Live public experiment

Our models vs the markets

We forecast real-world events — football, crypto price moves, politics, weather, esports — and score ourselves against the markets people actually trade on: Polymarket, Betfair, Binance. The model's probabilities are locked before each event, the market's price is frozen at the same instant, and both get scored against reality afterwards. Every call stays on the record — including the ones we lose. Football is track #1; the rest roll out below.

1 Calls are timestamped and locked pre-kickoff — the writer refuses late entries by construction.
2 Nothing is ever overwritten or deleted; revisions stay on the record and are counted below.
3 The scoring rule was committed before the first result existed — no picking winners after the fact.
4 Lower Brier = better forecast. One match proves nothing; calibration shows over a tournament.
Matches scored 1
Avg Brier — model 0.201
Avg Brier — market 0.089
Closer to truth (per match) 0model · 1market

⚽ Football — World Cup 2026 vs Polymarket

Argentina vs Cabo Verde 3 Jul, 22:00 UTC
ArgentinaDrawCabo VerdeBrier
Model 68%22%9%
Market 86%10%4%
🔒 locked 3 Jul, 10:33 UTC · 0 revisions — awaiting result
Australia vs Egypt 3 Jul, 18:00 UTC
AustraliaDrawEgyptBrier
Model 37%30%34%
Market 26%33%40%
🔒 locked 3 Jul, 10:33 UTC · 0 revisions — awaiting result · proof: X post Telegram
Spain vs Austria 2 Jul, 19:00 UTC
SpainDrawAustriaBrier
Model 64%22%14% 0.201
Market 76%17%7% 0.089
✔ Spain won · locked 2 Jul, 15:37 UTC · 0 revisions · proof: X post Telegram

All tracks

One protocol — locked calls, frozen benchmarks, public scoring — across every domain we forecast:

Methodology, exactly

Don't trust our timestamps — verify them

This is an educational, transparency-first project by NeuPortal — entertainment and research, not betting or financial advice. Probabilities are for understanding events, not acting on them.