Race Winner Prediction

4.940 km lap length19 cornersHigh overtaking difficulty

Race date October 11, 2026 · generated Oct 11, 2026, 7:21 PM UTC · rain scenario weighted at 20%

Predicted vs. Actual

Predicted winner

ANT (24.6%)

Actual winner

VER (missed)

Podium hits

3 / 3

Brier score (win)

0.0313

The model gave the actual winner (VER) a 24.3% chance beforehand. Predicted live, before the race. See the track record page for the model's accuracy across every graded race.

Our model: 0.0313 · Grid-based baseline (always pick the polesitter): 0.0000 (baseline beats the model this time — lower is better)

Predicted vs. actual finishing position, every driver

Green, right — finished better than the model expected (underrated). Red, left — finished worse (overrated). Sorted by size of the miss.

Win Probability — Top 10

Error bars show ± 1 standard deviation of win probability across batches of the Monte Carlo simulation — how much this estimate would wobble on a re-run, not a formal confidence interval.

Full Prediction Table

Click a row to see what drove that driver's score.

Pos	Driver	Team	Win %	Podium %	Points %	Exp. Pos	Actual
1	▸ANT	Mercedes	24.6% ± 0.5%	59%	97.2%	3.63	3
2	▸VER	Red Bull	24.3% ± 0.6%	59%	97.2%	3.63	1
3	▸LEC	Ferrari	17.3% ± 0.6%	49.4%	96.1%	4.2	2
4	▸HAM	Ferrari	9.9% ± 0.4%	34.6%	92.9%	5.22	6
5	▸RUS	Mercedes	8.9% ± 0.4%	31.7%	91.7%	5.5	9
6	▸NOR	McLaren	8.4% ± 0.5%	30.8%	91.5%	5.56	4
7	▸PIA	McLaren	2.6% ± 0.1%	12.7%	78.2%	7.78	5
8	▸LAW	RB F1 Team	1.3% ± 0.2%	6.7%	64.5%	9.3	7
9	▸GAS	Alpine F1 Team	0.7% ± 0.1%	3.6%	50.1%	10.72	11
10	▸LIN	RB F1 Team	0.7% ± 0.1%	4%	52.9%	10.44	15
11	▸HAD	Red Bull	0.4% ± 0.1%	2.6%	46.3%	11.05	10
12	▸HUL	Audi	0.2% ± 0.1%	1%	22%	13.82	8
13	▸BOR	Audi	0.2% ± 0.1%	1%	23.9%	13.54	22
14	▸COL	Alpine F1 Team	0.2% ± 0.1%	1.6%	32.2%	12.53	21
15	▸SAI	Williams	0.1%	0.4%	10%	15.91	14
16	▸OCO	Haas F1 Team	0.1% ± 0.1%	0.5%	13.7%	15.15	13
17	▸BEA	Haas F1 Team	0.1%	0.8%	19.1%	14.25	12
18	▸ALO	Aston Martin	0.1%	0.4%	11.4%	15.55	16
19	▸ALB	Williams	0%	0.1%	3.1%	18.28	18
20	▸STR	Aston Martin	0%	0.1%	2.2%	18.84	20
21	▸PER	Cadillac F1 Team	0%	0.1%	2%	18.95	17
22	▸BOT	Cadillac F1 Team	0%	0%	1.8%	19.16	19

Methodology

Each driver's simulated race pace is sampled from a distribution weighted by qualifying pace percentile, top-speed ranking, season championship standings, historical finishing position at this circuit, tire degradation, and pit-stop delta. The simulation runs 100,000 times, with 20% of runs treated as a wet-race scenario (estimated from this circuit's own rainfall history).

This circuit's weighting — Singapore Grand Prix: overtaking difficulty High, downforce dependency High, tire severity Medium. Higher overtaking difficulty shifts weight toward grid position; lower downforce dependency shifts weight toward straight-line top speed; higher tire severity shifts weight toward tire-degradation management. See circuit_profiles.py for every circuit's numbers and how they translate into weights.