Abu Dhabi Grand Prix

Round 22 · 2026

Yas Marina

v 0.3.012 Jun 2026, 17:58 UTC

Based onHistoryRegulations

Max Verstappen is the model's pick at 35.0% (80% CI 34.1–35.8%) for Abu Dhabi Grand Prix.

Top 10 drivers

bars are win / podium / points probabilitiesWeekend analytics →

1
VERMax Verstappen
Red Bull RacingLOW
WIN
35.0%
POD
71.6%
PTS
93.7%
Why · 3 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 1.00 (1.0 = top of grid). Familiarity discount ×0.85.
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.91, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.94 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
2
LECCharles Leclerc
FerrariLOW
WIN
10.2%
POD
33.1%
PTS
82.5%
Why · 4 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.65 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.77, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.77 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
- DNF riskw 0.12
  Per-race DNF probability ≈ 11.8% (team × circuit).
3
RUSGeorge Russell
MercedesLOW
WIN
9.7%
POD
33.5%
PTS
85.5%
Why · 3 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.64 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.80, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.95 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
4
NORLando Norris
McLarenLOW
WIN
7.8%
POD
24.9%
PTS
70.5%
Why · 4 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.61 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.92, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.92 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
- DNF riskw 0.20
  Per-race DNF probability ≈ 22.9% (team × circuit).
5
GASPierre Gasly
AlpineLOW
WIN
4.8%
POD
16.3%
PTS
59.9%
Why · 4 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.60 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.21, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.28 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
- DNF riskw 0.20
  Per-race DNF probability ≈ 25.0% (team × circuit).
6
ALOFernando Alonso
Aston MartinLOW
WIN
4.1%
POD
15.3%
PTS
65.3%
Why · 3 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.57 (1.0 = top of grid). Familiarity discount ×0.85.
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.58, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.64 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
7
ALBAlexander Albon
WilliamsLOW
WIN
3.6%
POD
12.0%
PTS
55.3%
Why · 4 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.55 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.24, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.50 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
- DNF riskw 0.20
  Per-race DNF probability ≈ 25.0% (team × circuit).
8
SAICarlos Sainz
WilliamsLOW
WIN
3.4%
POD
11.9%
PTS
51.7%
Why · 4 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.50 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.66, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.50 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
- DNF riskw 0.20
  Per-race DNF probability ≈ 25.0% (team × circuit).
9
PIAOscar Piastri
McLarenLOW
WIN
2.9%
POD
11.1%
PTS
50.1%
Why · 4 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.44 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.83, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.92 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
- DNF riskw 0.20
  Per-race DNF probability ≈ 22.9% (team × circuit).
10
COLFranco Colapinto
AlpineLOW
WIN
2.6%
POD
9.7%
PTS
44.7%
Why · 4 factors
- Driver skill (Elo)w 0.58
  Teammate-only Elo skill score 0.47 (1.0 = top of grid).
- Track history (time-weighted)w 0.07
  Recency-weighted finish score 0.30, 5-year window, 12-month half-life, reg-discounted ×0.24.
- Chassis at this circuitw 0.35
  Constructor-lineage best-finish score 0.28 at this circuit across the last 5 seasons. Driver-agnostic — captures chassis-DNA fit independent of who drove.
- DNF riskw 0.20
  Per-race DNF probability ≈ 25.0% (team × circuit).

Most-likely podium combinations

approximate · 1st-2nd-3rd

1
P1VER·P2RUS·P3LEC
3.52%
2
P1VER·P2LEC·P3RUS
3.50%
3
P1VER·P2RUS·P3NOR
2.65%
4
P1VER·P2LEC·P3NOR
2.60%
5
P1VER·P2NOR·P3RUS
2.42%
6
P1VER·P2NOR·P3LEC
2.39%
7
P1VER·P2RUS·P3GAS
1.73%
8
P1VER·P2LEC·P3GAS
1.70%
9
P1LEC·P2VER·P3RUS
1.68%
10
P1VER·P2RUS·P3ALO
1.63%

Methodology

Probabilities come from a layered ensemble: historical base rates per circuit, current constructor form (rolling 6 races), qualifying outcome, and a weather adjustment. Confidence intervals are 80% credible bands from the win-share posterior. See 0.3.0 for the active weights.