Spain arrive as the cleanest favorites in Group H: 92.9% to qualify, 61.3% to top the group, and a +1.54 Δ xG that places them among the top five differentials in the entire World Cup. Uruguay is the second-most likely runner-up of any non-Big Five group (83.9% qualification), thanks to a kind draw: Saudi Arabia and Cape Verde split barely 23% between them. The only match with real uncertainty is Spain–Uruguay in Guadalajara (MD3), which decides first place and which Round-of-16 fixture each side has to navigate.
The Ensemble v1 (DIBP + Dual SPI + Elo + Bayesian Poisson, calibrated with temperature T=0.85 over 364 validation matches, post-calibration ECE 3.3%) produces probability predictions for each of the 103 World Cup fixtures. The table below shows the average per-match win probability for each nation within the group (not the probability of finishing 1st — that is covered in section 4):
| Nation | Per-match win prob (avg) | xG for | xG against | Δ xG |
|---|---|---|---|---|
| 🇪🇸 Spain | 66.2% | 2.34 | 0.80 | +1.54 |
| 🇺🇾 Uruguay | 54.8% | 1.91 | 0.98 | +0.93 |
| 🇸🇦 Saudi Arabia | 17.0% | 0.86 | 1.86 | -1.00 |
| 🇨🇻 Cape Verde | 11.3% | 0.72 | 2.18 | -1.46 |
The gap between Uruguay (83.9%) and Saudi Arabia (14.9%) is 69 points — the second-largest intra-group step in the tournament behind the Spain/Cape Verde gap inside Group H itself. This is a two-tier group, not a four-tier one.
92.9% qualification, 61.3% to top the group. Δ xG +1.54 (2.34 for / 0.80 against). Sharpest attack on the western half of the bracket and the only defense in the group below 1.0 xG against. xP 6.61: basically a modelled ‘two wins and a draw’.
Model strengths — Strengths: possession control and high block. 2.34 xG per match is top-5 in the World Cup. Low block defensively: 0.80 xG against is the most solid number on the western half. The Roja wins 66% of their matches in simulation.
Risks — Risks: managing the MD3 draw with Uruguay if already qualified. Spain has rotated in previous World Cups and paid for it (Morocco 2022). If rotation goes badly, finishing second becomes realistic (31.6% p2).
Realistic path — Most likely path: 1st in group (61.3%). Round of 16 against a 2nd-place finisher from Group D — candidates: Croatia or Morocco. Both playable, but Morocco would be the known executioner.
83.9% to qualify, 34.9% to win the group. Δ xG +0.93. La Celeste does not need to beat Spain to get out — they just need to avoid 0–2 type results in MD1+MD2. xP 5.65 is the fourth-highest non-world-champion xP in the tournament.
Model strengths — Strengths: World Cup experience (15th appearance) and two-time historic champion. Δ xG +0.93 is #2 in the group. Bielsa runs a mid-to-high block with suffocating pressure: the weaker sides (KSA, CV) crack quickly against that system.
Risks — Risks: the opener vs Saudi Arabia (62.9% Uruguay) is the classic banana-skin fixture. If they draw or lose there, qual drops from 84% to ~55% instantly. The 0.98 xG against is not invulnerable.
Realistic path — Most likely path: 2nd in group (49.0%). Round of 16 against the winner of Group D — England, Croatia, or the actual top dog there. Tough crossover either way.
14.9% qualification. The bulk of that 14.9% depends on Cape Verde, not on stealing points from Spain or Uruguay. Δ xG −1.00 and 1.86 expected goals against per match: the defense is the problem, not the attack. Must beat Cape Verde in MD3 to have a real shot.
Model strengths — Strengths: the Al-Hilal/Al-Nassr backbone gives them spine and counter-attack pace. Mansoor and Salem Al-Dawsari (if fit) can punish. History: they beat Argentina in Qatar 2022 — the group should not write them off without a fight.
Risks — Risks: 1.86 xG against per match. Worst defense in the group by a wide margin. Δ xG −1.00 means any top-30 opponent is a problem. Only 2.5% chance of finishing first.
Realistic path — Most likely path: 3rd in group and elimination. To qualify, they need to win MD3 vs Cape Verde and have Uruguay sitting at no more than 4 points after MD1+MD2.
World Cup debut. 8.3% to qualify, second-lowest figure in the tournament after Jordan. Δ xG −1.46 and nearly 6 expected goals against across three games. The story is to avoid humiliation against Spain and arrive alive into the MD3 dogfight with Saudi Arabia.
Model strengths — Strengths: a debut is an emotional asset. No pressure, new generation, expatriate talent from Portugal/France. History: they made the AFCON 2024 group-stage exit, not naive.
Risks — Risks: 2.18 xG against per match — worst defense in the entire tournament alongside Jordan. Δ xG −1.46 says they lose all three. Spain and Uruguay can put 3+ on them each.
Realistic path — Most likely path: 4th and home with the experience. The MD3 vs Saudi Arabia (23.4/42.7/33.9) is the only fixture in the group where they have a real chance to score.
The six group games are spread across FIFA-assigned host cities. Probabilities reflect the model's 1X2 distribution for each match, expressed as percentages.
Up to this point we have shown the average per-match win probability (how strong each team is in any given individual match). But "who finishes 1st in the group?" is a different question: it requires winning ENOUGH matches AND scoring MORE points than the other 3 by the end of matchday 3. For that we simulate the entire group 100,000 times and count how often each team finishes 1st, 2nd, 3rd or 4th.
Tournament probabilities (Spain +1.54 Δ xG, Uruguay +0.93) reflect baseline strength. The Round-of-16 draw can flip the value: finishing 2nd in H could mean England in the next round, while finishing 1st means the 2nd-placer from Group D. The Spain–Uruguay MD3 fixture doesn’t just decide first place, it decides which crossover you avoid.
Simulating the six group matches 100,000 times under the model's probability distributions, we obtain the following probabilities of finishing the group in each position. These numbers are the most relevant probabilistic input for informed Battle Royale decisions.
| Nation | P (1st) | P (2nd) | P (qualify) | P (eliminated) |
|---|---|---|---|---|
| 🇪🇸 Spain | 61% | 32% | 93% | 7% |
| 🇺🇾 Uruguay | 35% | 49% | 84% | 16% |
| 🇸🇦 Saudi Arabia | 3% | 12% | 15% | 85% |
| 🇨🇻 Cape Verde | 1% | 7% | 8% | 92% |
Spain wins the group in 61% of the 100k simulations. Uruguay top in 35%. Joint probability of both advancing: 82%. The scenario where Saudi Arabia qualifies (15%) almost always requires Uruguay to drop points on MD1 and then lose MD3.
Another powerful metric: the expected number of points each nation will accumulate across its three group fixtures, weighted by the model's probabilities. Results:
| Nation | XP | Expected goals (for) | Expected goals (against) |
|---|---|---|---|
| 🇪🇸 Spain | 6.61 | 6.21 | 2.31 |
| 🇺🇾 Uruguay | 5.65 | 4.91 | 2.67 |
| 🇸🇦 Saudi Arabia | 2.41 | 2.45 | 4.87 |
| 🇨🇻 Cape Verde | 1.80 | 2.15 | 5.87 |
xP Spain 6.61: roughly ‘almost two wins and a draw’. xP Uruguay 5.65: ‘win + win + draw-or-tight-loss’. Both numbers in the top-10 of the entire World Cup. KSA (2.41) and CV (1.80) are bottom-half on the global xP table.
Saudi Arabia steals from Uruguay in the opener They are +0.8 expected goals away from a draw. One poorly defended corner and 1–1 is on. If KSA take a point from MD1, their qual jumps from 15% to ~33%. If they win it, they pass Uruguay outright.
Cape Verde holds Spain to 0–0 Tunisia did it to France in 2022. If Cape Verde set an extreme low block and survive 90 minutes without conceding, the group reopens because Spain finishes xP 5.4 instead of 6.6. Real probability ~3-4%, but the impact on the rest of the group is enormous.
The MatchScan Ensemble v1 combines four independent predictive models weighted by their historical accuracy over the consolidated 87,000+ match database:
DIBP (37.5%) — Dynamic Inter-team Bayesian Poisson. Goal-generation model with inter-team rating adjustment.Dual SPI (25%) — Soccer Power Index with separate attack/defense adjustment.Elo v0 (37.5%) — Dynamic Elo system with opponent-strength adjustment.Bayesian Poisson (0% in final blend) — Bayesian model showing instability on asymmetric matches.
The resulting probability set is calibrated via scalar temperature T=0.85 (Platt scaling) with Bayesian correction b_H=-0.24 and b_D=+0.38. Calibration is verified over 364 cross-validation matches, yielding pre-calibration ECE of 4.68% and post-calibration ECE of 3.26% — very low expected calibration error, indicating published probabilities are reliable as frequentist expectations.
World Cup 2026 Battle Royale on MatchScan. Pick 8 teams for the tournament. One per matchday. If your team loses or draws, eliminated. Last survivor wins €100 + 1 year MatchScan VIP. Free entry, no card, no spam.
Sign up →