Every Forecast. Every Outcome. No Cherry-Picking.
Our full prediction history is recorded before markets resolve and verified against outcomes. This page shows every forecast — including the ones we got wrong.
Balanced Mode
High-confidence forecasts with strong accuracy. This is the default mode powering the platform's advertised accuracy.
Calibration Curve
A well-calibrated forecaster's predictions should match actual outcomes. When we say 70%, the event should happen ~70% of the time.
Our nightly recalibration process adjusts per-category multipliers to minimize overconfidence and underconfidence.
Balanced Mode Resolved Forecasts
How We Measure Accuracy
CRPS: Beyond Win/Loss
Unlike simple win/loss tracking, Continuous Ranked Probability Score (CRPS) measures how close our probability estimates are to the actual outcome. A forecast of 90% on an event that happens scores better than a forecast of 55% on the same event — even though both are technically "correct." This incentivizes precision, not just being on the right side.
Nightly Recalibration
Every night, our system reviews all resolved forecasts and computes per-category calibration multipliers. If we're systematically overconfident in crypto forecasts (e.g., predicting 80% when the true rate is 75%), the multiplier corrects for this bias going forward. This process runs automatically and the results are reflected in the calibration chart above.
Model Disagreement Scoring
When all 6 models agree (6/6), our confidence is highest. When models disagree significantly (3/6 or 4/6), we flag lower confidence and adjust our calibrated output downward. Disagreement is itself a valuable signal — it often indicates genuinely uncertain markets where the outcome is harder to predict.
See something interesting? Start researching markets yourself.
⚠️ Disclaimer: PolyEdge AI provides AI-generated probability forecasts and market analysis for informational and educational purposes only. This is not financial advice. We do not facilitate trades or handle user funds. Past forecast accuracy does not guarantee future results.