Forma
Forma · World Cup 2026 Hub · as of 2026-07-12

Every club watches the World Cup.
Forma watches what it tells us.

Five modules built on the open WC26 dataset: a tactical fingerprint of all 48 nations, a Hidden Value Index of tournament risers, a club→country talent pipeline, a knockout simulator running 10,000 brackets, and a daily Forma Spotter. Refreshing daily.

100 of 104 matches inR32: 16 played, 0 to go28 eliminated
Matches
100/104
Teams placed
48
Bracket sims
10,000
Risers surfaced
30
Module 01 — Tactical Identity Map

48 nations. One map.

Six tactical axes turned into a single fingerprint per nation, then clustered into six archetypes. The dataset has no event coordinates, so a couple of axes are proxies — explained alongside the chart.

← STYLE AXIS 1 →← STYLE AXIS 2 →MEXRSAKORCZECANBIHQATSUIBRAMARHAISCOUSAPARAUSTURGERCUWCIVECUNEDJPNSWETUNBELEGYIRNNZLESPCPVKSAURUFRASENIRQNORARGALGAUTJORPORCODUZBCOLENGCROGHAPAN
What the axes mean

The map is a 2D projection (PCA) of six tactical axes. Two dots close together = two teams that play in a similar way. Dot colour is the cluster they fall into. Dot opacity reflects sample size.

  • PossessionShare of the ball. Higher = team likes to keep it; lower = team is happy without it.
  • Shot rateShots per ball-time. Higher = vertical, direct intent. A proxy for verticality (no event coordinates in the open dataset).
  • Foul rateFouls per match. A pressing proxy — high-press sides foul more, but referee profile shifts this too.
  • Set-piece relianceCorners per shot. Higher = chances disproportionately come from dead balls.
  • Defensive loadSaves per match. Higher = goalkeeper sees more work — usually a defending team.
  • ChaosVariance of opponent xG. Higher = scoreline swings game-to-game; lower = stable shape.
Possession control
· n=11
Long spells with the ball, sustained build-up, low chaos.
ARGBRACOLENGFRAGERMEXSENESPTURURU
Vertical transition
· n=8
Short possessions, fast attacks, high shots-per-touch.
ALGCIVIRQNORPORKORTUNUSA
Compact counter
· n=9
Cede possession, soak pressure, strike on turnover.
CPVCODCZEGHAIRNJORNZLPANRSA
Set-piece dependent
· n=4
Disproportionate share of chances from corners and dead balls.
AUSCANNEDUZB
Defensive lockdown
· n=5
High save and block load, low offensive output, ugly but effective.
CUWPARQATKSASWE
Chaos creators
· n=11
High variance both ways — entertaining, unpredictable, hard to model.
AUTBELBIHCROECUEGYHAIJPNMARSCOSUI
Tournament fingerprint, not season fingerprint. The dataset has no event coordinates, so 'Shot rate' is a verticality proxy and 'Foul rate' is a pressing proxy — both will move with referee profile and game state. Sample size shown per team; expect identities to sharpen through the knockouts. Dot opacity reflects sample size.

Style Matchup Predictor

From the Identity Map vector
Mexico
Possession control
xG/m
1.53
xGA/m
0.91
Expected goals (per match)
1.50 v 0.84
Forecast · blend of each side's attack & opponent's defence
South Africa
Compact counter
xG/m
0.76
xGA/m
1.48
Z-delta per axis · MEX → RSA
Possession
+0.43σ
A: 0.52σB: 0.09σ
Shot rate
+1.15σ
A: 0.39σB: -0.76σ
Foul rate
-0.21σ
A: 0.23σB: 0.44σ
Set-piece reliance
+0.07σ
A: -0.57σB: -0.64σ
Defensive load
-1.87σ
A: -1.35σB: 0.52σ
Chaos
+0.63σ
A: -0.56σB: -1.19σ
Predicted clash
  • Ball controlEven fight for possession — both sides within 0.43σ of each other.
  • Chance creation tempoMexico creates shots faster per ball-time — expect them to drive the chance count.
  • Defensive loadSouth Africa's keeper sees more work per match — usually a sign of conceding shape and territory.
  • Game-state varianceMexico's match scorelines swing more — higher xG variance, harder to model.

Same vector that powers the Identity Map, surfaced as a head-to-head. With ~3 group matches per team, treat as directional — individual quality and game-state can override the style signal.

Module 02 — Hidden Value Index

Whose tournament contribution outruns their market rank.

Tournament Contribution Score versus current Transfermarkt valuation, position-normalised. Tiered as Strong Riser, Riser, or Watch. The percentile-points delta is a directional signal — not a quoted valuation or transfer-fee forecast.

Recruitment use-case · one-click filters
Position
Age
5 of 30 shown · 527 eligible
Timothy John PayneDEF · 32 · NZL
Wellington Phoenix FC
Strong Riser
Current market value
€350K
+92 percentile pts above market rank

Plays for Wellington Phoenix FC. 1 g+a in 3 matches' worth of minutes. against above-average opposition.

Omar RekikDEF · 24 · TUN
NK Maribor
Strong Riser
Current market value
€600K
+90 percentile pts above market rank

Plays for NK Maribor. 1 g+a in 2 matches' worth of minutes. against above-average opposition.

Lionel Nzau MpasiGK · 31 · COD
Le Havre AC
Strong Riser
Current market value
€400K
+88 percentile pts above market rank

Plays for Le Havre AC. against above-average opposition.

Elijah Henry JustMID · 26 · NZL
Motherwell FC
Strong Riser
Current market value
€2.5M
+84 percentile pts above market rank

Plays for Motherwell FC. 3 g+a in 3 matches' worth of minutes. against above-average opposition.

Manel Farhan Ehsan HaddadDEF · 32 · JOR
Al Hussein SC
Strong Riser
Current market value
€2.2M
+84 percentile pts above market rank

Plays for Al Hussein SC. 1 g+a in 3 matches' worth of minutes. against above-average opposition.

This is what Forma's Recruitment module does year-round.
Across 30,000+ players. Not just the ones playing in a World Cup.
See how →

Players need 180+ tournament minutes to surface. The percentile-points delta compares a player's tournament contribution rank against their market-value rank within their position group. It is a directional signal, not a quoted valuation or transfer-fee forecast.

Module 03 — Talent Pipeline

Where the World Cup is actually trained.

Club→country minutes flow. Some nations run on the Premier League. Some on Liga MX. A few on leagues you wouldn't expect.

Callout #1
Premier League supplies the most WC26 minutes
37,690 minutes (19% of the tournament) — 11 points ahead of Bundesliga.
Callout #2
England's squad runs heaviest through Premier League
4,862 minutes — 78% of their squad's tournament minutes wear Premier League club shirts day-to-day.
Callout #3
England keeps it at home
78% of their WC26 minutes come from clubs in their own top flight.
Nation → club league
ARGSUIMARNORBELENGESPFRAPARCOLEGYBRACANPORMEXUSANEDCPVAUSGERSENSWECIVCROBIHJPNALGAUTGHARSAECUCODOtherPremier LeagueBundesligaLigue 1Serie ALa LigaMajor League SoccerEgyptian Premier LeagueLiga MXEredivisiePrimeira LigaBrasileirão Série ASouth African PSLEFL Championship
UEFACONMEBOLCONCACAFAFCCAFOFCOther

Clubs supplying the most WC26 minutes

  1. 1
    FC Bayern München
    Bundesliga · 17 players
    5,505 min
  2. 2
    Manchester City FC
    Premier League · 18 players
    5,116 min
  3. 3
    Arsenal FC
    Premier League · 13 players
    4,287 min
  4. 4
    Paris Saint-Germain
    Ligue 1 · 14 players
    4,170 min
  5. 5
    Real Madrid C. F.
    La Liga · 10 players
    3,912 min
  6. 6
    FC Barcelona
    La Liga · 12 players
    3,522 min
  7. 7
    Liverpool FC
    Premier League · 10 players
    3,488 min
  8. 8
    Aston Villa FC
    Premier League · 9 players
    2,758 min
  9. 9
    Crystal Palace FC
    Premier League · 11 players
    2,689 min
  10. 10
    PSV Eindhoven
    Eredivisie · 10 players
    2,575 min
  11. 11
    Manchester United FC
    Premier League · 9 players
    2,539 min
  12. 12
    Al Ahly FC
    Egyptian Premier League · 7 players
    2,441 min

Domestic share — squads who keep it at home

  1. ENGEngland
    78%
  2. EGYEgypt
    70%
  3. GERGermany
    67%
  4. RSASouth Africa
    67%
  5. KSASaudi Arabia
    63%
  6. MEXMexico
    47%
  7. ESPSpain
    42%
  8. QATQatar
    28%
  9. AUSAustralia
    22%
  10. FRAFrance
    17%
  11. SCOScotland
    16%
  12. BRABrazil
    15%

Minutes — not players — is the unit, because minutes is what wins matches. Club→league mapping covers 72% of squad-minutes; the rest sits in `Other`. Domestic-share treats a nation's own top-flight league as domestic via an exact league→nation map; partial substring matches are intentionally not used (they conflate North/South Korea and the two Congos). Club→league mapping coverage: 72%.

Module 04 — Knockout Simulator

10,000 brackets. A strength rating, not a literal forecast.

Per-team xG ratings blended with an elo prior, Poisson-sampled match outcomes across the real FIFA R16 bracket. Teams already eliminated in real knockouts are excluded.

Read this firstSimulator now runs the real FIFA R16 bracket. R32 results are already in the dataset; R16 through Final are Poisson-simulated 10,000 times from the actual pairings. Eliminated teams are excluded. Output is a strength rating, not a literal title probability.

Real R16 bracket — the path the simulator walks

8 R16 · 4 QF · 2 SF · 1 Final
R16-1
Paraguay
v
France
→ QF1
R16-2
Canada
v
Morocco
→ QF1
R16-3
Portugal
v
Spain
→ QF2
R16-4
USA
v
Belgium
→ QF2
R16-5
Brazil
v
Norway
→ QF3
R16-6
Mexico
v
England
→ QF3
R16-7
Argentina
v
Egypt
→ QF4
R16-8
Switzerland
v
Colombia
→ QF4
Simulations
10,000
As of
2026-07-12
Top model rating
ESP
31.3% strength share
Most likely final
Spain v Argent
32.1%

Model strength rating — top 12

From 10,000 sims
ESPSpain
31.3%
ARGArgentina
27.6%
FRAFrance
26.8%
ENGEngland
14.4%
Already eliminated · 28
BRAOUTPOROUTCOLOUTNEDOUTGEROUTBELOUTCROOUTMAROUTJPNOUTECUOUTSUIOUTUSAOUTSENOUTAUTOUTSWEOUTMEXOUTCANOUTAUSOUTNOROUTALGOUTCIVOUTEGYOUTPAROUTGHAOUTCODOUTBIHOUTCPVOUTRSAOUT

Most likely finals

  1. 1Spain v Argentina32.1%
  2. 2France v Argentina27.9%
  3. 3Spain v England20.5%
  4. 4France v England19.5%
What this model is missing

Per-team xG-for and xG-against from completed matches, blended with an elo-derived prior using confidence weighting. The elo→xG mapping (1.0 + (elo−1500)/600, clamped 0.5–2.2) is an uncalibrated heuristic, not fit to historical data. Match outcomes are Poisson-sampled with team-strength-adjusted expected goals; extra time uses a 30% scoring rate; penalties resolve with a coin flip. R32 results are deterministic (already played). R16 pairings and everything downstream follow the real FIFA draw defined in scripts/wc26/config/bracket.mjs — QF matchups are R16 slot-1-and-2, slot-3-and-4, and so on; SF and Final derive from those. The model ignores injuries, suspensions, travel, fatigue, home-crowd effects, goalkeeper identity, and provider bias. Treat outputs as a relative strength rating, not a literal title probability.

Module 05 — Forma's Pick

If the tournament ended today, our model says…

The payoff of the simulator: the highest-strength alive team, their expected path through the bracket, why the model likes them, and an honest note on how confident you should be.

🇪🇸
As it stands now, this is the winner

Spain

Strength rating 31.3% · elo 2120 (#2 in tournament) · CLEAR FAVOURITE
Path to the trophy — most likely opponents
R16
Portugal
Fixture confirmed
QF
Belgium
Most likely to advance
SF
France
Most likely to advance
Final
Argentina
Most likely finalist
Why the model picks Spain
  • Model strengthElo rating 2120, ranked #2 in the tournament of 48. xG-for 2.23 and xG-against 0.45 across the completed matches.
  • Bracket pathR16 opens against Portugal (elo 2010) — a real test but not the toughest possible R16 draw. Deeper rounds get harder — the model still gives them a 31.3% share out of 10,000 sims across the full bracket.
  • Field contextAhead of the runner-up Argentina (27.6%) by 3.7 points — a meaningful gap but not decisive.
This is a model output for analytical demonstration — not tipping advice. Group-stage xG signal is thin, penalties are essentially coin-flips in this model, and a single upset can flip the entire picture. Do not use this pick for betting or wagering of any kind.
Powered by Forma's analytical stack — the same layer we use for club-side decisions on recruitment, tactics, and value. The World Cup is a public showcase; the real work happens quietly with our club partners.
Module 06 — Daily Spotter

One card. Every match day. Through the final.

One number. One player. One tactical observation. Auto-surfaced from the data — nothing invented.

Forma Spotter · Sun 12 Jul
1 match this day
Number of the day
59%
Argentina dominated possession
ARG 59% — 41% SUI
Player of the day
Alexis Mac Allister
Argentina
1G · 0A · 120 min
Sat 11 Jul · 1 match
+1.0 goals vs xG
England overperformed their xG
Jude Victor William Bellingham · 2G 0A
Fri 10 Jul · 1 match
+0.6 goals vs xG
Belgium overperformed their xG
Fabian Ruiz · 1G 0A
Module 07 — Through a ClubOS lens

Same data. Product surface.

Pick any nation and see what their sporting director's ClubOS workspace would surface today — performance, recruitment, commercial, and financial cards, populated from the same live snapshot.

Through a ClubOS lens

What a Forma customer's workspace would surface for Spain today.

Performance
Possession control
Style archetype
xG / match2.23
xGA / match0.45
Possession64%
Sample6 matches
Same vector that powers the Identity Map — read the section above to see how this team sits among the 48.
Recruitment
No risers yet
From Hidden Value Index
No players from this squad have crossed the eligibility bar yet (180+ minutes).
ClubOS Recruitment runs the same engine on 30,000+ players year-round.
Commercial
31.3% strength share
Sponsor visibility forecast
Reach SF100.0%
Reach final52.6%
Lift TBCSponsor model
Drives sponsor-asset pricing and partner activation timing. Same lens we apply to club-season commercial dashboards.
Financial
6k squad minutes
Where the value is
Top leagueLa Liga
Share42%
Domestic42%
Plays into squad-valuation snapshots — concentration risk, league-tier exposure, succession depth.
This is the data through a product surface — not a dashboard demo, a workspace. The same flows run inside ClubOS year-round.
Open ClubOS →
Methodology · Footnote

What this hub is — and isn't.

Built on open WC26 data. Three group matches per team is a small sample — every output is directional, not predictive. None of this is a quoted valuation, transfer-fee forecast, or betting advice.

  • Identity Map — 6-axis style vector (PCA → k-means k=6). The dataset has no event coordinates; shot rate proxies verticality, foul rate proxies pressing.
  • Hidden Value Index — Tournament Contribution Score vs market-value rank, within position. 180+ tournament minutes required. Percentile delta is a directional signal.
  • Talent Pipeline — Minutes (not players) flow from club leagues to national squads. Domestic share uses an exact league→nation map.
  • Simulator — Confidence-weighted xG + elo prior; Poisson scoring; 10,000 bracket sims. Pairings use qualification-order seeding, not FIFA's draw. Eliminated teams are excluded.
  • Daily Spotter — Auto-surfaced extremes, rotated by category. Nothing invented.

Full methodology, formulas, and decisions: docs/plans/world-cup-2026-hub.md.

Open data — this much

Imagine what each of these becomes with full event & tracking feeds.

Identity Map — Real pressing intensity (PPDA from event coords), build-up zones, line-height by phase.
Hidden Value — Expected Threat per 90, on-ball value added, role-adjusted contribution, not just g+a.
Talent Pipeline — Full domestic-share by minutes weighted by competition level, not just league count.
Simulator — Injury, suspension, GK identity, fatigue, home-crowd, and the actual FIFA draw.
Spotter — Tactical observations from real event sequences (turnovers, high recoveries, build patterns).
Everything — Analyst-reviewed, not auto-generated. Findings that survive a real club's questions.

What you see here is Forma working with what's public. Here's what we do with what isn't.

Sources. Built from the Kaggle: mominullptr/fifa-world-cup-2026-dataset. Last refresh: 2026-07-12. Snapshot regenerated daily on every site rebuild.

Every model output on this page is presented as directional. None of it is a quoted projection, a recommendation, or a substitute for the kind of work Forma does for our clients — which is built on far richer data and analyst review.

Forma Football Analytics

We build analytics that survive a real club's questions.

This hub is a public snapshot of how we think. Our real work — for clubs, on full-season data, with analyst review — goes further than any open dataset can.

What we do →Get in touch