Deep insights

Model explainability and dataset analytics, computed live over the full World Cup 2026 database.

49k
International matches
901k
Player market valuations
143k
Player injury spells
27k
World Cup appearances
7k
ELO rating points
93k
Profiled players

150 years of international football

Era-level trends computed live over 49,256 international matches (1872–2026) and the full goalscorer record — scoring inflation, home advantage, goal timing and the most common scorelines in the sport’s history.

49,256
Matches analysed
1872–2026
144,802
Goals scored
4.5 → 2.7
Goals/game then → now
scoring has cooled
76-90'
Goals peak window
when nets bulge most
Goals per game, by decade

The defensive era: average goals per match has roughly halved since football’s wild early decades.

Home advantage & draw rate, by decade

Home-win % (non-neutral venues) has hovered near 50% for a century; the draw rate is the quiet constant.

When goals are scored

Across every recorded goal — teams score most in the final 15 minutes, when legs tire and games open up.

Most common scorelines in history

The tight 1–0 edges out the 1–1 and 0–0 — international football is a low-scoring game.

  • 1–0
    5,083
  • 1–1
    4,891
  • 0–0
    3,957
  • 2–0
    3,825
  • 2–1
    3,760
  • 0–1
    3,443
  • 1–2
    2,542
  • 3–0
    2,352

How the model decides — feature attribution

Each point is one of 72 fully-resolved fixtures; position is that feature’s signed contribution to the home-win logit, colour is the feature value (low → high). Bars show global mean |contribution|.

Global importance
Per-match attribution (beeswarm)
-2.20-1.100.00+1.10+2.20← lowers home-win logitraises home-win logit →Recent 10-match win rate2-year win rateHome-country advantageHead-to-head recordRecent 5-match formSquad market valueWorld Cup experienceInjury concerns (365d)ELO rating difference
feature valuelow
high

How a prediction is built — the causal chain

The full pipeline from raw data to outcome probabilities. Edge thickness is each feature’s real mean |contribution|; colour shows whether it tilts the result toward the home or away side.

Raw dataFeaturesWeighted logitSoftmaxPredictionELO ratingsRecent resultsSquad market valueInjury recordsHead-to-headHost / venueWorld Cup historyELO rating differenceWorld Cup experienceInjury concerns (365d)Recent 10-match win rateSquad market valueRecent 5-match form2-year win rateHome-country advantageHead-to-head recordΣ+ interceptσHome winDrawAway win
tilts home tilts away· edge thickness = mean |contribution|

Signal correlation — what overlaps

Pearson correlation between the model’s team-level inputs across all 48 nations. High values mean two signals carry overlapping information (multicollinearity) — useful for understanding what really drives a prediction.

ELO
Squad value
Form (10)
2yr win%
WC exp
Injuries
ELO
1.00
0.64
0.57
0.37
0.68
0.50
Squad value
0.64
1.00
0.53
0.38
0.60
0.52
Form (10)
0.57
0.53
1.00
0.69
0.36
0.40
2yr win%
0.37
0.38
0.69
1.00
0.19
0.25
WC exp
0.68
0.60
0.36
0.19
1.00
0.82
Injuries
0.50
0.52
0.40
0.25
0.82
1.00
positive negative· Pearson r across all 48 qualified nations

Predicted outcome mix

Across 72 matches. The draw share is structurally inflated — draw has a fixed logit of 0 while both win logits start at −0.35.

Home win37 · 51%
Draw9 · 13%
Away win26 · 36%

Win probability vs ELO edge

P(home win) against ELO difference; point size encodes prediction confidence, colour the predicted outcome.

Squad market value vs ELO rating

All 48 qualified nations; ordinary-least-squares fit on log₁₀(value). Colour by confederation.

R² = 0.41 · log₁₀(value) = 2.00e-3·ELO + 4.7

The Title Race, Visualized

Every contender as a bubble — ELO strength (x), model title probability (y), squad market value (size), confederation (colour). Spot favourites, value outliers and regional clusters at a glance.

Contenders across every axis

Each line is a team across the model's key metrics (title odds, ELO, form, win rate, squad value, World Cup pedigree). Hover to trace one team and spot trade-offs a single chart hides.

Hover a line to highlight a team
Title %25.4%1.3%ELO21711830Form PPM2.501.70Win% (10)80%50%Squad €€1210M€234MWC apps1148SpainArgentinaFranceEnglandNetherlandsPortugalCroatiaColombiaNorwayBrazilGermanySwitzerlandTurkeyMoroccoBelgiumJapan
UEFACONMEBOLCAFAFC

The field by confederation

Inner ring = confederations sized by qualified teams; outer ring = the teams, ordered by ELO. Hover any wedge for detail.

Hover a wedge — inner: confederation, outer: team
· 64UEFA · 16CAF · 10AFC · 9CONCACAF · 6CONMEBOL · 6OFC · 12A · ELO 15002B · ELO 15001E · ELO 15003A/B/C/D/F · ELO 15001F · ELO 15002C · ELO 15001C · ELO 15002F · ELO 15001I · ELO 15003C/D/F/G/H · ELO 15002E · ELO 15002I · ELO 15001A · ELO 15003C/E/F/H/I · ELO 15001L · ELO 15003E/H/I/J/K · ELO 15001D · ELO 15003B/E/F/I/J · ELO 15001G · ELO 15003A/E/H/I/J · ELO 15002K · ELO 15002L · ELO 15001H · ELO 15002J · ELO 15001B · ELO 15003E/F/G/I/J · ELO 15001J · ELO 15002H · ELO 15001K · ELO 15003D/E/I/J/L · ELO 15002D · ELO 15002G · ELO 1500W74 · ELO 1500W77 · ELO 1500W73 · ELO 1500W75 · ELO 1500W76 · ELO 1500W78 · ELO 1500W79 · ELO 1500W80 · ELO 1500W83 · ELO 1500W84 · ELO 1500W81 · ELO 1500W82 · ELO 1500W86 · ELO 1500W88 · ELO 1500W85 · ELO 1500W87 · ELO 1500W89 · ELO 1500W90 · ELO 1500W93 · ELO 1500W94 · ELO 1500W91 · ELO 1500W92 · ELO 1500W95 · ELO 1500W96 · ELO 1500W97 · ELO 1500W98 · ELO 1500W99 · ELO 1500W100 · ELO 1500L101 · ELO 1500L102 · ELO 1500W101 · ELO 1500W102 · ELO 1500Spain · ELO 2171France · ELO 2062England · ELO 2042Portugal · ELO 1976Netherlands · ELO 1959Croatia · ELO 1933Norway · ELO 1922Germany · ELO 1910Switzerland · ELO 1897Turkey · ELO 1880Belgium · ELO 1849Austria · ELO 1818Scotland · ELO 1790Czech Republic · ELO 1731Sweden · ELO 1660Bosnia & Herzegovina · ELO 1571Morocco · ELO 1830Senegal · ELO 1803Algeria · ELO 1726Tunisia · ELO 1641DR Congo · ELO 1616Ivory Coast · ELO 1607Egypt · ELO 1591Cape Verde · ELO 1560South Africa · ELO 1531Ghana · ELO 1509Japan · ELO 1878South Korea · ELO 1784Australia · ELO 1774Iran · ELO 1754Uzbekistan · ELO 1735Jordan · ELO 1687Saudi Arabia · ELO 1612Iraq · ELO 1582Qatar · ELO 1427Mexico · ELO 1835Canada · ELO 1802USA · ELO 1747Panama · ELO 1742Haiti · ELO 1542Curaçao · ELO 1467Argentina · ELO 2113Colombia · ELO 1998Brazil · ELO 1979Ecuador · ELO 1933Uruguay · ELO 1890Paraguay · ELO 1833New Zealand · ELO 1586112teams

Top-4 strength profiles

Five dimensions normalised 0–100 across all qualified nations.

ELO power ranking (top 16)

Radial bars sized by ELO rating, coloured by confederation.

Geographic distribution of qualified nations

All 48 nations at their real coordinates, sized by ELO and coloured by confederation — the global spread of strength heading into the tournament.

ESPSpain (UEFA) · ELO 2171ARGArgentina (CONMEBOL) · ELO 2113FRAFrance (UEFA) · ELO 2062ENGEngland (UEFA) · ELO 2042COLColombia (CONMEBOL) · ELO 1998BRABrazil (CONMEBOL) · ELO 1979PORPortugal (UEFA) · ELO 1976NEDNetherlands (UEFA) · ELO 1959ECUEcuador (CONMEBOL) · ELO 1933CROCroatia (UEFA) · ELO 1933NORNorway (UEFA) · ELO 1922GERGermany (UEFA) · ELO 1910SUISwitzerland (UEFA) · ELO 1897URUUruguay (CONMEBOL) · ELO 1890TURTurkey (UEFA) · ELO 1880JPNJapan (AFC) · ELO 1878BELBelgium (UEFA) · ELO 1849MEXMexico (CONCACAF) · ELO 1835PARParaguay (CONMEBOL) · ELO 1833MARMorocco (CAF) · ELO 1830AUTAustria (UEFA) · ELO 1818SENSenegal (CAF) · ELO 1803CANCanada (CONCACAF) · ELO 1802Scotland (UEFA) · ELO 1790South Korea (AFC) · ELO 1784Australia (AFC) · ELO 1774Iran (AFC) · ELO 1754USA (CONCACAF) · ELO 1747Panama (CONCACAF) · ELO 1742Uzbekistan (AFC) · ELO 1735Czech Republic (UEFA) · ELO 1731Algeria (CAF) · ELO 1726Jordan (AFC) · ELO 1687Sweden (UEFA) · ELO 1660Tunisia (CAF) · ELO 1641DR Congo (CAF) · ELO 1616Saudi Arabia (AFC) · ELO 1612Ivory Coast (CAF) · ELO 1607Egypt (CAF) · ELO 1591New Zealand (OFC) · ELO 1586Iraq (AFC) · ELO 1582Bosnia & Herzegovina (UEFA) · ELO 1571Cape Verde (CAF) · ELO 1560Haiti (CONCACAF) · ELO 1542South Africa (CAF) · ELO 1531Ghana (CAF) · ELO 1509Curaçao (CONCACAF) · ELO 1467Qatar (AFC) · ELO 1427
UEFA CONMEBOL CONCACAF CAF AFC OFC· bubble size = ELO

ELO distribution by confederation

Box = inter-quartile range with median; whiskers = min/max; dots = individual nations.

14271613179919852171CONMEBOLn=6UEFAn=16CONCACAFn=6AFCn=9CAFn=10OFCn=1

Model validation — backtested on 49,256 real international matches

ballsignal-v1 has never been scored against reality. Here it is, against every recorded international result (1872–2026).

Real World Cup draw rate
22.2%
across 964 historical WC matches — stable for decades
ballsignal-v1 mean draw prob
27.3%
+5.1 pts vs reality — draw logit fixed at 0 inflates even matchups
Home advantage (real)
+0.37
goals/match, home vs neutral venue (12,995 neutral games)

Calibration: model vs reality

ballsignal-v1 mean predicted probability vs the empirical World Cup base rate. Well-calibrated bars match.

Draw rate is stable — the model isn’t

Real draw rate by decade (≈22–23%, rock-steady) against ballsignal-v1’s mean predicted draw share.

Goal-margin distribution

How matches actually finish (49k results). Mean 2.94 goals/game (1.76 home / 1.18 away). The draw column is the only outcome the model over-weights.