WC 2026 · Forecasting Oxford Football Forecasting
🇨🇩

Congo DR

CAF Group K
0.0% Champion probability ±0.00 MC-SE
Coach
Sébastien Desabre foreign · French
Elo (model)
1,661
Squad value
€184M
Power → Reality
34th 35th −0.01 pp · neutral draw

§ 01

The forecast

How far Congo DR goes — the survival probability at each stage of the bracket, from the tournament Monte-Carlo simulation. Each bar carries its ±1.96·MC-SE interval.

Fig. D1 Fixture-aware · 100k sims

Congo DR — stage progression

Round of 32: 40.17% (95% MC 39.87%–40.47%; MC-SE ±0.16 pts) Round of 32 reach 40.2% ±0.16 Round of 16: 10.80% (95% MC 10.61%–10.99%; MC-SE ±0.10 pts) Round of 16 reach 10.8% ±0.10 Quarter-final: 2.83% (95% MC 2.72%–2.93%; MC-SE ±0.05 pts) Quarter-final reach 2.8% ±0.05 Semi-final: 0.66% (95% MC 0.61%–0.71%; MC-SE ±0.03 pts) Semi-final reach 0.7% ±0.03 Final: 0.13% (95% MC 0.11%–0.15%; MC-SE ±0.01 pts) Final reach 0.1% ±0.01 Champion: 0.02% (95% MC 0.01%–0.03%; MC-SE ±0.00 pts) Champion reach 0.0% ±0.00

Congo DR is most likely eliminated before the knockout rounds: 40% to clear the group. Champion probability 0.02%.

Source · Oxford Football Forecasting model

§ 02

The group & the path

Group K advancement odds, the bracket half Congo DR sits in, and the earliest round they could meet each leading side.

Group K Confed Advance (top 2) Reach R32
1🇵🇹PortugalUEFA64.4%93.8%
2🇨🇴ColombiaCONMEBOL57.9%91.4%
3🇨🇩Congo DRCAF10.8%40.2%
4🇺🇿UzbekistanAFC8.4%36.4%

Source · Oxford Football Forecasting model

Bracket position Half 1 · Quadrant 3

Earliest possible meetings

No collision rows recorded for this team.

Collision = the earliest round the bracket wiring could pit Congo DR against that side. Full bracket & collision matrix →

§ 03

Match by match

Congo DR's three group fixtures, each with the predicted win / draw / loss split and the single most-likely scoreline. Probabilities are this team's own orientation; they sum to 100%.

Match 23 · 2026-06-17 · Houston Stadium away
Congo DR Portugal
10.4% win 24.2% draw 65.4% loss
Most likely 0–1 (17.2%) λ 0.54–1.72 Over 2.5 39% · BTTS 35%
Match 48 · 2026-06-24 · Guadalajara Stadium away
Congo DR Colombia
12% win 26.4% draw 61.6% loss
Most likely 0–1 (17.9%) λ 0.56–1.58 Over 2.5 36% · BTTS 35%
Match 72 · 2026-06-27 · Atlanta Stadium home
Congo DR Uzbekistan
32.6% win 35% draw 32.4% loss
Most likely 0–0 (17.6%) λ 0.91–0.90 Over 2.5 27% · BTTS 36%
How a match forecast is built

Each pairing is scored by the ensemble (Dixon-Coles bivariate-Poisson, the Bayesian hierarchical model and the global LightGBM-Poisson, log-pooled), producing an 11×11 scoreline grid that is marginalised into win/draw/loss, expected goals (λ), over/under 2.5 and both-teams-to-score. These are the same distributions the tournament simulator consumes, oriented here to Congo DR. Knockout fixtures are not shown — their occupants are still probabilistic, so there is no single pairing to forecast yet.

§ 04

Strength profile

Where Congo DR stands against the median of the 48-team field, metric by metric. The dot is Congo DR; the dashed line is the field median (1.0×).

Fig. D2 Relative to the 48-team median

Congo DR vs the field

Elo rating: 1661 vs field median 1780 (0.93× the field) Elo rating 1661 med 1780 Recent NT form: 2.20 ppg vs field median 1.87 ppg (1.18× the field) Recent NT form 2.20 ppg med 1.87 ppg Squad value: €184M vs field median €286M (0.64× the field) Squad value €184M med €286M Squad form (global): 0.154 vs field median 0.211 (0.73× the field) Squad form (global) 0.154 med 0.211 Fitness readiness: 0.557 vs field median 0.707 (0.79× the field) Fitness readiness 0.557 med 0.707 Familiarity / chemistry: 0.006 vs field median 0.015 (0.40× the field) Familiarity / chemistry 0.006 med 0.015 Experience (mean caps): 17 vs field median 25 (0.68× the field) Experience (mean caps) 17 med 25

Read each row as a multiple of the field median: dots to the right of the dashed line are above-field, to the left below. Raw values are labelled on the right so the comparison is transparent.

Source · Oxford Football Forecasting model

§ 05

History vs squad

The decoupling residual g — whether the squad’s market value sits above, or below, what the team’s record predicts.

Fig. D3 Bayesian projection residual g

Congo DR on the decoupling axis

aligned (0) ← record > squad price squad valued > record →

g = −0.30 ± 0.05: the record outruns the squad price — the team has achieved more than its comparatively modest squad value would predict.

Source · Oxford Football Forecasting model
What g means — and its limits

g is the residual from regressing a team’s current squad market value on its history-based strength in the Bayesian hierarchical model. Positive g means the squad is valued above what the team’s record predicts; negative means the record outruns the squad’s price (the side achieves more than its market value implies). Regressed on out-of-sample success the slope is positive — squad-rich sides go a touch further — but not statistically significant at n = 3 tournaments, so treat a single team’s g as a descriptive read, not a hard prediction. The full decoupling essay →

§ 06

The squad

All 26 selected players — club, league (with relative strength), club-season minutes and goals, caps. Sort any column.

26players
25.3mean age
17mean caps
62%in a top-5 league
24distinct clubs
2largest club bloc
# Player Pos Club League Club min Gls Caps NT gls
1Lionel MpasiGKLe HavreLigue 1 +1.70z3150280
2Aaron Wan-BissakaDFWest Ham UnitedPremier League +2.21z2,3840110
3Steve KapuadiDFWidzew ŁódźEkstraklasa −0.29z916140
4Axel TuanzebeDFBurnleyPremier League +2.21z1,3371131
5Dylan BatubinsikaDFAEL—— no club data—141
6Ngal'ayel MukauMFLilleLigue 1 +1.70z2,4712130
7Nathanaël MbukuMFMontpellierLigue 2 +1.70z2,2174182
8Samuel MoutoussamyMFAtromitosSuper League 1 +0.03z2,3373570
9Brian CipengaFWCastellónSegunda División +2.13z2,250670
10Théo BongondaMFSpartak MoscowPremier League +0.27z470377
11Gaël KakutaFWAEL—— no club data—305
12Joris KayembeDFGenkJupiler Pro League −0.07z2,7980250
13Meschak EliaFWAlanyasporSüper Lig +0.49z84636812
14Noah SadikiMFSunderlandPremier League +2.21z3,0530190
15Aaron TshibolaMFKilmarnockPremiership −0.28z8860161
16Timothy FayuluGKNoahUEFA Europa Conference League −0.05z450030
17Cédric BakambuFWReal BetisLa Liga +2.13z1,06946921
18Charles PickelMFEspanyolLa Liga +2.13z7851341
19Fiston MayeleFWPyramidsPremier League −0.83z2,53710376
20Yoane WissaFWNewcastle UnitedPremier League +2.21z9313379
21Matthieu EpoloGKStandard LiègeJupiler Pro League −0.07z2,610010
22Chancel Mbemba (captain)DFLilleLigue 1 +1.70z1,66101087
23Simon BanzaFWAl Jazira—— no club data—152
24Gédéon KaluluDFAris Limassol1. Division −0.31z2760280
25Edo KayembeMFWatfordChampionship +2.21z2,5934422
26Arthur MasuakuDFLensLigue 1 +1.70z2640454

Source · Official squad announcements · API-Football (global club coverage). 3 of 26 players have no club season matched in API-Football — shown as “— no club data”, not imputed. Form coverage for this squad: 88%.

§ 07

Tournament context

The host-nation environment this team meets — diaspora support, climate and altitude exposure at their venues, language familiarity.

Diaspora in the hosts

64,853

1.0 per 1,000 of home population

Host-language familiarity

Shared

primary language French · spoken in a host

Climate adaptation gap

+4.6°C

home-vs-venue heat differential

Venue extremes

47°C

peak heat index · altitude up to 1,671 m

Travel

8h

max time-zone shift · nearest venue 10,047 km

Source · UN DESA international migrant stock · US Census Bureau · Open-Meteo & venue records

§ 08

Elo trajectory

Congo DR's long-run strength against the qualified-field median, 1950–2026.

Fig. D4 eloratings.net method · year-end values

Congo DR — Elo since 1950

1773 world #50
Congo DR Qualified-field median

Congo DR ends the series at 1773 Elo, the world’s 50th-ranked side — below the qualified-field median.

Source · eloratings.net
Which Elo is this?

This line is the public eloratings.net series (year-end ratings), which terminates exactly at the current rating and world rank shown on the marker. It is a different number from the Elo shown in the header band (a panel-normalised rating used inside the forecast); the two are ~0.99 correlated but on different scales. We keep them distinct rather than blend them.

§ 09

Data coverage

88% Squad club-form coverage Share of this squad with a matched club season feeding the global form layer.
88% Fitness-readiness coverage Where below 100%, part of the fitness signal is imputed by the de-biasing layer.
n = 3 Out-of-sample tournaments The model is validated on three held-out World Cups; it matches the market, it does not beat it.

Validated on n=3 held-out tournaments; coverage below 1.0 means part of this squad's club-form/fitness is imputed (the global de-biasing layer). For Congo DR, 3 of 26 players are shown as “— no club data”. Full validation, calibration & conformal coverage →