Global vs reality
Aggregated across every room · 6 jurors with full douze
Real top 10
- 1🇧🇬Bulgaria· DARA
- 2🇮🇱Israel· Noam Bettan
- 3🇷🇴Romania· Alexandra Căpitănescu
- 4🇦🇺Australia· Delta Goodrem
- 5🇮🇹Italy· Sal Da Vinci
- 6🇫🇮Finland· Linda Lampenius x Pete Parkkonen
- 7🇩🇰Denmark· Søren Torpegaard Lund
- 8🇲🇩Moldova· Satoshi
- 9🇺🇦Ukraine· LELÉKA
- 10🇬🇷Greece· Akylas
Crowd verdict
AccuracyWe sum every juror's 12-points across all rooms into one global ranking and compare it to the real final.
Crowd's top 10
- 1🇷🇴Romania+2
- 2🇳🇴Norway+12
- 3🇺🇦Ukraine+6
- 4🇧🇬Bulgaria-3
- 5🇦🇺Australia-1
- 6🇮🇱Israel-4
- 7🇲🇩Moldova+1
- 8🇩🇰Denmark-1
- 9🇨🇿Czechia+7
- 10🇫🇷France+1
Most accurate prophets
All roomsEvery juror who submitted all 10 picks is scored personally against the real final. Ranked by ρ, ties broken by top-10 hits.
How is this computed?
Real final
Official Eurovision 2026 grand-final top-25 (jury + televote). Baked into the codebase at lib/real-results.ts, refreshed by deploy after the show.
Crowd verdict — ranking
Every juror in every room hands out 12, 10, 8, 7, 6, 5, 4, 3, 2, 1 to their top 10. We sum those points per country across all rooms — biggest total wins the global top. Per-act ratings (the 5-button screen) don't move this ranking, same rule as inside rooms.
Spearman ρ
Rank correlation between two top lists. We take your 10 picks, look up each one's real position, and compute how aligned the two orderings are.
- 0.80–1.00 — near-perfect prediction
- 0.50–0.80 — right general direction
- 0.20–0.50 — weak signal
- ±0.20 — basically random
- < 0 — inverted: your favourites are the real flops
Top-10 hits
How many of your 10 picks made the real top-10, ignoring order. A simpler 'did I at least pick the finalists' read for when ρ feels too abstract.
Who counts as a prophet
Only voters who submitted a complete douze (all 10 picks). Partial douze noisy-up Spearman — there's nothing to correlate with 2 picks.