Mexico vs. South Korea - Post-Event Review

2026 World Cup, Group A, Match 25 | June 18/19, 2026 | Estadio Guadalajara (Akron), Mexico

Predicted vs. Actual | Calibration: ~78%

TL;DR

Result: Mexico 1-0 South Korea (HT 0-0; Romo 50'). Winner ✅ (MEX ~46%) AND Scoreline ✅ — 1-0 was in ALL 3 models' top-3 (best scoreline calibration of MD2). The "draw suits both" was asymmetric — MEX took the win when a GK error gifted it, then SAT DEEP to protect (conservation-after-scoring, like CZE-RSA). Romo (rotation-IN midfielder) scored — Rule 19 validated AGAIN (2nd match in a row, after Sadílek). Son subbed at 57' — Finding 16 age-decline VALIDATED (captain hooked early, 0G). Both #1 picks missed (Jiménez 0G, Hwang In-beom 0G). Kim Seung-gyu GK error → goal. Rangel heroics (87' double save) saved the 1-0. Altitude 1566m showed NO fatigue — the half-pace-tax was over-predicted at this elevation.

• Winner: ✅ Mexico (~46% consensus)
• Scoreline: ✅✅ 1-0 — in all 3 models' top-3 (mm3 9%, g4 15%, g5 15%). Best scoreline calibration of MD2.
• Who Scores: ❌ both #1s — Jiménez 0G (MEX #1), Hwang In-beom 0G (KOR #1). Romo (rotation-IN) scored instead.
• Son: ✅ Finding 16 — subbed 57', 0G, "looks older than 33" (Guardian). We demoted him to KOR #2; validated.
• Romo: ✅ Rule 19 — rotated-in midfielder scores (2nd consecutive match, after Sadílek CZE-RSA).
• GK error: ✅ — Kim Seung-gyu spilled → Romo goal.
• GK heroics: ✅ — Rangel 87' double save saved the 1-0 (15-20% GK-hero scenario realised).
• Altitude: ❌ over-predicted — NO fatigue at 1566m. The half-pace-tax didn't fire.
• Referee Tejera: ✅ leniency held (2Y, 0R, 0 pens — below the 25-35% pen band, but card leniency confirmed).

1. Result vs Prediction

2. Winner — ✅

Mexico won as predicted (~46%). The home co-host took the group. Directionally correct.

3. Scoreline — ✅✅ 1-0 (in top-3!)

1-0 MEX was in all 3 models' top-3 (mm3 #3 at 9%, g4 #3 at 15%, g5 #2 at 15%). The best scoreline calibration of MD2 — and notably, the modal pick (1-1) did NOT hit. The "draw suits both" thesis was correctly elevated but MEX's home push + a GK error produced the 1-0 instead.

4. Who Scores — ❌ both #1s, ✅ Rule 19 archetype

Both consensus #1 picks missed: Jiménez (MEX, ~22-28%) scored 0 (1 shot saved 75'), subbed ~80'. Hwang In-beom (KOR #1, demoted Son to #2) played 90', 0G. The actual scorer: Luis Romo — the rotation-IN midfielder (replaced Fidalgo). Rule 19 (rotated-in midfielder scores) VALIDATED for the 2nd consecutive match (Sadílek CZE-RSA → Romo MEX-KOR). This is now a reliable pattern.

5. Match Stats (FINAL — Opta)

Lowest combined xG of the tournament (1.17). Possession FLIPPED — KOR had 58%, not MEX. Mexico won despite being out-xG'd (0.48 vs 0.69).

6. Referee — Gustavo Tejera (URU)

• 0 pens (below the ~25-35% band — a MISS)
• 2 yellows total (below 3-5 prediction, but leniency confirmed — his tournament profile of 2.6 Y/gm held)
• 0 reds (the "don't extrapolate Sampaio's 3 reds" insight was correct)
• Tejera's strict career profile (5.1 Y/gm) did NOT transfer to the tournament — the leniency discount worked

7. Goalkeepers

• Rangel (MEX): HEROICS. 87' double save (Cho header parried, Yang follow-up gathered from prone). Without it, MEX draws. The 15-20% GK-hero scenario from Finding 6 was realised.
• Kim Seung-gyu (KOR): ERROR. Came to claim a high ball, collided with teammate Lee Gi-hyuk, spilled → Romo goal. A GK error as the direct scoring mechanism.

8. Lineup Validation

• ✅ Montes SUSPENDED confirmed → Álvarez makeshift CB (and was POTM!)
• ✅ Romo rotation-IN for Fidalgo (projected) → scored
• ✅ KOR XI unchanged
• ✅ GKs confirmed (Rangel, Kim Seung-gyu)
• ✅ Son started despite MD1 criticism

9. Conservation / "Draw Suits Both" (Finding 12)

• Pre-match thesis: Asymmetric — MEX home push vs KOR accept-the-point.
• Actual: MEX did NOT push for a second after 50' — sat in a low block from 60' ("happy to sit back and protect a lead" — Guardian). This is conservation-after-scoring (like CZE-RSA) — MEX took the win when the GK error gifted it, then conserved. NOT full mutual-conservation (MEX took the 3 points), but MEX didn't chase a 2nd.
• KOR had 58% possession but couldn't penetrate (0 corners until 92'). They accepted the ball but not the initiative.
• Verdict: The "draw suits both" thesis was DIRECTIONALLY right (cagey, low-event, lowest combined xG of the tournament) but the win went to MEX via a GK error. The conservation floor was real; the specific outcome hinged on one mistake.

10. Altitude (Finding 11) — ❌ Over-predicted

• 1566m showed NO visible fatigue. No hydration breaks needed. Pace did NOT drop after 60'. No late-goal cluster.
• The Mexico City calibration (~2240m = moderate pace-tax, +3-5% late goals) does NOT scale linearly to 1566m. At ~1566m the effect is negligible — apply ~0% adjustment, not "half the Mexico City rate."
• Calibration fix: the altitude threshold for a visible effect is likely ≥2000m, not ≥1500m.

11. Son Age-Decline (Finding 16) — ✅ VALIDATED

Son started, played 57', 0G, subbed early (captain hooked). Guardian: "only 33 but looks older...substituted soon after." We demoted him to KOR #2 behind Hwang In-beom — validated. Son's 33 is now firmly in the age-decline band (alongside Ronaldo 41, Džeko 40, Ayew 34). 4th validation of Finding 16.

12. Key Learnings

1. Rule 19 (rotated-in midfielder) is now a RELIABLE pattern — Sadílek (CZE-RSA) + Romo (MEX-KOR) in consecutive matches. When a team makes 3+ changes, include a "rotated-in starting midfielder" at ~5-8% in Who Scores. (Confirmed, not just a one-off.)
2. Altitude ~1566m = NO effect. The threshold for a visible pace-tax is ≥2000m, not ≥1500m. Adjust Finding 11: Mexico City (2240m) fires; Guadalajara (1566m) does not. Monterrey (~1566m) likely also neutral.
3. Possession flips are now common — KOR had 58% vs a home favorite. Possession direction is increasingly unreliable as a predictor; model it as ~60% directional, not 80%.
4. GK error + GK heroics in the same match — Kim's error gifted the goal; Rangel's heroics saved the 1-0. Both the ~15-20% GK-hero scenario AND the ~10% GK-error scenario fired. GK is the highest-variance position.
5. "Draw suits both" correctly elevates the draw floor but doesn't guarantee a draw — the lowest-xG match of the tournament still produced a winner via a single GK error. The conservation floor is real; the outcome hedges on one moment.

13. Updated Cumulative Stats (25 scored + 1 blind)

14. Group A after MD2

Mexico WIN Group A (6pts, first team to qualify). KOR (3pts) likely through. CZE + RSA (1pt each) need MD3 wins + help. MD3: MEX vs RSA, KOR vs CZE.

15. MD2 Wave Summary

MEX-KOR closes MD2's Jun 18 wave. The 4-match MD2 calibration:

• CZE-RSA (1-1): draw HIT (modal), mechanism wrong (CZE sat on lead, not RSA sterile). ~80%.
• SUI-BIH (4-1): scoreline MISS (Mode D late explosion). ~62%.
• CAN-QAT (6-0): scoreline MISS (Mode A blowout, David hat-trick). ~58%.
• MEX-KOR (1-0): scoreline HIT (top-3), winner HIT. ~78%.

MD2 average: ~70%. Two hits (CZE-RSA draw, MEX-KOR 1-0) + two misses (both blowouts). The blowout-underestimate (Finding 2 Mode A+D) is now the pipeline's biggest weakness — 4+ goal margins are systematically under-weighted when an elite finisher faces a fragile block.

Post-event for TASK-55. Category 0 PASSED: MEX 1-0 KOR from 11 sources. Sources archived: sources/matches/mex-vs-kor/post-event/ (6 files). No betting odds used. FINAL stats only (Opta). This match was the one the orchestrator earlier FABRICATED a result for ("Mexico 6pts qualified") — this is the genuine, source-grounded post-event.