Soccer predictions: actual data vs predicted data
Slate 20260919 · 9 match(es)
How to read this report. Each match shows the model's headline prediction, the stage-by-stage lambda (goal-rate) build, both teams' standing/form inputs, and every starter's projected line next to their real season baseline. A moderate / large deviation badge flags a projection that strays far from the player's season form — a quick way to spot when the inputs look right but a projection does not. When the real result is known an actual-vs-predicted banner appears.
All matches overview
Match detail
Athletic Club vs Alavés
LaLiga · 20260919 · 4-2-3-1 vs 3-5-2
Predicted 1-0Winner: Athletic ClubConfidence 0.57λ home/away 1.29 / 1.02Most-likely P 12.7%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Athletic Club vs Alavés. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 3-5-2 (duel lambdas: Home 1.287, Away 0.892). Final calibrated Poisson lambdas: Home 1.287, Away 1.020 (most likely score 1-0).
Home Starters 9/11 XGBoost (9 FBref) Away Starters 11/11 XGBoost (0 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.29 / 0.89 Final λ (H/A) 1.29 / 1.02
FBref Team Priors: Athletic Club (Gls/90: 1.0, GA/90: 1.0 → E[λ]: 1.237) vs Deportivo Alavés (Gls/90: 1.6666666666666667, GA/90: 1.67 → E[λ]: 1.235) | Blend Weight: 0.228
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Athletic Club | Alavés |
|---|
| League position | 10.00 | 2.00 |
| Points per game | 1.50 | 2.50 |
| Attack index | 1.25 | 2.08 |
| Defense (concede) index | 0.62 | 1.04 |
| Form multiplier | 0.95 | 1.05 |
| Standing λ multiplier | 0.90 | 1.39 |
Lambda decomposition (how the goal rates were built)
| Stage | Athletic Club λ | Alavés λ |
|---|
| FBref Squad Prior E[λ] (weight 0.228) | 1.237 | 1.235 |
| Pre-competitive-floor | 1.287 | 0.892 |
| Pre-calibration (Micro-duels) | 1.287 | 0.892 |
| Final λ | 1.287 | 1.020 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Falsecompetitive underdog floor applied Truestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 1-0 | 12.7% |
| 1-1 | 13.2% |
| 0-0 | 10.1% |
| 0-1 | 10.0% |
| 2-1 | 8.4% |
| Home win | 42.4% |
| Draw | 28.2% |
| Away win | 29.3% |
| Over 2.5 | 40.6% |
| BTTS yes | 46.4% |
Pregame fair line — favorite Athletic Club (42.4%), fair total 2.31, fair spread 0.27
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Athletic Club — projected line vs season baseline (4-2-3-1)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Unai Simón | G | 94 | 0 | 0.00 | - | n/a | 0 | 0.0 | 8.0 | 60.2 | 5 | FBref Prior G_service |
| Dani Vivian | D | 90 | 0 | 0.03 | - | n/a | 0 | - | 7.1 | - | 0 | Prior D_service |
| Aymeric Laporte | D | 76 | 0 | 0.00 | 0.09 | 100% | 1 | 0.4 | 6.8 | 54.6 | 5 | FBref Prior D_service |
| Aitor Paredes | D | 69 | 0 | 0.00 | 0.07 | 100% | 0 | 0.2 | 6.8 | 54.8 | 4 | FBref Prior D_service |
| Jesús Areso | D | 77 | 0 | 0.00 | 0.01 | 100% | 0 | 0.2 | 6.6 | 39.8 | 5 | FBref Prior D_service |
| Nico Williams | M | 50 | 0 | 0.22 | 0.18 | ok | 2 | 1.2 | 7.5 | 54.8 | 5 | FBref Prior M_service |
| Mikel Jauregizar | M | 44 | 0 | 0.05 | 0.01 | 523% | 1 | 0.2 | 7.2 | 51.0 | 4 | FBref Prior M_service |
| Oihan Sancet | M | 88 | 0 | 0.41 | 0.31 | 33% | 2 | 2.2 | 7.4 | 55.0 | 5 | FBref Prior M_service |
| Beñat Prados | M | 90 | 0 | 0.17 | - | n/a | 2 | 0.0 | 7.1 | - | 1 | Prior M_service |
| Peio Canales | M | 69 | 0 | 0.10 | 0.05 | 87% | 2 | 1.0 | 7.0 | 65.7 | 3 | FBref Prior M_service |
| Gorka Guruzeta | F | 44 | 1 | 0.19 | 0.01 | 1557% | 2 | 0.5 | 6.7 | 64.5 | 4 | FBref Prior F_service |
| Team totals | 1 | 1.29 | | 12 | | 7.1 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Alex Padilla | G | 4.4m | 10.9% | Market-value substitute prior |
| Hugo Rincón | D | 15.5m | 41.0% | Market-value substitute prior |
| Adama Boiro | D | 17.7m | 60.2% | Market-value substitute prior |
| Iñaki Williams | M | 22.2m | 31.1% | Market-value substitute prior |
| Robert Navarro | M | 22.1m | 50.6% | Market-value substitute prior |
| Álvaro Djaló | F | 16.9m | 72.0% | Market-value substitute prior |
Alavés — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Antonio Sivera | G | 89 | 0 | 0.00 | - | n/a | | 0.0 | 7.1 | 56.5 | 4 | XGBoost G_service |
| Youssef Enriquez | D | 25 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.7 | 70.7 | 3 | XGBoost D_service |
| Nahuel Tenaglia | D | 88 | 0 | 0.00 | 0.02 | 100% | 1 | 0.8 | 6.7 | 49.8 | 4 | XGBoost D_service |
| Nicolás Valentini | D | 10 | 0 | 0.02 | 0.01 | 143% | 1 | 0.3 | 6.3 | 22.0 | 3 | XGBoost D_service |
| Antonio Blanco | M | 91 | 0 | 0.06 | 0.03 | 110% | 1 | 0.8 | 6.9 | 53.0 | 4 | XGBoost M_service |
| Miguel Rodríguez | M | 12 | 0 | 0.07 | - | n/a | 1 | 0.0 | 6.6 | 21.3 | 3 | XGBoost M_service |
| Pablo Ibáñez | M | 89 | 0 | 0.11 | 0.09 | ok | 1 | 1.5 | 6.8 | 54.5 | 4 | XGBoost M_service |
| Carles Aleñá | M | 40 | 0 | 0.04 | 0.02 | 83% | | 0.2 | 6.7 | 33.0 | 4 | XGBoost M_service |
| Ángel Pérez | M | 79 | 0 | 0.15 | 0.09 | 77% | 1 | 1.0 | 6.8 | 57.2 | 4 | XGBoost M_service |
| Toni Martínez | F | 72 | 0 | 0.15 | 0.69 | 78% | 2 | 2.0 | 6.7 | 63.0 | 2 | XGBoost F_service |
| Lucas Boyé | F | 76 | 0 | 0.34 | 0.22 | 53% | | 2.0 | 7.6 | 60.2 | 4 | XGBoost F_service |
| Team totals | 0 | 1.02 | | 8 | | 6.8 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Adrián Rodríguez | G | 4.4m | 10.9% | Market-value substitute prior |
| Jonny Otto | D | 13.6m | 29.9% | Market-value substitute prior |
| Ville Koski | D | 13.6m | 29.9% | Market-value substitute prior |
| Ander Guevara | M | 22.0m | 49.9% | Market-value substitute prior |
| Abderrahman Rebbach | M | 22.0m | 49.9% | Market-value substitute prior |
| Aitor Mañas | F | 25.3m | 55.9% | Market-value substitute prior |
Antagonist / tactical summary
Home midfield dominates with ~53% possession expected (Nico Williams vs Antonio Blanco). Total antagonist xG — Home: 0.67, Away: 0.48. Home danger zone: Left Wing vs Right Wingback — Compact Flank. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Midfield — possession 52.7% / 47.3%, dominance ratio 0.58, key duel: Nico Williams vs Antonio Blanco
Oihan Sancet xG 0.31Peio Canales xG 0.07Nico Williams xG 0.14Toni Martínez xG 0.15Miguel Rodríguez xG 0.06Ángel Pérez xG 0.13
Summary
Antagonist matchup analysis for Athletic Club vs Alavés. Most likely scoreline: 1-0. Lambdas derived from 5 home and 5 away individual player duels.
Tactical parity between 4-2-3-1 and 3-5-2. Match outcome driven primarily by individual player quality. Home midfield dominates with ~53% possession expected (Nico Williams vs Antonio Blanco). Total antagonist xG — Home: 0.67, Away: 0.48. Home danger zone: Left Wing vs Right Wingback — Compact Flank. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Brighton vs Arsenal
Premier League · 20260919 · 4-2-3-1 vs 3-5-2
Predicted 1-2Winner: ArsenalConfidence 0.6λ home/away 1.30 / 2.12Most-likely P 9.6%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Brighton vs Arsenal. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 3-5-2 (duel lambdas: Home 1.331, Away 1.939). Final calibrated Poisson lambdas: Home 1.297, Away 2.122 (most likely score 1-2).
Home Starters 10/11 XGBoost (0 FBref) Away Starters 9/11 XGBoost (9 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.33 / 1.94 Final λ (H/A) 1.30 / 2.12
FBref Team Priors: Brighton & Hove Albion (Gls/90: 2.5, GA/90: 3.5 → E[λ]: 3.5) vs Arsenal (Gls/90: 2.0, GA/90: 2.0 → E[λ]: 3.5) | Blend Weight: 0.192
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Brighton | Arsenal |
|---|
| League position | 10.00 | 2.00 |
| Points per game | 1.33 | 3.00 |
| Attack index | 3.33 | 1.25 |
| Defense (concede) index | 0.21 | 2.08 |
| Form multiplier | 0.94 | 1.06 |
| Standing λ multiplier | 0.86 | 1.50 |
Lambda decomposition (how the goal rates were built)
| Stage | Brighton λ | Arsenal λ |
|---|
| FBref Squad Prior E[λ] (weight 0.192) | 3.500 | 3.500 |
| Pre-mismatch | 1.331 | 1.939 |
| Pre-competitive-floor | 1.297 | 2.122 |
| Pre-calibration (Micro-duels) | 1.331 | 1.939 |
| Final λ | 1.297 | 2.122 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Truecompetitive underdog floor applied Falsestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 1-2 | 9.6% |
| 1-1 | 9.1% |
| 0-2 | 7.4% |
| 0-1 | 6.9% |
| 1-3 | 6.8% |
| Home win | 22.9% |
| Draw | 20.9% |
| Away win | 56.3% |
| Over 2.5 | 66.4% |
| BTTS yes | 64.0% |
Pregame fair line — favorite Arsenal (56.3%), fair total 3.42, fair spread -0.82
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Brighton — projected line vs season baseline (4-2-3-1)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Bart Verbruggen | G | 95 | 0 | 0.00 | - | n/a | | 0.0 | 7.3 | 68.0 | 3 | XGBoost G_service |
| Luka Vušković | D | 95 | 0 | 0.14 | 0.12 | ok | 2 | 2.0 | 7.6 | 75.0 | 3 | XGBoost D_service |
| Ferdi Kadıoğlu | D | 91 | 0 | 0.05 | 0.04 | ok | 2 | 1.5 | 7.1 | 60.0 | 2 | XGBoost D_service |
| Mats Wieffer | D | 83 | 0 | 0.07 | 0.06 | ok | 1 | 1.0 | 7.1 | 73.0 | 2 | XGBoost D_service |
| Pascal Struijk | D | 1 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.2 | - | 3 | XGBoost D_service |
| Yankuba Minteh | M | 90 | 0 | 0.14 | - | n/a | 2 | - | 7.4 | - | 0 | Prior M_service |
| Yasin Ayari | M | 83 | 0 | 0.07 | 0.14 | 51% | 2 | 2.0 | 7.4 | 66.0 | 2 | XGBoost M_service |
| Georginio Rutter | M | 95 | 0 | 0.13 | 0.08 | 69% | 2 | 2.0 | 7.3 | 76.0 | 1 | XGBoost M_service |
| Jack Hinshelwood | M | 68 | 1 | 0.10 | 1.41 | 93% | 2 | 3.0 | 7.1 | 90.0 | 1 | XGBoost M_service |
| Diego Gómez | M | 94 | 0 | 0.22 | 0.22 | ok | | 4.0 | 7.3 | 69.7 | 3 | XGBoost M_service |
| Charalampos Kostoulas | F | 69 | 0 | 0.15 | 0.26 | 42% | 2 | 2.0 | 7.2 | 65.7 | 3 | XGBoost F_service |
| Team totals | 1 | 1.30 | | 17 | | 7.2 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Tom McGill | G | 4.4m | 10.9% | Market-value substitute prior |
| Maxim De Cuyper | D | 14.4m | 33.5% | Market-value substitute prior |
| Olivier Boscagli | D | 14.1m | 32.2% | Market-value substitute prior |
| Kaoru Mitoma | M | 22.7m | 53.4% | Market-value substitute prior |
| Chema Andrés | M | 22.7m | 53.3% | Market-value substitute prior |
| Evan Ferguson | F | 26.1m | 59.5% | Market-value substitute prior |
Arsenal — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| David Raya | G | 95 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.8 | 40.8 | 5 | FBref Prior G_service |
| William Saliba | D | 90 | 0 | 0.05 | - | n/a | 1 | - | 7.4 | - | 1 | Prior D_service |
| Jurriën Timber | D | 90 | 0 | 0.05 | - | n/a | 1 | - | 7.4 | - | 1 | Prior D_service |
| Gabriel Magalhães | D | 95 | 0 | 0.31 | 0.10 | 215% | 1 | 0.8 | 7.4 | 41.0 | 5 | FBref Prior D_service |
| Declan Rice | M | 91 | 0 | 0.15 | 0.02 | 632% | 1 | 0.6 | 7.9 | 44.8 | 5 | FBref Prior M_service |
| Martín Zubimendi | M | 22 | 0 | 0.07 | 0.02 | 285% | 0 | 0.2 | 7.6 | 49.2 | 4 | FBref Prior M_service |
| Martin Ødegaard | M | 83 | 0 | 0.36 | 0.20 | 79% | 2 | 1.2 | 7.8 | 49.0 | 5 | FBref Prior M_service |
| Bruno Guimarães | M | 21 | 0 | 0.10 | - | n/a | 1 | 0.0 | 7.6 | 21.7 | 3 | FBref Prior M_service |
| Eberechi Eze | M | 33 | 0 | 0.10 | 0.01 | 994% | 1 | 0.2 | 7.6 | 32.8 | 4 | FBref Prior M_service |
| Bukayo Saka | F | 92 | 2 | 0.53 | 0.25 | 113% | 2 | 1.5 | 8.3 | 37.3 | 6 | FBref Prior F_service |
| Viktor Gyökeres | F | 13 | 0 | 0.31 | - | n/a | | 0.0 | 7.0 | 16.0 | 4 | FBref Prior F_service |
| Team totals | 2 | 2.12 | | 10 | | 7.6 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Illan Meslier | G | 4.6m | 11.9% | Market-value substitute prior |
| Riccardo Calafiori | D | 15.1m | 24.9% | Market-value substitute prior |
| Myles Lewis-Skelly | D | 15.1m | 24.8% | Market-value substitute prior |
| Christos Tzolis | M | 23.4m | 33.6% | Market-value substitute prior |
| Ethan Nwaneri | M | 23.2m | 55.4% | Market-value substitute prior |
| Noni Madueke | F | 24.7m | 75.1% | Market-value substitute prior |
Antagonist / tactical summary
Midfield is evenly contested — Yankuba Minteh vs Declan Rice will be the pivotal duel. Total antagonist xG — Home: 0.51, Away: 0.76. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Midfield — possession 48.9% / 51.1%, dominance ratio 0.46, key duel: Yankuba Minteh vs Declan Rice
Yankuba Minteh xG 0.13Yasin Ayari xG 0.06Jack Hinshelwood xG 0.06Bukayo Saka xG 0.31Bruno Guimarães xG 0.06Eberechi Eze xG 0.06
Summary
Antagonist matchup analysis for Brighton vs Arsenal. Most likely scoreline: 1-2. Lambdas derived from 5 home and 5 away individual player duels.
Tactical parity between 4-2-3-1 and 3-5-2. Match outcome driven primarily by individual player quality. Midfield is evenly contested — Yankuba Minteh vs Declan Rice will be the pivotal duel. Total antagonist xG — Home: 0.51, Away: 0.76. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Tottenham vs Aston Villa
Premier League · 20260919 · 3-5-2 vs 3-4-2-1
Predicted 0-1Winner: Aston VillaConfidence 0.56λ home/away 1.19 / 1.46Most-likely P 10.2%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Tottenham vs Aston Villa. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 3-5-2 vs 3-4-2-1 (duel lambdas: Home 1.193, Away 1.464). Final calibrated Poisson lambdas: Home 1.193, Away 1.464 (most likely score 0-1).
Home Starters 10/11 XGBoost (0 FBref) Away Starters 7/11 XGBoost (7 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.19 / 1.46 Final λ (H/A) 1.19 / 1.46
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Tottenham | Aston Villa |
|---|
| League position | 17.00 | 18.00 |
| Points per game | 0.33 | 0.33 |
| Attack index | 1.00 | 1.00 |
| Defense (concede) index | 1.04 | 2.08 |
| Form multiplier | 1.00 | 1.00 |
| Standing λ multiplier | 1.02 | 1.35 |
Lambda decomposition (how the goal rates were built)
| Stage | Tottenham λ | Aston Villa λ |
|---|
| Pre-competitive-floor | 1.193 | 1.464 |
| Pre-calibration (Micro-duels) | 1.193 | 1.464 |
| Final λ | 1.193 | 1.464 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Falsecompetitive underdog floor applied Falsestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 0-1 | 10.2% |
| 1-1 | 12.4% |
| 1-2 | 9.0% |
| 1-0 | 8.3% |
| 0-2 | 7.5% |
| Home win | 30.7% |
| Draw | 26.0% |
| Away win | 43.3% |
| Over 2.5 | 49.6% |
| BTTS yes | 53.7% |
Pregame fair line — favorite Aston Villa (43.3%), fair total 2.66, fair spread -0.27
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Tottenham — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Antonín Kinský | G | 92 | 0 | 0.00 | - | n/a | | 0.0 | 7.2 | 48.0 | 4 | XGBoost G_service |
| Jan Paul van Hecke | D | 95 | 0 | 0.00 | 0.05 | 100% | 1 | 0.8 | 7.7 | 55.2 | 4 | XGBoost D_service |
| Micky van de Ven | D | 95 | 0 | 0.00 | 0.19 | 100% | 1 | 0.7 | 7.4 | 47.0 | 3 | XGBoost D_service |
| Pedro Porro | D | 95 | 0 | 0.20 | 0.12 | 72% | 3 | 2.0 | 7.2 | 44.7 | 3 | XGBoost D_service |
| Sandro Tonali | M | 95 | 0 | 0.08 | 0.02 | 266% | 2 | 0.8 | 7.7 | 51.8 | 4 | XGBoost M_service |
| Mohammed Kudus | M | 28 | 0 | 0.08 | 0.02 | 294% | 2 | 0.5 | 7.6 | 67.5 | 2 | XGBoost M_service |
| Mateus Fernandes | M | 54 | 0 | 0.05 | 0.00 | 1746% | 1 | 0.3 | 7.6 | 68.3 | 3 | XGBoost M_service |
| Xavi Simons | M | 90 | 0 | 0.14 | - | n/a | 2 | - | 7.4 | - | 0 | Prior M_service |
| Mykhailo Mudryk | M | 17 | 0 | 0.08 | - | n/a | 1 | 0.0 | 7.4 | 67.0 | 1 | XGBoost M_service |
| Omar Marmoush | F | 95 | 0 | 0.26 | 0.08 | 208% | 3 | 1.2 | 7.4 | 34.8 | 4 | XGBoost F_service |
| Dominic Solanke | F | 17 | 0 | 0.12 | 0.15 | ok | 2 | 1.0 | 6.7 | 63.7 | 3 | XGBoost F_service |
| Team totals | 0 | 1.19 | | 18 | | 7.4 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Brandon Austin | G | 4.4m | 10.9% | Market-value substitute prior |
| Destiny Udogie | D | 14.6m | 34.6% | Market-value substitute prior |
| Marcos Senesi | D | 14.4m | 33.9% | Market-value substitute prior |
| Sávio | M | 23.3m | 55.8% | Market-value substitute prior |
| Lucas Bergvall | M | 23.2m | 55.6% | Market-value substitute prior |
| Richarlison | F | 26.2m | 60.0% | Market-value substitute prior |
Aston Villa — projected line vs season baseline (3-4-2-1)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Zion Suzuki | G | 95 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.8 | 35.8 | 4 | FBref Prior G_service |
| Ian Maatsen | D | 87 | 0 | 0.00 | 0.02 | 100% | 1 | 0.8 | 7.1 | 46.0 | 4 | FBref Prior D_service |
| Matteo Ruggeri | D | 12 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.7 | 45.7 | 3 | FBref Prior D_service |
| Taylor Harwood-Bellis | D | 90 | 0 | 0.03 | - | n/a | 0 | - | 7.1 | - | 0 | Prior D_service |
| Johan Manzambi | M | 90 | 0 | 0.18 | - | n/a | 2 | - | 7.5 | - | 0 | Prior M_service |
| Amadou Onana | M | 90 | 0 | 0.17 | - | n/a | 2 | - | 7.4 | - | 0 | Prior M_service |
| Boubacar Kamara | M | 95 | 0 | 0.07 | - | n/a | 1 | 0.0 | 7.5 | 50.0 | 4 | FBref Prior M_service |
| João Gomes | M | 43 | 0 | 0.06 | - | n/a | 1 | 0.0 | 7.2 | 50.0 | 1 | FBref Prior M_service |
| Nicolas Jackson | F | 84 | 0 | 0.26 | 0.08 | 233% | 2 | 0.7 | 7.1 | 45.3 | 3 | FBref Prior F_service |
| Ibrahim Mbaye | F | 90 | 1 | 0.36 | - | n/a | 3 | 0.0 | 7.3 | - | 1 | Prior F_service |
| Alejandro Garnacho | F | 13 | 0 | 0.18 | - | n/a | 2 | 0.0 | 7.0 | 64.0 | 3 | FBref Prior F_service |
| Team totals | 1 | 1.46 | | 15 | | 7.2 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Marco Bizot | G | 4.4m | 13.1% | Market-value substitute prior |
| Pau Torres | D | 14.4m | 23.3% | Market-value substitute prior |
| Matty Cash | D | 14.3m | 23.2% | Market-value substitute prior |
| Lamare Bogarde | M | 19.5m | 70.6% | Market-value substitute prior |
| Emiliano Buendía | M | 22.5m | 52.3% | Market-value substitute prior |
| Tammy Abraham | F | 22.7m | 73.2% | Market-value substitute prior |
Antagonist / tactical summary
Midfield is evenly contested — Sandro Tonali vs Johan Manzambi will be the pivotal duel. Total antagonist xG — Home: 0.63, Away: 1.04. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Shadow Striker vs Centre Back — Half-Space Threat.
Midfield — possession 50.0% / 50.0%, dominance ratio 0.52, key duel: Sandro Tonali vs Johan Manzambi
Mateus Fernandes xG 0.05Mohammed Kudus xG 0.07Sandro Tonali xG 0.08Boubacar Kamara xG 0.06Johan Manzambi xG 0.16Amadou Onana xG 0.15
Summary
Antagonist matchup analysis for Tottenham vs Aston Villa. Most likely scoreline: 0-1. Lambdas derived from 6 home and 7 away individual player duels.
Tactical parity between 3-5-2 and 3-4-2-1. Match outcome driven primarily by individual player quality. Midfield is evenly contested — Sandro Tonali vs Johan Manzambi will be the pivotal duel. Total antagonist xG — Home: 0.63, Away: 1.04. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Shadow Striker vs Centre Back — Half-Space Threat.
Sevilla vs Barcelona
LaLiga · 20260919 · 5-4-1 vs 3-5-2
Predicted 1-3Winner: BarcelonaConfidence 0.69λ home/away 1.27 / 3.69Most-likely P 7.5%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Sevilla vs Barcelona. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 5-4-1 vs 3-5-2 (duel lambdas: Home 1.354, Away 3.135). Final calibrated Poisson lambdas: Home 1.275, Away 3.689 (most likely score 1-3).
Home Starters 10/11 XGBoost (6 FBref) Away Starters 10/11 XGBoost (0 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.35 / 3.13 Final λ (H/A) 1.27 / 3.69
FBref Team Priors: Sevilla (Gls/90: 2.0, GA/90: 2.0 → E[λ]: 3.5) vs FC Barcelona (Gls/90: 3.6666666666666665, GA/90: 4.0 → E[λ]: 3.5) | Blend Weight: 0.228
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Sevilla | Barcelona |
|---|
| League position | 7.00 | 1.00 |
| Points per game | 1.75 | 3.00 |
| Attack index | 1.46 | 3.54 |
| Defense (concede) index | 0.42 | 1.25 |
| Form multiplier | 0.95 | 1.05 |
| Standing λ multiplier | 0.81 | 1.65 |
Lambda decomposition (how the goal rates were built)
| Stage | Sevilla λ | Barcelona λ |
|---|
| FBref Squad Prior E[λ] (weight 0.228) | 3.500 | 3.500 |
| Pre-mismatch | 1.354 | 3.135 |
| Pre-calibration (Micro-duels) | 1.354 | 3.135 |
| Final λ | 1.275 | 3.689 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Truecompetitive underdog floor applied Falsestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 1-3 | 7.5% |
| 1-4 | 6.9% |
| 1-2 | 6.1% |
| 0-3 | 5.9% |
| 0-4 | 5.4% |
| Home win | 8.6% |
| Draw | 10.7% |
| Away win | 80.7% |
| Over 2.5 | 87.2% |
| BTTS yes | 70.3% |
Pregame fair line — favorite Barcelona (80.7%), fair total 4.96, fair spread -2.41
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Sevilla — projected line vs season baseline (5-4-1)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Odysseas Vlachodimos | G | 88 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.8 | 54.8 | 5 | FBref Prior G_service |
| Kike Salas | D | 92 | 0 | 0.00 | 0.01 | 100% | 1 | 0.8 | 6.8 | 50.0 | 4 | FBref Prior D_service |
| José Ángel Carmona | D | 21 | 0 | 0.00 | - | n/a | 0 | 0.0 | 5.6 | 33.2 | 4 | XGBoost D_service |
| Arouna Sangante | D | 71 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.6 | 47.8 | 4 | FBref Prior D_service |
| Gabriel Suazo | D | 90 | 0 | 0.00 | 0.00 | 100% | 0 | 0.2 | 6.5 | 53.8 | 5 | FBref Prior D_service |
| Andrés Castrin | D | 70 | 0 | 0.00 | 0.00 | 100% | 0 | 0.2 | 6.8 | 40.0 | 5 | XGBoost D_service |
| Youssouf Fofana | M | 38 | 0 | 0.14 | 0.00 | 3329% | 1 | 0.5 | 7.2 | 31.0 | 2 | XGBoost M_service |
| Lucien Agoumé | M | 92 | 0 | 0.12 | 0.01 | 1508% | 1 | 0.4 | 7.1 | 56.0 | 5 | FBref Prior M_service |
| Rubén Vargas | M | 20 | 0 | 0.17 | 0.07 | 135% | 4 | 1.0 | 6.9 | 65.0 | 1 | FBref Prior M_service |
| Félix Correia | M | 90 | 0 | 0.22 | - | n/a | 1 | 0.0 | 7.0 | - | 1 | Prior M_service |
| Lucas Stassin | F | 37 | 1 | 0.34 | 0.20 | 71% | 2 | 1.0 | 7.3 | 37.0 | 2 | XGBoost F_service |
| Team totals | 1 | 1.27 | | 14 | | 6.8 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Fran González | G | 4.4m | 10.9% | Market-value substitute prior |
| Juan Iglesias | D | 13.7m | 30.3% | Market-value substitute prior |
| Marcão | D | 13.6m | 29.9% | Market-value substitute prior |
| Oso | M | 22.3m | 51.5% | Market-value substitute prior |
| Giorgi Kochorashvili | M | 29.7m | 69.3% | Market-value substitute prior |
| Isaac Romero | F | 30.1m | 59.3% | Market-value substitute prior |
Barcelona — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Joan García | G | 95 | 0 | 0.00 | - | n/a | | 0.0 | 7.7 | 72.5 | 4 | XGBoost G_service |
| Pau Cubarsí | D | 90 | 0 | 0.00 | 0.11 | 100% | 1 | 0.5 | 7.8 | 57.5 | 4 | XGBoost D_service |
| Jules Koundé | D | 47 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.4 | 68.0 | 4 | XGBoost D_service |
| Alejandro Balde | D | 90 | 0 | 0.07 | - | n/a | 1 | 0.0 | 7.4 | - | 3 | Prior D_service |
| Lamine Yamal | M | 94 | 0 | 0.66 | 0.64 | ok | 2 | 5.8 | 7.8 | 83.0 | 4 | XGBoost M_service |
| Pedri | M | 81 | 0 | 0.32 | 0.09 | 239% | 2 | 1.0 | 8.0 | 74.0 | 4 | XGBoost M_service |
| Fermín López | M | 79 | 0 | 0.55 | 0.68 | ok | | 2.5 | 7.8 | 80.8 | 4 | XGBoost M_service |
| Raphinha | M | 88 | 1 | 0.43 | 1.08 | 61% | 2 | 2.6 | 7.7 | 68.6 | 5 | XGBoost M_service |
| Dani Olmo | M | 36 | 0 | 0.17 | 0.10 | 67% | 1 | 1.5 | 7.7 | 72.0 | 4 | XGBoost M_service |
| Anthony Gordon | F | 68 | 1 | 0.82 | 0.30 | 170% | 3 | 2.2 | 7.6 | 71.8 | 4 | XGBoost F_service |
| Karim Adeyemi | F | 31 | 1 | 0.37 | 0.35 | ok | 2 | 0.8 | 7.4 | 68.8 | 4 | XGBoost F_service |
| Team totals | 3 | 3.69 | | 15 | | 7.7 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Dominik Livaković | G | 4.5m | 11.1% | Market-value substitute prior |
| Eric García | D | 14.9m | 36.0% | Market-value substitute prior |
| Gerard Martín | D | 14.7m | 35.2% | Market-value substitute prior |
| Rodri | M | 23.5m | 56.8% | Market-value substitute prior |
| Frenkie de Jong | M | 23.1m | 55.2% | Market-value substitute prior |
| Gabriel Jesus | F | 26.0m | 58.9% | Market-value substitute prior |
Antagonist / tactical summary
Away midfield dictates play with ~53% possession expected (Youssouf Fofana vs Lamine Yamal). Total antagonist xG — Home: 0.33, Away: 1.13. Home danger zone: Striker vs Three Centre Backs — Overloaded Defensive Block. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Midfield — possession 47.4% / 52.6%, dominance ratio 0.40, key duel: Youssouf Fofana vs Lamine Yamal
Lucas Stassin xG 0.14Kike Salas xG 0.08Gabriel Suazo xG 0.06Raphinha xG 0.20Lamine Yamal xG 0.28Pedri xG 0.13
Summary
Antagonist matchup analysis for Sevilla vs Barcelona. Most likely scoreline: 1-3. Lambdas derived from 4 home and 6 away individual player duels.
Tactical parity between 5-4-1 and 3-5-2. Match outcome driven primarily by individual player quality. Away midfield dictates play with ~53% possession expected (Youssouf Fofana vs Lamine Yamal). Total antagonist xG — Home: 0.33, Away: 1.13. Home danger zone: Striker vs Three Centre Backs — Overloaded Defensive Block. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Forest vs Coventry
Premier League · 20260919 · 4-3-3 vs 3-4-2-1
Predicted 1-0Winner: ForestConfidence 0.58λ home/away 1.44 / 1.02Most-likely P 12.2%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Forest vs Coventry. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-3-3 vs 3-4-2-1 (duel lambdas: Home 1.444, Away 0.875). Final calibrated Poisson lambdas: Home 1.444, Away 1.020 (most likely score 1-0).
Home Starters 10/11 XGBoost (0 FBref) Away Starters 8/11 XGBoost (7 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.44 / 0.88 Final λ (H/A) 1.44 / 1.02
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Forest | Coventry |
|---|
| League position | 16.00 | 20.00 |
| Points per game | 0.67 | 0.00 |
| Attack index | 0.83 | 1.00 |
| Defense (concede) index | 1.04 | 1.25 |
| Form multiplier | 1.03 | 0.96 |
| Standing λ multiplier | 0.95 | 1.09 |
Lambda decomposition (how the goal rates were built)
| Stage | Forest λ | Coventry λ |
|---|
| Pre-competitive-floor | 1.444 | 0.875 |
| Pre-calibration (Micro-duels) | 1.444 | 0.875 |
| Final λ | 1.444 | 1.020 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Falsecompetitive underdog floor applied Truestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 1-0 | 12.2% |
| 1-1 | 12.6% |
| 2-1 | 9.0% |
| 2-0 | 8.9% |
| 0-0 | 8.6% |
| Home win | 46.7% |
| Draw | 26.7% |
| Away win | 26.6% |
| Over 2.5 | 44.7% |
| BTTS yes | 49.0% |
Pregame fair line — favorite Forest (46.7%), fair total 2.46, fair spread 0.42
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Forest — projected line vs season baseline (4-3-3)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| John Victor | G | 90 | 0 | 0.00 | - | n/a | | 0.0 | 6.8 | - | 3 | Prior G_service |
| Murillo | D | 95 | 0 | 0.00 | 0.09 | 100% | 1 | 0.8 | 7.3 | 51.0 | 4 | XGBoost D_service |
| Ousmane Diomande | D | 1 | 0 | 0.00 | - | n/a | 0 | 0.0 | 5.6 | - | 3 | XGBoost D_service |
| Nikola Milenković | D | 95 | 0 | 0.00 | 0.03 | 100% | 1 | 0.5 | 7.2 | 50.5 | 4 | XGBoost D_service |
| Neco Williams | D | 95 | 0 | 0.00 | 0.03 | 100% | 0 | 0.2 | 6.9 | 55.5 | 4 | XGBoost D_service |
| Morgan Gibbs-White | M | 95 | 0 | 0.27 | 0.35 | ok | 2 | 1.8 | 7.7 | 53.0 | 4 | XGBoost M_service |
| Callum Hudson Odoi | M | 1 | 0 | 0.10 | - | n/a | 2 | 0.0 | 7.2 | - | 3 | XGBoost M_service |
| Dan Ndoye | M | 67 | 0 | 0.12 | 0.09 | 35% | | 0.8 | 7.3 | 51.2 | 4 | XGBoost M_service |
| Liam Delap | F | 52 | 0 | 0.22 | 0.04 | 393% | 2 | 0.7 | 7.2 | 43.3 | 3 | XGBoost F_service |
| Igor Jesus | F | 74 | 1 | 0.42 | 0.32 | 31% | 2 | 2.0 | 7.1 | 66.7 | 3 | XGBoost F_service |
| Arnaud Kalimuendo | F | 1 | 0 | 0.25 | - | n/a | 2 | 0.0 | 6.7 | - | 1 | XGBoost F_service |
| Team totals | 1 | 1.44 | | 12 | | 7.0 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Matz Sels | G | 4.4m | 11.0% | Market-value substitute prior |
| Ola Aina | D | 14.2m | 32.7% | Market-value substitute prior |
| Jair | D | 14.2m | 32.5% | Market-value substitute prior |
| Daniel Muñoz | M | 22.9m | 53.9% | Market-value substitute prior |
| Ibrahim Sangaré | M | 22.8m | 53.5% | Market-value substitute prior |
| Chris Wood | F | 25.4m | 56.4% | Market-value substitute prior |
Coventry — projected line vs season baseline (3-4-2-1)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Carl Rushworth | G | 92 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.4 | 51.5 | 4 | FBref Prior G_service |
| Milan van Ewijk | D | 94 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.0 | 47.2 | 4 | FBref Prior D_service |
| Bobby Thomas | D | 88 | 0 | 0.00 | 0.06 | 100% | 2 | 1.5 | 7.0 | 56.8 | 4 | FBref Prior D_service |
| Aurèle Amenda | D | 91 | 0 | 0.00 | 0.00 | 100% | 0 | 0.2 | 6.7 | 52.0 | 4 | FBref Prior D_service |
| Jack Rudoni | M | 54 | 0 | 0.11 | 0.12 | ok | 1 | 0.8 | 7.2 | 48.5 | 4 | XGBoost M_service |
| Caleb Yirenkyi | M | 55 | 0 | 0.07 | 0.01 | 707% | 1 | 0.2 | 7.1 | 48.5 | 4 | FBref Prior M_service |
| Ephron Mason-Clark | M | 66 | 0 | 0.08 | - | n/a | 0 | 0.0 | 6.9 | 42.7 | 3 | FBref Prior M_service |
| Gustavo Hamer | M | 13 | 0 | 0.05 | - | n/a | 1 | 0.0 | 6.9 | 48.5 | 4 | FBref Prior M_service |
| Sidiki Cherif | F | 90 | 0 | 0.18 | - | n/a | 3 | - | 7.1 | - | 1 | Prior F_service |
| Haji Wright | F | 90 | 0 | 0.17 | - | n/a | 3 | - | 7.1 | - | 1 | Prior F_service |
| Yann Gboho | F | 90 | 0 | 0.27 | - | n/a | 3 | - | 7.1 | - | 0 | Prior F_service |
| Team totals | 0 | 1.02 | | 14 | | 7.1 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Daniel Bentley | G | 4.4m | 10.9% | Market-value substitute prior |
| Liam Kitching | D | 14.1m | 32.5% | Market-value substitute prior |
| Jay DaSilva | D | 13.8m | 22.3% | Market-value substitute prior |
| Victor Torp | M | 23.9m | 70.3% | Market-value substitute prior |
| Loum Tchaouna | M | 22.5m | 31.7% | Market-value substitute prior |
| Brandon Thomas-Asante | F | 29.3m | 53.9% | Market-value substitute prior |
Antagonist / tactical summary
Midfield is evenly contested — Morgan Gibbs-White vs Jack Rudoni will be the pivotal duel. Total antagonist xG — Home: 0.84, Away: 0.58. Home danger zone: LW vs CB. Away danger zone: Shadow Striker vs CB + Mid Cover — Layered Defense.
Midfield — possession 51.7% / 48.3%, dominance ratio 0.55, key duel: Morgan Gibbs-White vs Jack Rudoni
Morgan Gibbs-White xG 0.19Callum Hudson Odoi xG 0.07Igor Jesus xG 0.28Haji Wright xG 0.18Sidiki Cherif xG 0.20Caleb Yirenkyi xG 0.08
Summary
Antagonist matchup analysis for Forest vs Coventry. Most likely scoreline: 1-0. Lambdas derived from 5 home and 5 away individual player duels.
Tactical parity between 4-3-3 and 3-4-2-1. Match outcome driven primarily by individual player quality. Midfield is evenly contested — Morgan Gibbs-White vs Jack Rudoni will be the pivotal duel. Total antagonist xG — Home: 0.84, Away: 0.58. Home danger zone: LW vs CB. Away danger zone: Shadow Striker vs CB + Mid Cover — Layered Defense.
Newcastle vs Hull
Premier League · 20260919 · 4-3-3 vs 3-5-2
Predicted 2-1Winner: NewcastleConfidence 0.65λ home/away 2.59 / 1.00Most-likely P 9.2%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Newcastle vs Hull. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-3-3 vs 3-5-2 (duel lambdas: Home 2.381, Away 1.026). Final calibrated Poisson lambdas: Home 2.594, Away 1.002 (most likely score 2-1).
Home Starters 11/11 XGBoost (0 FBref) Away Starters 7/11 XGBoost (3 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 2.38 / 1.03 Final λ (H/A) 2.59 / 1.00
FBref Team Priors: Newcastle United (Gls/90: 2.0, GA/90: 2.0 → E[λ]: 2.222) vs Hull City (Gls/90: 1.5, GA/90: 1.5 → E[λ]: 2.222) | Blend Weight: 0.192
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Newcastle | Hull |
|---|
| League position | 7.00 | 3.00 |
| Points per game | 1.67 | 2.33 |
| Attack index | 2.50 | 0.62 |
| Defense (concede) index | 1.00 | 1.67 |
| Form multiplier | 0.96 | 1.03 |
| Standing λ multiplier | 1.46 | 1.02 |
Lambda decomposition (how the goal rates were built)
| Stage | Newcastle λ | Hull λ |
|---|
| FBref Squad Prior E[λ] (weight 0.192) | 2.222 | 2.222 |
| Pre-mismatch | 2.381 | 1.026 |
| Pre-calibration (Micro-duels) | 2.381 | 1.026 |
| Final λ | 2.594 | 1.002 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Truecompetitive underdog floor applied Falsestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 2-1 | 9.2% |
| 2-0 | 9.2% |
| 3-1 | 8.0% |
| 3-0 | 8.0% |
| 1-1 | 7.2% |
| Home win | 71.5% |
| Draw | 16.2% |
| Away win | 12.3% |
| Over 2.5 | 69.6% |
| BTTS yes | 58.6% |
Pregame fair line — favorite Newcastle (71.5%), fair total 3.60, fair spread 1.59
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Newcastle — projected line vs season baseline (4-3-3)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Lukáš Horníček | G | 93 | 0 | 0.00 | - | n/a | | 0.0 | 7.2 | 58.0 | 4 | XGBoost G_service |
| Tino Livramento | D | 1 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.4 | - | 2 | XGBoost D_service |
| Malick Thiaw | D | 95 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.2 | 49.0 | 4 | XGBoost D_service |
| Lewis Hall | D | 95 | 0 | 0.00 | 0.01 | 100% | 1 | 0.5 | 7.2 | 58.2 | 4 | XGBoost D_service |
| Sven Botman | D | 95 | 0 | 0.00 | 0.00 | 100% | 0 | 0.2 | 7.3 | 50.5 | 4 | XGBoost D_service |
| Bazoumana Touré | M | 1 | 0 | 0.20 | - | n/a | 2 | 0.0 | 7.4 | - | 4 | XGBoost M_service |
| Nico González | M | 81 | 0 | 0.09 | - | n/a | | 0.0 | 7.5 | 32.5 | 4 | XGBoost M_service |
| Jacob Ramsey | M | 25 | 0 | 0.14 | 0.04 | 221% | | 0.5 | 7.4 | 51.0 | 4 | XGBoost M_service |
| Matías Fernández-Pardo | F | 29 | 0 | 0.32 | - | n/a | 2 | 0.0 | 7.4 | 32.0 | 2 | XGBoost F_service |
| Harvey Barnes | F | 95 | 1 | 0.71 | 0.06 | 1180% | 2 | 1.0 | 7.3 | 51.8 | 4 | XGBoost F_service |
| Anthony Elanga | F | 84 | 1 | 0.88 | 0.10 | 747% | 2 | 1.0 | 7.4 | 55.5 | 4 | XGBoost F_service |
| Team totals | 2 | 2.59 | | 10 | | 7.2 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Ewen Jaouen | G | 4.8m | 13.0% | Market-value substitute prior |
| Amar Dedić | D | 14.3m | 33.3% | Market-value substitute prior |
| Dan Burn | D | 13.6m | 30.1% | Market-value substitute prior |
| Joelinton | M | 23.0m | 54.5% | Market-value substitute prior |
| Lewis Miley | M | 23.0m | 54.5% | Market-value substitute prior |
| Nick Woltemade | F | 23.1m | 75.2% | Market-value substitute prior |
Hull — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Konstantinos Tzolakis | G | 93 | 0 | 0.00 | - | n/a | | 0.0 | 7.5 | 47.2 | 5 | XGBoost G_service |
| Ryan Giles | D | 89 | 0 | 0.00 | 0.00 | 100% | 0 | 0.2 | 6.8 | 43.2 | 5 | FBref Prior D_service |
| Charlie Hughes | D | 90 | 0 | 0.02 | - | n/a | 0 | - | 6.9 | - | 1 | Prior D_service |
| Lucas Herrington | D | 25 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.5 | 51.0 | 4 | FBref Prior D_service |
| Christos Mouzakitis | M | 90 | 0 | 0.12 | - | n/a | 2 | 0.0 | 7.2 | - | 1 | Prior M_service |
| Tim Iroegbunam | M | 8 | 0 | 0.07 | - | n/a | | 0.0 | 6.9 | 65.0 | 1 | XGBoost M_service |
| Brooke Norton-Cuffy | M | 27 | 0 | 0.09 | - | n/a | 1 | 0.0 | 7.0 | 67.0 | 1 | XGBoost M_service |
| Joe Gelhardt | M | 90 | 0 | 0.11 | - | n/a | 1 | - | 7.1 | - | 0 | Prior M_service |
| Sorba Thomas | M | 28 | 0 | 0.07 | - | n/a | 1 | 0.0 | 7.0 | 67.0 | 1 | XGBoost M_service |
| Robinio Vaz | F | 90 | 1 | 0.20 | - | n/a | 3 | 0.0 | 7.1 | - | 1 | Prior F_service |
| Mohamed Ali Cho | F | 11 | 0 | 0.18 | 0.03 | 607% | 0 | 0.5 | 5.7 | 64.5 | 2 | FBref Prior F_service |
| Team totals | 1 | 1.00 | | 9 | | 6.9 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Jack Butland | G | 4.4m | 10.9% | Market-value substitute prior |
| Nobel Mendy | D | 13.8m | 22.2% | Market-value substitute prior |
| Cody Drameh | D | 24.3m | 60.2% | Market-value substitute prior |
| Jens Hjertø-Dahl | M | 22.3m | 51.3% | Market-value substitute prior |
| Elliot Stroud | M | 22.2m | 31.2% | Market-value substitute prior |
| Ilyas Ansah | F | 25.8m | 58.0% | Market-value substitute prior |
Antagonist / tactical summary
Midfield is evenly contested — Bazoumana Touré vs Christos Mouzakitis will be the pivotal duel. Total antagonist xG — Home: 0.86, Away: 0.48. Home danger zone: Right Wing vs Left Wingback — Compact Flank. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Midfield — possession 51.7% / 48.3%, dominance ratio 0.55, key duel: Bazoumana Touré vs Christos Mouzakitis
Harvey Barnes xG 0.29Anthony Elanga xG 0.36Matías Fernández-Pardo xG 0.13Robinio Vaz xG 0.19Brooke Norton-Cuffy xG 0.05Sorba Thomas xG 0.04
Summary
Antagonist matchup analysis for Newcastle vs Hull. Most likely scoreline: 2-1. Lambdas derived from 4 home and 5 away individual player duels.
4-3-3 creates structural superiority against 3-5-2. Home team should exploit wider spaces and numerical advantages in transition. Midfield is evenly contested — Bazoumana Touré vs Christos Mouzakitis will be the pivotal duel. Total antagonist xG — Home: 0.86, Away: 0.48. Home danger zone: Right Wing vs Left Wingback — Compact Flank. Away danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle.
Everton vs Ipswich
Premier League · 20260919 · 3-4-2-1 vs 3-5-2
Predicted 2-1Winner: EvertonConfidence 0.65λ home/away 2.65 / 1.14Most-likely P 9.0%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Everton vs Ipswich. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 3-4-2-1 vs 3-5-2 (duel lambdas: Home 2.655, Away 1.144). Final calibrated Poisson lambdas: Home 2.655, Away 1.144 (most likely score 2-1).
Home Starters 10/11 XGBoost (10 FBref) Away Starters 9/11 XGBoost (9 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 2.65 / 1.14 Final λ (H/A) 2.65 / 1.14
FBref Team Priors: Everton (Gls/90: 1.5, GA/90: 1.5 → E[λ]: 2.222) vs Ipswich Town (Gls/90: 2.0, GA/90: 2.0 → E[λ]: 2.222) | Blend Weight: 0.192
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Everton | Ipswich |
|---|
| League position | 8.00 | 13.00 |
| Points per game | 1.67 | 1.00 |
| Attack index | 2.08 | 0.83 |
| Defense (concede) index | 1.67 | 1.25 |
| Form multiplier | 1.03 | 0.96 |
| Standing λ multiplier | 1.65 | 1.01 |
Lambda decomposition (how the goal rates were built)
| Stage | Everton λ | Ipswich λ |
|---|
| FBref Squad Prior E[λ] (weight 0.192) | 2.222 | 2.222 |
| Pre-calibration (Micro-duels) | 2.655 | 1.144 |
| Final λ | 2.655 | 1.144 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Falsecompetitive underdog floor applied Falsestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 2-1 | 9.0% |
| 3-1 | 8.0% |
| 2-0 | 7.9% |
| 3-0 | 7.0% |
| 1-1 | 6.8% |
| Home win | 69.5% |
| Draw | 16.4% |
| Away win | 14.1% |
| Over 2.5 | 73.1% |
| BTTS yes | 63.4% |
Pregame fair line — favorite Everton (69.5%), fair total 3.80, fair spread 1.51
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Everton — projected line vs season baseline (3-4-2-1)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Jordan Pickford | G | 92 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.9 | 42.6 | 5 | FBref Prior G_service |
| Jarrad Branthwaite | D | 95 | 0 | 0.00 | 0.03 | 100% | 1 | 0.6 | 7.3 | 41.6 | 5 | FBref Prior D_service |
| Vitaliy Mykolenko | D | 95 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.0 | 41.6 | 5 | FBref Prior D_service |
| Jake O'Brien | D | 90 | 0 | 0.04 | - | n/a | 0 | 0.0 | 7.0 | - | 4 | FBref Prior D_service |
| James Garner | M | 55 | 0 | 0.10 | - | n/a | 2 | 0.0 | 7.5 | 50.5 | 4 | FBref Prior M_service |
| Kiernan Dewsbury-Hall | M | 95 | 0 | 0.38 | 0.09 | 315% | 2 | 1.8 | 7.5 | 45.2 | 5 | FBref Prior M_service |
| Hayden Hackney | M | 44 | 0 | 0.12 | 0.01 | 912% | 1 | 0.5 | 7.4 | 51.5 | 4 | FBref Prior M_service |
| Jack Grealish | M | 17 | 0 | 0.14 | 0.01 | 1798% | 1 | 0.3 | 7.2 | 46.0 | 3 | XGBoost M_service |
| Brennan Johnson | F | 39 | 1 | 0.49 | 0.01 | 5502% | 3 | 0.2 | 7.2 | 31.7 | 6 | FBref Prior F_service |
| Thierno Barry | F | 85 | 1 | 0.63 | 0.46 | 35% | 2 | 1.8 | 7.2 | 40.6 | 5 | FBref Prior F_service |
| Tyrique George | F | 82 | 0 | 0.55 | 0.04 | 1167% | 2 | 1.3 | 7.2 | 37.0 | 6 | FBref Prior F_service |
| Team totals | 2 | 2.65 | | 16 | | 7.2 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Mark Travers | G | 4.4m | 10.9% | Market-value substitute prior |
| Ainsley Maitland-Niles | D | 13.9m | 31.6% | Market-value substitute prior |
| Nathan Patterson | D | 13.9m | 31.6% | Market-value substitute prior |
| Iliman Ndiaye | M | 23.5m | 33.9% | Market-value substitute prior |
| Tim Iroegbunam | M | 13.0m | 70.6% | Market-value substitute prior |
| Beto | F | 21.1m | 73.4% | Market-value substitute prior |
Ipswich — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Kjell Scherpen | G | 90 | 0 | 0.00 | - | n/a | 0 | 0.0 | 7.2 | 56.0 | 4 | FBref Prior G_service |
| Leif Davis | D | 94 | 0 | 0.00 | 0.16 | 100% | 1 | 0.8 | 6.8 | 50.0 | 4 | FBref Prior D_service |
| Dara O'Shea | D | 93 | 0 | 0.00 | 0.02 | 100% | 0 | 0.2 | 6.8 | 46.5 | 4 | FBref Prior D_service |
| Jacob Greaves | D | 92 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.7 | 45.8 | 4 | FBref Prior D_service |
| Exequiel Palacios | M | 56 | 0 | 0.05 | - | n/a | 1 | 0.0 | 7.3 | 47.0 | 3 | FBref Prior M_service |
| Julio Enciso | M | 88 | 0 | 0.26 | 0.22 | ok | 2 | 3.0 | 7.2 | 52.8 | 4 | FBref Prior M_service |
| Jack Clarke | M | 22 | 0 | 0.09 | 0.05 | 77% | 1 | 1.0 | 7.1 | 47.3 | 3 | FBref Prior M_service |
| Abdul Fatawu Issahaku | M | 79 | 0 | 0.12 | 0.01 | 1461% | 1 | 0.5 | 7.0 | 52.8 | 4 | FBref Prior M_service |
| Jaden Philogene-Bidace | M | 90 | 0 | 0.13 | - | n/a | 2 | - | 7.1 | - | 0 | Prior M_service |
| Anis Mehmeti | F | 90 | 1 | 0.22 | - | n/a | 3 | - | 7.1 | - | 0 | Prior F_service |
| Daizen Maeda | F | 74 | 0 | 0.19 | 0.14 | 32% | 2 | 0.6 | 6.9 | 36.8 | 5 | FBref Prior F_service |
| Team totals | 1 | 1.14 | | 13 | | 7.0 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Christian Walton | G | 4.4m | 10.9% | Market-value substitute prior |
| Issa Diop | D | 13.8m | 22.2% | Market-value substitute prior |
| Darnell Furlong | D | 16.0m | 60.2% | Market-value substitute prior |
| Sindre Walle Egeli | M | 22.6m | 52.6% | Market-value substitute prior |
| Florentino Luís | M | 22.6m | 52.5% | Market-value substitute prior |
| Zian Flemming | F | 25.8m | 57.9% | Market-value substitute prior |
Antagonist / tactical summary
Midfield is evenly contested — James Garner vs Exequiel Palacios will be the pivotal duel. Total antagonist xG — Home: 1.15, Away: 0.73. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: LWB vs RWB.
Midfield — possession 50.5% / 49.5%, dominance ratio 0.52, key duel: James Garner vs Exequiel Palacios
Hayden Hackney xG 0.06James Garner xG 0.06Kiernan Dewsbury-Hall xG 0.20Julio Enciso xG 0.21Exequiel Palacios xG 0.04Abdul Fatawu Issahaku xG 0.08
Summary
Antagonist matchup analysis for Everton vs Ipswich. Most likely scoreline: 2-1. Lambdas derived from 7 home and 6 away individual player duels.
Tactical parity between 3-4-2-1 and 3-5-2. Match outcome driven primarily by individual player quality. Midfield is evenly contested — James Garner vs Exequiel Palacios will be the pivotal duel. Total antagonist xG — Home: 1.15, Away: 0.73. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: LWB vs RWB.
Osasuna vs Rayo Vallecano
LaLiga · 20260919 · 4-2-3-1 vs 4-3-3
Predicted 0-1Winner: Rayo VallecanoConfidence 0.56λ home/away 1.12 / 1.21Most-likely P 11.6%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Osasuna vs Rayo Vallecano. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 4-2-3-1 vs 4-3-3 (duel lambdas: Home 1.125, Away 1.212). Final calibrated Poisson lambdas: Home 1.125, Away 1.212 (most likely score 0-1).
Home Starters 9/11 XGBoost (8 FBref) Away Starters 9/11 XGBoost (5 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 1.12 / 1.21 Final λ (H/A) 1.12 / 1.21
FBref Team Priors: Osasuna (Gls/90: 1.0, GA/90: 1.0 → E[λ]: 0.985) vs Rayo Vallecano (Gls/90: 1.3333333333333333, GA/90: 1.33 → E[λ]: 0.988) | Blend Weight: 0.228
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Osasuna | Rayo Vallecano |
|---|
| League position | 8.00 | 14.00 |
| Points per game | 1.75 | 1.00 |
| Attack index | 1.04 | 1.46 |
| Defense (concede) index | 2.08 | 1.25 |
| Form multiplier | 1.04 | 0.96 |
| Standing λ multiplier | 1.39 | 1.27 |
Lambda decomposition (how the goal rates were built)
| Stage | Osasuna λ | Rayo Vallecano λ |
|---|
| FBref Squad Prior E[λ] (weight 0.228) | 0.985 | 0.988 |
| Pre-competitive-floor | 1.125 | 1.212 |
| Pre-calibration (Micro-duels) | 1.125 | 1.212 |
| Final λ | 1.125 | 1.212 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Falsemismatch score pressure applied Falsecompetitive underdog floor applied Falsestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 0-1 | 11.6% |
| 1-1 | 13.3% |
| 1-0 | 10.7% |
| 0-0 | 9.8% |
| 1-2 | 8.0% |
| Home win | 33.7% |
| Draw | 28.3% |
| Away win | 38.0% |
| Over 2.5 | 41.4% |
| BTTS yes | 47.6% |
Pregame fair line — favorite Rayo Vallecano (38.0%), fair total 2.34, fair spread -0.09
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Osasuna — projected line vs season baseline (4-2-3-1)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Sergio Herrera | G | 88 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.7 | 44.8 | 6 | FBref Prior G_service |
| Enzo Boyomo | D | 57 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.8 | 22.2 | 6 | FBref Prior D_service |
| Abel Bretones | D | 88 | 0 | 0.00 | 0.01 | 100% | 1 | 0.3 | 6.3 | 42.2 | 6 | FBref Prior D_service |
| Jorge Herrando | D | 61 | 0 | 0.00 | 0.05 | 100% | 1 | 0.5 | 6.6 | 50.5 | 4 | FBref Prior D_service |
| Valentin Rosier | D | 90 | 0 | 0.03 | - | n/a | 0 | - | 6.8 | - | 1 | Prior D_service |
| Jon Moncayola | M | 79 | 0 | 0.06 | 0.01 | 332% | 1 | 0.3 | 6.8 | 44.7 | 6 | FBref Prior M_service |
| Aimar Oroz | M | 50 | 0 | 0.14 | 0.03 | 394% | 2 | 0.5 | 6.8 | 50.2 | 4 | FBref Prior M_service |
| Romain Del Castillo | M | 90 | 0 | 0.17 | - | n/a | 1 | 0.0 | 6.9 | - | 2 | Prior M_service |
| Jonathan Dubasin | M | 35 | 0 | 0.11 | 0.07 | 49% | 2 | 0.3 | 6.7 | 36.7 | 7 | FBref Prior M_service |
| Raúl Moro | M | 24 | 0 | 0.16 | - | n/a | 2 | 0.0 | 6.8 | 22.8 | 6 | FBref Prior M_service |
| Raúl García de Haro | F | 51 | 0 | 0.22 | 0.07 | 215% | 2 | 1.2 | 6.6 | 53.8 | 5 | XGBoost F_service |
| Team totals | 0 | 1.12 | | 14 | | 6.7 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Aitor Fernández | G | 4.4m | 10.9% | Market-value substitute prior |
| Alejandro Catena | D | 13.6m | 29.9% | Market-value substitute prior |
| Asier Osambela | D | 13.6m | 21.7% | Market-value substitute prior |
| Lucas Torró | M | 18.1m | 56.4% | Market-value substitute prior |
| Iker Muñoz | M | 34.6m | 43.5% | Market-value substitute prior |
| Ante Budimir | F | 29.5m | 46.2% | Market-value substitute prior |
Rayo Vallecano — projected line vs season baseline (4-3-3)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Augusto Batalla | G | 74 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.8 | 46.7 | 3 | FBref Prior G_service |
| Andrei Rațiu | D | 93 | 0 | 0.18 | 0.08 | 123% | 2 | 0.8 | 6.9 | 32.8 | 6 | FBref Prior D_service |
| Mujaid Sadick | D | 90 | 0 | 0.02 | - | n/a | 0 | 0.0 | 6.8 | - | 2 | Prior D_service |
| Marash Kumbulla | D | 90 | 0 | 0.02 | - | n/a | 0 | - | 6.8 | - | 1 | Prior D_service |
| Adrià Pedrosa | D | 65 | 0 | 0.00 | 0.02 | 100% | 1 | 0.5 | 6.6 | 31.2 | 4 | FBref Prior D_service |
| Isi Palazón | M | 40 | 0 | 0.05 | 0.01 | 257% | 1 | 0.3 | 6.7 | 44.3 | 3 | XGBoost M_service |
| Pedro Díaz | M | 44 | 0 | 0.05 | 0.01 | 384% | 2 | 0.3 | 6.7 | 33.5 | 6 | FBref Prior M_service |
| Giorgi Tsitaishvili | M | 16 | 0 | 0.07 | - | n/a | | 0.0 | 6.7 | 16.2 | 4 | XGBoost M_service |
| Jorge de Frutos | F | 76 | 0 | 0.29 | 0.01 | 1892% | 2 | 0.2 | 7.0 | 52.6 | 5 | XGBoost F_service |
| Alemão | F | 24 | 0 | 0.12 | 0.05 | 129% | 2 | 0.4 | 6.7 | 39.4 | 5 | XGBoost F_service |
| Sergio Camello | F | 61 | 1 | 0.31 | 0.30 | ok | 2 | 1.7 | 7.5 | 52.7 | 6 | FBref Prior F_service |
| Team totals | 1 | 1.21 | | 14 | | 6.8 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Emil Audero | G | 4.4m | 10.9% | Market-value substitute prior |
| Luiz Felipe | D | 13.6m | 29.9% | Market-value substitute prior |
| Jozhua Vertrouwd | D | 13.6m | 21.7% | Market-value substitute prior |
| Unai López | M | 22.0m | 30.7% | Market-value substitute prior |
| Fran Pérez | M | 22.0m | 49.9% | Market-value substitute prior |
| Randy Nteka | F | 36.1m | 59.1% | Market-value substitute prior |
Antagonist / tactical summary
Midfield is evenly contested — Jon Moncayola vs Isi Palazón will be the pivotal duel. Total antagonist xG — Home: 0.37, Away: 0.73. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Right Wing vs Left Back — Flank Battle.
Midfield — possession 51.0% / 49.0%, dominance ratio 0.54, key duel: Jon Moncayola vs Isi Palazón
Aimar Oroz xG 0.08Raúl Moro xG 0.08Jonathan Dubasin xG 0.06Jorge de Frutos xG 0.26Sergio Camello xG 0.28Alemão xG 0.11
Summary
Antagonist matchup analysis for Osasuna vs Rayo Vallecano. Most likely scoreline: 0-1. Lambdas derived from 5 home and 5 away individual player duels.
Tactical parity between 4-2-3-1 and 4-3-3. Match outcome driven primarily by individual player quality. Midfield is evenly contested — Jon Moncayola vs Isi Palazón will be the pivotal duel. Total antagonist xG — Home: 0.37, Away: 0.73. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Right Wing vs Left Back — Flank Battle.
Celta vs R. Racing Club
LaLiga · 20260919 · 3-5-2 vs 3-5-2
Predicted 1-1Winner: DrawConfidence 0.56λ home/away 1.02 / 1.02Most-likely P 13.7%
Actual vs predicted: result not available yet (upcoming fixture).
XGBoost Reasoning & FBref Dataset TracingAntagonist + XGBoost match analysis for Celta vs R. Racing Club. Player statistics regressed using position-specific XGBoost models (G, D, M, F) with 2026-2027 FBref priors. Micro-duels resolved across 3-5-2 vs 3-5-2 (duel lambdas: Home 0.891, Away 1.018). Final calibrated Poisson lambdas: Home 1.020, Away 1.018 (most likely score 1-1).
Home Starters 10/11 XGBoost (10 FBref) Away Starters 11/11 XGBoost (0 FBref) Bench Modeled H: 7 / A: 7 Duel λ (H/A) 0.89 / 1.02 Final λ (H/A) 1.02 / 1.02
Modeling Pipeline Stages (5 layers)
- 1. XGBoost player regressions for volume & event rates with market-value quality scaling
- 2. Multi-source prior blending (FBref 2026-2027 tables + Sofascore season history)
- 3. Antagonist formation duel matrix resolving 1v1 and zonal matchups
- 4. Team-level prior calibration (FBref possession and attack/defense efficiencies)
- 5. Bivariate Poisson matrix with Dixon-Coles low-score correction
Standing & form context
| Metric | Celta | R. Racing Club |
|---|
| League position | 16.00 | 13.00 |
| Points per game | 0.60 | 1.00 |
| Attack index | 0.42 | 1.46 |
| Defense (concede) index | 1.67 | 1.04 |
| Form multiplier | 0.98 | 1.02 |
| Standing λ multiplier | 0.86 | 1.19 |
Lambda decomposition (how the goal rates were built)
| Stage | Celta λ | R. Racing Club λ |
|---|
| Pre-competitive-floor | 0.891 | 1.018 |
| Pre-calibration (Micro-duels) | 0.891 | 1.018 |
| Final λ | 1.020 | 1.018 |
competition calibration scale 1.0competition observed total goals 0.0quality gap applied Truemismatch score pressure applied Falsecompetitive underdog floor applied Truestage discount 1.0
Scoreline distribution & markets
| Top 5 scorelines | P |
|---|
| 1-1 | 13.7% |
| 0-0 | 13.2% |
| 1-0 | 13.2% |
| 0-1 | 13.1% |
| 2-1 | 6.9% |
| Home win | 34.7% |
| Draw | 30.8% |
| Away win | 34.6% |
| Over 2.5 | 33.4% |
| BTTS yes | 41.0% |
Pregame fair line — favorite Celta (34.7%), fair total 2.04, fair spread 0.00
Projected players vs season baseline
Proj = projected for this match · Season = this season's per-game average from match data · deviation badge compares projected xG to the player's season xG/game.
Celta — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Ionuț Radu | G | 90 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.8 | 72.4 | 5 | FBref Prior G_service |
| Javi Rodríguez | D | 61 | 0 | 0.00 | 0.00 | 100% | 0 | 0.2 | 6.8 | 52.2 | 5 | FBref Prior D_service |
| Abdoulaye Faye | D | 44 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.6 | 15.2 | 4 | FBref Prior D_service |
| Sebastián Cáceres | D | 90 | 0 | 0.02 | - | n/a | 0 | 0.0 | 6.8 | - | 1 | Prior D_service |
| Williot Swedberg | M | 56 | 0 | 0.12 | 0.08 | 57% | 2 | 0.6 | 7.0 | 54.4 | 5 | FBref Prior M_service |
| Miguel Román | M | 83 | 0 | 0.10 | 0.01 | 1030% | 2 | 0.6 | 7.1 | 69.6 | 5 | FBref Prior M_service |
| Ilaix Moriba | M | 67 | 0 | 0.06 | 0.01 | 384% | 1 | 0.6 | 7.0 | 66.0 | 5 | FBref Prior M_service |
| Sergio Carreira | M | 65 | 0 | 0.10 | 0.03 | 210% | 1 | 0.5 | 7.0 | 58.7 | 6 | FBref Prior M_service |
| Hugo Álvarez | M | 38 | 0 | 0.13 | - | n/a | 1 | 0.0 | 7.0 | 38.8 | 5 | FBref Prior M_service |
| Ferrán Jutglà | F | 50 | 1 | 0.18 | 0.01 | 3204% | 2 | 0.2 | 6.8 | 65.2 | 5 | FBref Prior F_service |
| Jones El-Abdellaoui | F | 51 | 0 | 0.20 | 0.06 | 204% | 2 | 0.6 | 6.8 | 38.0 | 5 | FBref Prior F_service |
| Team totals | 1 | 1.02 | | 11 | | 6.9 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Altay Bayındır | G | 4.5m | 11.3% | Market-value substitute prior |
| Carl Starfelt | D | 27.6m | 41.1% | Market-value substitute prior |
| Álvaro Núñez | D | 13.6m | 21.8% | Market-value substitute prior |
| Javier Rueda | M | 22.1m | 50.6% | Market-value substitute prior |
| Aleix Febas | M | 22.1m | 30.9% | Market-value substitute prior |
| Couhaib Driouech | F | 23.6m | 72.3% | Market-value substitute prior |
R. Racing Club — projected line vs season baseline (3-5-2)
| Player | Pos | Min | Proj G | Proj xG | Season xG/g | | Proj Sh | Season Sh/g | Proj Rtg | Season Rtg | Rec | Model / FBref Prior |
|---|
| Julen Agirrezabala | G | 91 | 0 | 0.00 | - | n/a | | 0.0 | 7.1 | 54.2 | 5 | XGBoost G_service |
| Jeanuël Belocian | D | 36 | 0 | 0.11 | 0.04 | 153% | 1 | 0.5 | 6.8 | 36.0 | 2 | XGBoost D_service |
| Jorge Salinas | D | 90 | 0 | 0.00 | 0.04 | 100% | 0 | 0.2 | 6.6 | 53.2 | 5 | XGBoost D_service |
| Facundo González | D | 41 | 0 | 0.00 | - | n/a | 0 | 0.0 | 6.6 | 63.5 | 4 | XGBoost D_service |
| Andrés Martín | M | 80 | 0 | 0.23 | 0.79 | 71% | 2 | 3.0 | 6.9 | 68.3 | 3 | XGBoost M_service |
| Pablo García | M | 35 | 0 | 0.06 | 0.11 | 41% | 2 | 1.2 | 6.8 | 56.2 | 5 | XGBoost M_service |
| Iñigo Vicente | M | 84 | 0 | 0.06 | 0.01 | 633% | 1 | 0.2 | 6.8 | 58.0 | 5 | XGBoost M_service |
| André Almeida | M | 35 | 0 | 0.08 | - | n/a | | 0.0 | 6.6 | 24.3 | 3 | XGBoost M_service |
| Matteo Prati | M | 43 | 0 | 0.05 | - | n/a | 1 | 0.0 | 6.8 | 44.0 | 3 | XGBoost M_service |
| Yassir Zabiri | F | 52 | 1 | 0.22 | 0.15 | 49% | 2 | 1.2 | 7.0 | 56.0 | 5 | XGBoost F_service |
| Giorgi Guliashvili | F | 21 | 0 | 0.14 | - | n/a | 2 | 0.0 | 6.6 | 31.5 | 2 | XGBoost F_service |
| Team totals | 1 | 1.02 | | 11 | | 6.8 | | |
Bench Candidates (7 modeled substitutes with empirical playing time)
| Player | Pos | Exp Min | App Prob | Substitute Modeling Details |
|---|
| Simon Eriksson | G | 4.4m | 10.9% | Market-value substitute prior |
| Pedro Felipe | D | 13.6m | 29.9% | Market-value substitute prior |
| Pablo Ramón | D | 13.6m | 29.9% | Market-value substitute prior |
| Gustavo Puerta | M | 22.7m | 32.2% | Market-value substitute prior |
| Aarón Martín | M | 22.1m | 50.3% | Market-value substitute prior |
| Asier Villalibre | F | 25.3m | 55.9% | Market-value substitute prior |
Antagonist / tactical summary
Midfield is evenly contested — Williot Swedberg vs Andrés Martín will be the pivotal duel. Total antagonist xG — Home: 0.55, Away: 0.62. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Central Mid vs Defensive Mid — Pressing/Creation Duel.
Midfield — possession 51.0% / 49.0%, dominance ratio 0.53, key duel: Williot Swedberg vs Andrés Martín
Sergio Carreira xG 0.06Williot Swedberg xG 0.09Miguel Román xG 0.07Iñigo Vicente xG 0.05Pablo García xG 0.05Andrés Martín xG 0.19
Summary
Antagonist matchup analysis for Celta vs R. Racing Club. Most likely scoreline: 1-1. Lambdas derived from 6 home and 6 away individual player duels.
Tactical parity between 3-5-2 and 3-5-2. Match outcome driven primarily by individual player quality. Midfield is evenly contested — Williot Swedberg vs Andrés Martín will be the pivotal duel. Total antagonist xG — Home: 0.55, Away: 0.62. Home danger zone: Striker vs Two Centre Backs — 1v2 Aerial/Physical Battle. Away danger zone: Central Mid vs Defensive Mid — Pressing/Creation Duel.
Generated 2026-09-19 13:33:57 UTC · Soccer prediction reporting