RaoscaffResearch
Prediction Series · Lock · Issue P-20
Prediction Series · P-20

The F1 title, the back half run 40,000 times.

We simulated the 15 remaining rounds of the 2026 season 40,000 times — race-by-race finishing orders, sprint points, DNFs — from the real post-Barcelona standings (Antonelli 156, Hamilton 115, Russell 106). The current-form model makes Antonelli a heavy 81% champion. The betting market is more cautious at ~61%, pricing in a Verstappen/Red Bull resurgence the model doesn't yet see. We lock a blended ~70% — and show our work on the disagreement.

Type · Prediction Lock · rest-of-season Monte-Carlo (40k runs) Locked · 2026-06-21 · after Round 7, before Round 8 Resolves · ~Dec 2026 · final round, official FIA standings Scored · Brier on the champion binary + distribution
F1 2026 Drivers' title · our locked call
ANTONELLI ~70%
model 81% (current form) · market ~61% · blended lock

40,000 simulated run-outs of the 15 remaining rounds from the post-Barcelona table give Antonelli 81%. The outright market sits at ~61% (Verstappen ~25% despite being 7th, 101 points back) — it prices a Red Bull pace resurgence the current-form model doesn't assume. We lock the blend, ~70%, and disclose both inputs.

— 1 · The Locked Forecast

Antonelli champion — ~70%, model-vs-market blended.

We lock Kimi Antonelli to win the 2026 Drivers' Championship at ~70% — a deliberate blend of our 40,000-run current-form simulation (81%) and the more cautious outright market (~61%). The full simulated distribution is below; the headline is the blend, with the disagreement disclosed rather than hidden.

F1 2026 title · 40,000-run simulation from the post-Barcelona standings
DriverPts (R7)P(champion) — model
Antonelli15681.0%
Hamilton1156.7%
Russell1065.4%
Verstappen553.0%
Piastri681.6%
Norris621.3%
Leclerc751.0%
— 2 · The Engine

15 rounds, race by race — run in parallel.

Each simulated season plays the 15 remaining rounds in turn: a Plackett-Luce finishing order from driver pace + Gumbel noise, top-10 points (25-18-15-…), an independent ~9% per-race DNF that can drop a contender, and sprint points across the four remaining sprint weekends — accumulated onto the real Round-7 table. Repeat 40,000 times; the champion is whoever tops the final table. A 41-point lead plus a competitive car makes Antonelli the heavy favourite under current form.

Verified 2026-06-21: standings after the Barcelona-Catalunya GP (Antonelli 156, Hamilton 115, Russell 106, Leclerc 75, Verstappen 55, 7th); 15 rounds + 4 sprints remaining of a 22-race season. Source: Formula1.com / FIA standings.

— 3 · The Disagreement

Model 81% vs market 61% — we show the gap.

This is the honest centre of the brief. Our current-form model is more bullish on Antonelli (81%) than the betting market (~61%). The market has Verstappen second-favourite at ~25% despite his being seventh and 101 points adrift — a bet that Red Bull will field the season's dominant car from here and Verstappen will simply out-pace the field for fifteen rounds. Our model doesn't assume that pace shift; it runs the cars roughly as they are now. We don't pretend to resolve who's right — we lock the blend (~70%) and let the season score it.

— 4 · Methodology

Rest-of-season Monte-Carlo: reproducible, fixed-seed.

The simulator is a committed, fixed-seed script (scripts/sim/f1-2026-title.py) — 40,000 parallel season run-outs, deterministic. We synthesise current standings + a pace model and disclose the market blend; we do not claim a proprietary F1 model. Scored by Brier on the champion binary against the final FIA standings.

— 5 · Pre-Committed Post-Mortem

How this resolves wrong — named in advance.

A Red Bull resurgence — if Verstappen's pace takes over and he claws back 101 points, the market's caution is vindicated and our ~70% looks high. Antonelli cruises — if Mercedes stays dominant, the model's 81% looks right and the blend was too cautious. Either way the champion binary is scored honestly against the final standings. (Companion to P-13, which locks only the lead after the British GP.)

The season simulated 40,000 times, locked — scored against the final standings.

RAOSCAFF locks P-20 on 2026-06-21, after Round 7. We lock Antonelli at ~70% to win the title — a disclosed blend of the 40,000-run current-form simulation (81%) and the outright market (~61%). Scored by Brier on the champion binary against the final FIA classification.

Locked
2026-06-21 (commit timestamp on origin/main)
Resolves
~Dec 2026 — final round, official FIA Drivers' Championship
Source
FIA / Formula1.com final standings
Scored by
Brier on the champion binary + the simulated distribution

Fixed seed, committed script, and the model-vs-market gap disclosed in the open — the season, simulated and checkable.