Group A · Matchday 3

South AfricavsSouth Korea

2026-06-24·19:00 localPredictions finalised

Snapshot · 2026-07-14Model 1.0.0Final prediction · locked 24 Jun, 22:56 UTCSouth Africa·South Korea·Head-to-head →·
Full time · forecast gradedSouth Africa 1 0 South KoreaThe locked pre-match forecast has been graded against this result.See the calibration recap →

The forecast

Match-outcome probability

  • South Africa win
    20.5%
  • Draw
    28.5%
  • South Korea win
    51.0%

A 228-point Elo gap frames this as a significant mismatch, yet the model still gives South Africa a 20% probability of a result — enough to make this more than a formality.

Rank checkFIFA ranks South Africa #61 in the world; the model ranks them #34 in this tournament field, 27 places higher than the FIFA list suggests. All 48 compared →
Likeliest score0–116.8%
First goal0-15'26.6%
Both teams score35.9%
Over 2.5 goals28.5%
Top scorerHeung-min12.6%
Expected goals0.7 - 1.1
Loading pitch visualisation...

Why the model says this

Favoring South Africa

  • ·South Africa has a better goal difference of -1 over their last six matches, compared to South Korea's -5 over the same period.
  • ·South Africa scored 8 goals in their last six matches, exceeding South Korea's 5 goals during the same timeframe.

Favoring South Korea

  • ·South Korea holds a significant Elo rating advantage, with a delta of 228 points over South Africa.
  • ·South Korea is ranked 22nd in the FIFA rankings, substantially higher than South Africa's 61st position.
  • ·South Korea's expected goals (xG) of 1.58 are nearly double South Africa's 0.81, indicating superior attacking output.
  • ·The Elo model gives South Korea a 67.8% chance of winning, and the stacking model projects a 61.2% win probability, highlighting strong model consensus for an away victory.

What the model can't fully price

  • ·Squad availability issues, with 2 players carrying fitness doubts across both squads, including 1 projected starter, are not fully integrated into the model's current lineup channel.
  • ·The specific venue and city for the match are not provided, which means factors such as travel impact or local conditions are not accounted for in the probabilities.
  • ·Team motivation for this Group A Matchday 3 fixture, which can be heavily influenced by prior group results and qualification scenarios, is not a factor the model explicitly prices.

Form check

South Africa

Declining

South Africa's recent form has been inconsistent, recording two wins, one draw, and three losses in their last six matches. They have struggled to build momentum, with their last two results being a loss and a draw.

8 goals scored in last 6 matches

South Korea

Declining

South Korea's form is mixed, with three wins and three losses in their last six outings. They enter this fixture on the back of two consecutive defeats in friendly matches, conceding 5 goals across those two games.

2 consecutive losses in recent friendlies

Analysis

How it plays out

South Africa's balanced setup will need to hold shape against South Korea's direct transition game. The risk for South Africa: getting caught between attacking and defending. South Africa will expect to hold 52% possession. South Korea need their shape to stay compact without the ball and be clinical when they win it back.

What decides it

South Korea will concede possession willingly and attack in transition. Their defensive block needs to hold without fouling in dangerous areas. Son Heung-min's 12.6% scoring probability is the highest in this fixture. Containing that output is South Africa's primary defensive task.

Off the pitch

Hugo Broos (5 years in charge of South Africa) vs Hong Myung-bo (2 years). That tenure gap shows up in squad familiarity and set-piece coordination.

The angle

Likely the last World Cup for Kim Seung-gyu. Tournament experience at this level is hard to quantify but hard to replace.

Goals & scorelines

Likeliest score 0–1 (16.8%) · xG 0.7 - 1.1

Expected goals

South Africa
0.73
South Korea
1.13

Mean of the Dixon-Coles joint goal distribution. Same fit that produces the most-likely-scoreline list below.

Most likely scorelines

  • 0–1
    16.8%
  • 0–0
    16.4%
  • 1–1
    13.6%
  • 1–0
    10.6%
  • 0–2
    9.9%

From the Dixon-Coles joint Poisson with the low-score correction. Scorelines are listed in probability order; this is a description of the model's distribution, not a recommendation.

Most likely half-time scorelines

  • 0–0
    40.0%
  • 0–1
    21.8%
  • 1–0
    13.9%
  • 1–1
    8.6%
  • 0–2
    6.3%

Same Dixon-Coles fit as the full-time list above, with rates halved to a 45-minute window and the low-score correction applied to that 1st-half block. The 0-0 row sits higher here than at full-time because fewer minutes have elapsed.

Goal totals

  • More than 0.5 goals
    83.6%
  • More than 1.5 goals
    56.2%
  • More than 2.5 goals
    28.5%
  • More than 3.5 goals
    11.8%
  • More than 4.5 goals
    4.1%
  • More than 5.5 goals
    1.2%
  • Both teams score
    35.9%

Each row is the probability the match finishes with more than the listed number of goals. Both-teams-to-score is the probability each side scores at least once. All values are marginals of the Dixon-Coles joint goal grid that produces the scoreline list above — not market lines or any other operator construct.

Event-typed probabilities

  • South Africa clean sheetOpposing team scores zero32.4%
  • South Korea clean sheetOpposing team scores zero48.2%

Derived from the same Dixon-Coles joint distribution as the scoreline list. These are descriptive event probabilities — see CLAUDE.md §3/§4 (formerly COMPLIANCE.md §4.2.7) for the framing the project uses.

Win-margin probability

  • South Africa by 4+
    0.3%
  • South Africa by 3+
    1.5%
  • South Africa by 2+
    6.9%
  • South Africa by 1+
    22.9%
  • Draw
    32.9%
  • South Korea by 1+
    44.2%
  • South Korea by 2+
    19.1%
  • South Korea by 3+
    6.1%
  • South Korea by 4+
    1.6%

Each row is the probability the match ends with the listed margin or larger in that direction. Marginal of the Dixon-Coles joint goal grid; the “by 1+” rows plus the draw row sum to 1.

How the match unfolds

Over 2.5 goals 28.5% · BTTS 35.9%

Game state through the match

0%25%50%75%100%0'15'30'45'60'75'90'
  • South Africa ahead23.7%
  • Level31.3%
  • South Korea ahead45.0%

Probability of each game state at minutes 0, 15, 30, 45, 60, 75, 90 — derived from two independent thinned-Poisson processes with the Dixon-Coles per-team rates. The three lines always sum to 1 at each minute. The right column shows the state at the match's closing minute.

When the first goal arrives

  • 0–15
    26.6%
  • 15–30
    19.5%
  • 30–45
    14.3%
  • 45–60
    10.5%
  • 60–75
    7.7%
  • 75–90
    5.7%
  • No goal
    15.6%

Probability the match's first goal arrives in each 15-minute window. Homogeneous Poisson with combined rate λ = λh + λa from the Dixon-Coles fit; the seven rows (six windows + no-goal tail) sum to 1.

Half-time / full-time grid

Joint probability of half-time and full-time results
HT ↓ / FT →HSouth Africa winDDrawASouth Korea win
HSouth Africa ahead13.3%4.0%1.3%
DLevel9.4%23.6%16.1%
ASouth Korea ahead0.8%4.0%27.5%

Each cell is P(half-time result, full-time result). All nine cells sum to 1. Derived from a halved-λ Dixon-Coles fit for the first half plus an independent-Poisson second-half convolution.

Comeback probability

  • South Africa trail at HT, avoid defeat at FT
    4.8%
  • South Korea trail at HT, avoid defeat at FT
    5.3%

Joint probability — P(side trailing at half-time AND avoiding defeat at full-time). NOT conditional on trailing at HT. Derived from the same half-time / full-time decomposition that produces the HT/FT grid above; a tied first half is neither a home nor an away comeback opportunity.

Cards

  • Expected yellow cardsMean of the Poisson on total yellow cards.3.45
  • Total yellows over 2.567.0%
  • Total yellows over 3.545.3%
  • Total yellows over 4.526.5%
  • Any red cardP(at least one red card in the match).9.5%

Referee not yet assigned. Using the 2026 pool-mean per-match rate as a placeholder; the model picks up the referee's personal rate once the assignment is published. Total yellow cards modelled as a Poisson with mean equal to two team baselines plus the referee's deviation from the pool mean. Reds are modelled the same way, independently. See /docs/methodology/.

Teams & players

Top scorer: Heung-min (12.6%)

Match detail

South Africa

Model-rated key players: Evidence Makgopa (FW) — P(scores) 4.7%; Oswin Appollis (FW) — P(scores) 4.7%; Thapelo Morena (FW) — P(scores) 4.7%.

How they play

South Africa under Hugo Broos play a balanced game with 52% possession. They apply moderate pressing intensity (PPDA 23.9). They favour high-quality chances (xG/shot 0.189, among the best in the field).

What they must execute

South Africa will need to leverage their strengths while managing the physical demands of a tournament spread across three host countries.

Storylines
Club core: 9 of 26 predicted-squad players play their club football for Mamelodi Sundowns — a single-club spine on the international side.
Model bold: Model rates them #44 by tournament-winner probability — 17 places higher than FIFA #61.
Local-league core: Only 1 of 26 predicted-squad players played in a top-5 European league last season — the rest play home or in non-top-5 leagues.

South Korea

Model-rated key players: Son Heung-min (FW) — P(scores) 12.6%; Oh Hyeon-gyu (FW) — P(scores) 3.5%; Joo Min-kyu (FW) — P(scores) 2.9%.

How they play

South Korea under Hong Myung-bo play a counter attacker game, with just 44% possession — among the lowest in the field. Their likely shape is a 4-2-3-1, though they have also used 4-3-3 and 4-4-2. They apply moderate pressing intensity (PPDA 25.0).

What they must execute

South Korea rely on defensive discipline and quick transitions — absorbing pressure and converting turnovers into attacking chances. Concentration and defensive organisation for full 90-minute stretches will determine whether the approach holds against top opposition. Managing minutes for Kim Seung-gyu across what could be seven matches will test the coaching staff's rotation planning.

Storylines
Dead-ball: Son Heung-minTakes corners, free kicks, and penalties — the team's dead-ball threat.
Long-haul: Travels 34,978 km across 2 venues in the group stage — one of the longest itineraries in the field.
Local-league core: Only 4 of 24 predicted-squad players played in a top-5 European league last season — the rest play home or in non-top-5 leagues.
Set-piece outlook

South Africa historically converts 10.9% of xG from set-pieces, contributing 0.08 expected set-piece goals in this fixture. South Korea converts 12.6% from set-pieces (0.14 expected). Combined, the model expects 0.22 set-piece goals across the 90 minutes.

  • P(South Africa scores set-piece goal) 7.7%
  • P(South Korea scores set-piece goal) 13.3%
  • P(set-piece goal in match) 20.0%

South Korea: Son Heung-min on corners (43 corners) (per fbref 2022 23)

Penalty outlook

If a penalty is awarded to South Africa, the model gives 71.4% conversion, 72.5% for South Korea.

South Korea primary PK: Son Heung-min (1/1 in 2020-21, per fbref 2022 23).

Derived from the model's per-fixture forecast joint and supporting reference data (predicted squads, set-piece xG share, PK posteriors, club minutes). See /docs/methodology/ for the full methodology.

Tactical forecast

South Africabalanced
PPDA
23.9
Possession
52%
Directness (yds/pass)
6.8
Long balls/90
44
Set-piece xG
11%
South Koreacounter-attacker
PPDA
25.0
Possession
44%
Directness (yds/pass)
7.1
Long balls/90
40
Set-piece xG
13%

Style profile per side from StatsBomb open-data aggregation across recent international tournaments (Euro 2020/2024, Copa America 2024, AFCON 2023, World Cup 2018/2022). The tactical-fingerprint badge maps each team’s observed style vector into one of eight canonical archetypes via a rule-based classifier; teams with fewer than three matches of qualifying coverage carry an “insufficient-data” label rather than being forced into a default. Sides outside the StatsBomb-open corpus use FotMob team match stats from recent qualifiers and friendlies instead (possession and shot volume only), marked as partial coverage. PPDA = passes the side allows per defensive action (lower = more intense press). Formation distributions are not yet produced — that head of the §2.7 classifier is pending its own data pull. See /docs/methodology/.

Squad depth

Most irreplaceable starters

South Africa

  1. Nkosinathi SibisiCentre-backCover: Mbekezeli Mbokazi · 0.000.41gap
  2. Aubrey ModibaFull-backCover: Thabang Matuludi · 0.180.28gap
  3. Khuliso MudauFull-backCover: Thabang Matuludi · 0.180.24gap

South Korea

  1. Hwang In-beomDefensive midfieldCover: Park Jin-seob · 0.280.46gap
  2. Lee Kang-inAttacking midfieldCover: Lee Jae-sung · 0.410.46gap
  3. Cho Gue-sungStrikerNo natural backup0.31gap

Gap = how far a side's rating at the position falls from the starter to his likely in-squad replacement (named under each name). Larger = harder to replace. Descriptive metric, does not feed the published probabilities. Methodology →

Match conditions

  • AltitudeNear sea level521 m
  • Avg temperatureFive-year mean over the tournament window27.7 °C
  • Avg humidity65%
  • Heat stressShade WBGT ~29.1 °CModerate heat stress
  • Pitch surfacenatural grass

Natural-grass football stadium; the pitch was refreshed ahead of the tournament.

Heat stress is a shade Wet-Bulb Globe Temperature proxy from the venue's climatology mean temperature and humidity; FIFA mandates cooling breaks at WBGT 32 °C. Evening kickoff (local time). These are long-window averages, not a match-day forecast, and they are not inputs to the forecast.

Top scorers · P(scores in this match)

South Africa
South Korea

Per-player scoring rate from Model #5 (`p_score_per_match`). Reflects each player's npxG/90, expected minutes, team xG share, and the average opposing-team defence. See /docs/methodology/.

Recent match form

Last match player ratings

South Africa

vs Canada · avg 7.7

9
Ronwen WilliamsGK
ATK
DEF
PAS
7
Khuliso MudauRB
ATK
DEF
PAS
7
Yaya SitholeCM
ATK
DEF
PAS

South Korea

vs Mexico · avg

Player scores from official highlight analysis of each team's most recent match. Observational, not a model input. Methodology →

Video analysis: player performance

Per-player ratings and event breakdowns from official highlights analysis. Tap a player to see their full match timeline.

South Africa
8
Siphesihle Sithole20'–20'

Scored a crucial goal and was involved in winning a foul that led to a booking.

1goals

Match timeline

20'Siphesihle Sithole scores for South Africa with a close-range finish after a well-worked attacking move.
8
Sipho Zondi

Made crucial saves, particularly in the second half, to secure the clean sheet and preserve the lead.

8
Mashego

Scored a goal and was a constant threat with his dribbling and close-range efforts.

7
Apolisi

A constant offensive threat with numerous shots and a powerful header, though unable to score himself.

7
Ronwen Williams

Made important saves in his time on the field, contributing to the team's defensive solidity and clean sheet.

7
Modiba

Made a vital goal-line block that prevented a certain goal, a key defensive contribution.

South Korea
9
KOR Goalkeeper15'–15'

Made numerous crucial and spectacular saves, single-handedly keeping his team in the match despite constant pressure.

1saves

Match timeline

15'Korean Republic goalkeeper makes a save from a close-range shot
6
KOR #3

Made a defensive block but was part of a defense that ultimately conceded two goals.

4
Do-hyun Ryu

Received a yellow card for a foul and had no other notable positive contributions.

4
Kyung-min Bae

Received a yellow card for a late challenge and had no other notable positive contributions.

Match observations

  • The match was a tightly contested affair, with South Africa securing a narrow 1-0 victory.
  • Both teams had periods of attacking pressure, but South Africa's early goal proved to be the difference.
  • The second half saw Republic of Korea push for an equaliser, but they were ultimately thwarted by solid defending and goalkeeping.

Under the hood

Model-by-model comparison

South Africa vs South Korea

High disagreement (17.8%)
ModelWeightHomeDrawAway
EloRating-based strength estimate32%
15.9%
22.0%
62.1%
Dixon-ColesGoal-process model with low-score correction63%
23.0%
32.1%
44.9%
Hierarchical PoissonBayesian model with confederation pooling6%
24.2%
31.5%
44.3%
Bayesian stackingLearned-weight combination
15.2%
32.4%
52.3%
Ensemble (published)Uniform average + isotonic calibration
19.9%
30.3%
49.8%
Home spread: 8.3%
Draw spread: 10.1%
Away spread: 17.8%
How each model works
Elo
Each team carries a single strength rating updated after every match by a margin-aware K-factor. Match probabilities come from the logistic function of the rating gap. Elo is fast-adapting but coarse — it sees only who won and by how much, not how the goals were scored.
Dixon-Coles
A Poisson regression on team-level attack and defence parameters, fitted via maximum likelihood with an exponential time-decay weighting. The Dixon-Coles correction adjusts the four low-score cells (0-0, 1-0, 0-1, 1-1) where independent Poisson underestimates dependence. Produces full scoreline distributions, not just H/D/A.
Hierarchical Poisson
A Bayesian Poisson model fitted via MCMC (PyMC) with hierarchical priors that pool attack and defence parameters within confederations. Shrinks small-sample teams toward their confederation mean — helpful for nations with few recent competitive fixtures. Slower to fit but better-calibrated on the tails.
Bayesian stacking
Optimises simplex weights (w_elo, w_dc, w_hp) to maximise the leave-one-out log-score across a walk-forward backtest (Yao et al. 2018). The result is a weighted average of the three component models' probabilities, then isotonic-calibrated. Adds no extra features — just learns which component to trust more from historical accuracy.
Ensemble (published)
Equal-weight average of all three component models, followed by per-class isotonic regression calibration fitted on 24 months of walk-forward out-of-fold predictions. This is the probability published on the site. The uniform mean is deliberately simple — it avoids overfitting to the stacking weights' training window.

Three independent component models feed two combination strategies. The uniform ensemble is the published probability; Bayesian stacking uses learned weights. Amber bars flag >5pp divergence from the published number. Full methodology

Probability decomposition (transparency surface)

  • Baseline ensemble — P(South Africa win)18.5%
  • + Lineup contribution0.0pp
  • + Style-matchup contribution0.0pp
  • Published P(South Africa win)18.5%
South Africa
18.5%
Draw
26.8%
South Korea
54.7%

Decomposition of the published P(South Africa win) into the calibrated- baseline plus contributions from the §2.3 expected-XI lineup delta and the §2.7 style-matchup interaction. The §2.7 roadmap is explicit that style effects are second-order to team strength — single-digit-percentage P(win) shifts on extreme style matchups, near-zero on balanced ones. We surface the decomposition for transparency even when the contributions are small; the baseline carries the prediction. Methodology: /docs/methodology.

For this fixture both contributions round to under 0.05pp — the fitted style-matchup pair effect is in the small-magnitude regime the model expects to dominate.

Latest news & match context

Team news

No recent headlines for South Africa or South Korea.

Match conditions
Stage:
Group A · Matchday 3
Date:
24 Jun
Availability

South Africa

South Africa come in at close to full strength.

South Korea

South Korea come in at close to full strength.

What it means

South Africa and South Korea both come in at close to full strength, so the forecast rests on baseline team strength rather than late team-news swings.

Availability from the predicted squads and injury feed; forecast adjustments from the model's own decomposition. See /docs/methodology/.

Standard Pass

This match is a free preview

You're seeing the model's full forecast for this fixture for free. Unlock the same depth: probabilities, expected goals, scoreline distributions, and per-player scoring, for all 104 matches with a Standard Pass, valid through the tournament.

Get the Pass, $15

Every forecast graded against the real result, scored on 987 matches since 2014. See the scorecard.

24h money-back, no questions asked·No subscription, no auto-renewal·Access through 31 Dec 2026. See refund policy.