD gold by more than $10–$20 USD for more than a minute — is that a real, repeatable dislocation, or just noise? Three weeks of tick-level data will answer it better than anecdotes.
The verdict
On day 21 we publish our honest read — real edge, friction, or inconclusive. The full dataset ships alongside it. No cherry-picking, no vibes-based conclusions. You can argue with our read; we'll show our work.
Everyone says retail arbitrage is dead. Maybe it is. But nobody's actually shown their working — so I'm showing mine. A solo build: OANDA ticks for USD gold, AUD gold, and AUD/USD, a UTC session-open baseline, a MySQL capture log, a paper-trading engine. Measured in public for twenty-one days.
One-time support. Not a subscription.
Both passes are one-time payments that fund the trial. No recurring charges. If the anomaly is real, pass holders get first access to the live product.
- Complete captured dataset: every tick, 21 days, timestamped.
- Honest verdict: edge, friction, or inconclusive.
- Full backtest runner on captured data.
- First access + discount if a live product ships.
- Everything in the Results pass, plus:
- Live spread ticker to watch the gap move in real time.
- Every signal the instant it fires, with context.
- Private build log: what I tried, broke, and learned.
- Priority invite + larger discount if a live product ships.
Before you pay for anything, read this.
What am I actually paying for?
A seat in an open experiment. Every tick, every signal, every surprise, plus the final dataset and our honest read. If the gap turns out to be trade-able at retail size, existing pass holders get first dibs on whatever ships next — at a discount.
Can I get a refund?
No. You're paying to help fund an open trial that runs for exactly 21 days. If you're not comfortable with that, please don't buy a pass. The simulator is free and will always be.
Is this investment advice?
Absolutely not. Nothing on this site is financial advice. Always verify the data yourself before acting on anything. We're not responsible for trades you place.
What if the anomaly turns out to be real?
We'll write it up openly, then launch a separate paid monthly product with live trade data. Existing pass holders get first dibs and a discount. If it turns out to be nothing, we'll say that too.
How does login work?
You enter your email and we send a one-time sign-in link. No passwords. Pass holders unlock the Live Data and Backtest pages after signing in. Results-pass holders get access automatically on the trial end date.
What's the actual data pipeline?
No fund, no promotion, no pitch deck — just a real capture stack running 24/7:
- OANDA feedXAU/USD · XAU/AUD · AUD/USD every 5s
- UTC 00:00 snapshotfirst tick pinned as session open
- Gap engineUSD move − AUD move in USD, every tick
- MySQL capturetimestamped, reproducible, downloadable
Nothing on this site is financial, investment, tax, or trading advice. The data, simulations, and signals are provided for research and educational purposes only. Past performance — simulated or otherwise — is not indicative of future results. Markets can and do move against you faster than you can react.
You are solely responsible for any trading decision you make. Always verify data directly with primary sources. We make no representations and offer no guarantees regarding the accuracy, completeness, or usefulness of any information presented here.
Trial passes are one-time payments to support an open research trial. They are non-refundable under any circumstances. By purchasing a pass you acknowledge that the trial outcome is unknown, results may show no anomaly at all, and you accept full responsibility for any actions you take based on anything you read here.