Montran World Cup 2026 · who wins the pool

Who wins
the pool

Where the table lands once the final is over — the most total points wins. With 104 of all 104 matches scored and 0 to go (group + knockout), each remaining match is bootstrapped from a player's own scoring history (20,000 simulated tournaments) for a projected final total, an 80% range, and the odds of winning. Totals include every bonus (group finishing position, Round-of-32 → champion, top scorer), each a simulated outcome so players who backed the same teams move together; players whose bracket isn't modelled keep a 0-future-bonus floor, so their projection is a lower bound.

Backtest — forecasting matches 15–104 from 1–14: mean error ±141.2 pts per player. Raw bootstrap bands covered only 58% of outcomes, so intervals are widened ×2 to a calibrated 81% coverage of the 80% range (target ≈ 80%).
Projected to win
100% title odds · proj. 988 pts
Most uncertain run-in
80% range 988–988 (0 pts wide)

Title race · odds to finish #1

100%Radu V

How the title odds moved

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% 105% 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 96 100 104 matches scored marcus23 — now 0% mbogdan — now 0% Raisa — now 0% Gullit26 — now 0% ri2c — now 0% The Hound — now 0% Rocky Gratiano — now 0% Pikachu — now 0% Arhentinaa — now 0% Cristi — now 0% Tay Tay — now 0% Labuta — now 0% gnita — now 0% Inoasca — now 0% HoratiuT — now 0% loaded fries — now 0% Alexandrai — now 0% vladivostok — now 0% Cri — now 0% Tricolorii — now 0% Radu V — now 100% Stochastic Parrot — now 0% Dali002 — now 0% raluca — now 0% raresFarcas — now 0% Montran — now 0%

Everyone opens at 1/26 ≈ 4% before a ball is kicked, then the field fans out as results land. Odds of finishing #1 are recomputed after each scored match by replaying the same joint simulation on the standings at that point, with each player's full simulated tournament bonus added on top (0 for unmodelled brackets). The first 5 matches are too thin for the bootstrap to give any spread, so each line runs from the common origin into the series from match 6 on. Top six by current title odds are highlighted — hover any line for the name.

Projected final standings

ProjPlayerProjected final (80% range) FinalWinTop 3
#1Radu Vnow #1 · 988
988988–988100%100%
#2Stochastic Parrotnow #2 · 980
980980–9800%100%
#3Dali002now #3 · 929
929929–9290%100%
#4ralucanow #4 · 902
902902–9020%0%
#5raresFarcasnow #5 · 901
901901–9010%0%
#6Montrannow #6 · 897
897897–8970%0%
#7marcus23now #7 · 897
897897–8970%0%
#8mbogdannow #8 · 895
895895–8950%0%
#9Raisanow #9 · 874
874874–8740%0%
#10Gullit26now #10 · 872
872872–8720%0%
#11ri2cnow #11 · 855
855855–8550%0%
#12The Houndnow #12 · 855
855855–8550%0%
#13Rocky Gratianonow #13 · 853
853853–8530%0%
#14Pikachunow #14 · 840
840840–8400%0%
#15Arhentinaanow #15 · 825
825825–8250%0%
#16Cristinow #16 · 812
812812–8120%0%
#17Tay Taynow #17 · 803
803803–8030%0%
#18Labutanow #18 · 794
794794–7940%0%
#19gnitanow #19 · 779
779779–7790%0%
#20Inoascanow #20 · 745
745745–7450%0%
#21HoratiuTnow #21 · 731
731731–7310%0%
#22loaded friesnow #22 · 720
720720–7200%0%
#23Alexandrainow #23 · 717
717717–7170%0%
#24vladivostoknow #24 · 624
624624–6240%0%
#25Crinow #25 · 584
584584–5840%0%
#26Tricoloriinow #26 · 177
177177–1770%0%
█ bar = 80% range| bright tick = projection | faint tick = current points

Each unplayed match — group and knockout, through the final — is a random draw from that player's own per-match points so far (skipped games count as 0, so low engagement carries forward). All 20,000 tournaments are simulated jointly, tie-broken by the pool tiebreaker, to read off win / top-3 odds. The full bonus (group finishing position, Round-of-32 → champion, top scorer) is drawn from one shared simulated tournament per trial, so co-pickers rise and fall together. Caveats: it assumes scoring rate and engagement hold and that knockout games keep the same MOTD/scorer mix (the model can't see fixtures it doesn't yet know); the bonus treats each team's fate as independent (no two-teams-can't-both-win joint); players whose bracket isn't modelled are credited 0 future bonus, which understates them; and thin samples get wide bands. Treat it as the shape of the race, not a prophecy.