Prediction accuracy
The final call before each session, fully scored races only. White tick: what a random pick from the field would get.
Better than guessing
Each bar is the engine's hit rate across fully scored races. The white tick is the fair baseline: a random pick from the field for skill calls, and the best fixed guess for calls where one outcome dominates.
Easy calls are shown for completeness and never count toward the headline.
Race-by-race scoreboard
The call sharpens through the weekend
The same 10 races, scored twice: the race call made after the last practice session, and the one made after qualifying.
10 races with both calls scored. Race score, core calls only.
How scoring works
Every call is generated before the session and locked the moment it is saved. After the chequered flag it is scored against the official classification using the published weights, and the result is never edited: the scoreboard above is computed from the same records you can open on each race page.
Races and sprints reward the winner, the podium (presence and exact order), the full running order, team results and the driver who leads the most laps. Qualifying revolves around pole and the top 5. Weights are relative: each bucket is normalized against its own total. Lower-predictability calls (safety car, weather, top speed, pit stops, margin) are scored separately as easy calls and never lower the headline.
Race
The headline prediction: winner, full podium in order, the complete top 10 and the driver who leads the most laps. Scored on winner call, exact podium order, top-10 presence, how far each pick landed from its predicted position and the most-laps-led call, plus separate easy calls like fastest lap, top speed, safety car and margin of victory.
Qualifying
Pole position and the top 5 on Saturday pace. Qualifying is the purest test of single-lap form, and the post-qualifying race prediction re-anchors on this result, so quali accuracy compounds into race accuracy.
Sprint
On sprint weekends the short race gets its own locked prediction and its own score (winner and podium presence) completely separate from the Grand Prix call.
Head-to-head
Driver-vs-driver calls (mostly teammate battles) scored as simple right-or-wrong verdicts. Small individual weights, but across a season they reveal whether the engine reads intra-team form correctly.
How is prediction accuracy measured?
Each prediction is compared against the official classification once a session is complete. The headline score measures core race outcomes only: the winner call, podium order, the full running order, the driver who leads the most laps, and team results, producing a 0 to 100% figure per session. Lower-predictability calls (safety car, top speed, pit stops, margin, weather) are scored separately, shown as easy calls against the best fixed guess, and never blend into the headline. The same rubric is applied to every race, with calls added mid-season counting from the race they were introduced, and every score is public.
Are predictions ever edited after the results?
No. Predictions are locked before the session starts and kept exactly as published: they are never rewritten or quietly deleted. What you see on the scoreboard is exactly what was published before lights out.
How often does the AI pick the race winner?
The live winner hit rate is shown at the top of this page and updates after every scored race. Formula 1 winners are genuinely hard to call: treat any single-race result as noise and judge the rate over a full season.
What is a good baseline to compare against?
A random pick from the 22-driver grid wins about 4.5% of the time. Rain, safety cars and first-lap contact can upend any pick, so a winner hit rate is best read across a full season against that baseline.
What data do the predictions use?
The AI combines championship standings, recent driver form, circuit history, weather forecasts, and results from earlier sessions of the same weekend, plus grounded news search for late-breaking developments. It never sees the future: every input predates the session.
How is the accuracy score calculated?
Each session type has a weighted rubric (winner, podium order, top-10 accuracy for races; pole and top-5 for qualifying) normalized to a 0–100% score. The exact weights are published in the scoring section above, and every race page shows which individual calls hit.
Why do scores vary so much from race to race?
Because Formula 1 does. Chaotic races (rain, safety cars, first-lap incidents) scramble the field and punish every forecast, while processional races reward it. That is why the season-long average and hit rates matter more than any single weekend.
Do you ever remove or rewrite bad predictions?
Never. Each prediction is written once before its session and is never overwritten, and the scoreboard is computed from those same locked records. Misses stay on the board with the hits: that is the point of a public track record.
How do the results compare to just guessing?
A random winner pick from a 22-car grid succeeds about 4.5% of the time. The Calls vs chance section above puts the engine's live hit rates next to those odds: any sustained gap between the two is real predictive signal.
Think you can beat it?
Make your picks for the next race and get scored with the same rubric as the AI.