We publish our accuracy

How we measure this

• Before each session opens, FXBias publishes a directional bias (bullish or bearish) for each pair.
• At session close we compare the predicted direction against the realized move from session open to close.
• A call is correct when the market made a directional move our way, wrong when it moved the opposite way.
• Flat sessions — where price barely moved either way — are reported separately and excluded from the hit rate. A flat session is undecided, not a directional error: a position would have broken even, not lost. We show the flat count so nothing is hidden.
• Stand-aside / neutral calls are not graded — when there's no clear edge, we say so rather than forcing a guess.
• Hit rate = correct ÷ (correct + wrong). Pairs with fewer than 5 graded sessions are hidden to avoid small-sample noise.

No service is right every time. A consistent directional edge above 50%, applied with disciplined risk management, is what compounds. FXBias is an analytical tool for educational purposes — not financial advice, and not a guarantee of future results.