.jpg){kind=link}
At a glance
Across 431 paired grid and classified-finish results from the 24 Grands Prix of 2024, Pearson r was 0.7845. Yet only 67 drivers (15.5%) finished in their exact starting position.
- Paired results431numeric grid and finish positions
- Pearson correlation0.7845grid rank vs. classified finish
- Exact position retained15.5%67 of 431 driver-race pairs
- Within two places62.9%271 of 431 pairs
A race preview often starts with the grid. That is a sensible anchor, but how close did the 2024 Sunday classifications stay to Saturday’s starting order? We paired each driver’s official numbered grid position with their final numbered race classification at all 24 Grands Prix. Sprints are excluded. The comparison covers repeated driver-race rows, not 431 different drivers.
A strong overall relationship, with a wide margin for movement
For 431 pairs with a numeric grid slot and numeric classified finish, Pearson’s linear correlation was 0.7845. A one-variable linear fit gives R² = 0.6154. This describes association in the observed ranks; it is not a causal explanation or an out-of-sample forecast test. The same underlying car and driver competitiveness influence both the grid and the eventual classification, while strategy, incidents, reliability and race circumstances can change the order.
Exact position retention was much less common than the correlation alone might suggest: 67 of 431 pairs, or 15.5%, ended on the same number. Another 204 (47.3%) finished one or two places from their grid slot; 160 (37.1%) moved at least three. Mean absolute movement was 2.58 positions. So the aggregate relationship is strong, but a preview that treats each grid slot as a predicted finish would be too precise.
How far 2024 classified finishers moved from their grid slot (n = 431)
Driver Event Count
driver-event pairs; share of 431 numeric grid/finish pairs · All 24 Grands Prix of the 2024 F1 season
| Movement Bucket | Driver Event Count | Share Percent |
|---|---|---|
| Exact same position | 67 | 15.5 |
| Shift of 1–2 places | 204 | 47.3 |
| Shift of 3+ places | 160 | 37.1 |
Front positions often held, but not as an exact result
Across all 24 rounds, 96 of the 120 drivers who started in the top five also finished in the top five (80.0%). For the top ten, 201 of 240 grid positions resulted in a top-ten classified finish (83.8%). These pooled position opportunities are not individual-driver win probabilities: each grid spot appears once per Grand Prix, and all 24 races are combined. The counts show that front-grid placement usually remained inside its broad band, even when the exact order changed.
The relationship varied by event
Event-level correlations also differed. Monaco had r = 0.9740 among 16 paired classifications, while the São Paulo Grand Prix had r = 0.3707 among its 15 pairs. These describe within-event linear alignment only. They do not explain why one event moved more cars; the official tables provide final positions, not a causal record of weather, pit strategy, safety-car timing or mechanical problems.
Event-level correlation between starting position and finish in 2024
Pearson R Grid Finish
Pearson correlation r of numbered starting grid and final classified finish; paired classifications per event shown · 24 Grands Prix of the 2024 F1 season
| Round | Grand Prix | Paired Classifications | Pearson R Grid Finish |
|---|---|---|---|
| 1 | Bahrain | 20 | 0.8857142857142857 |
| 2 | Saudi Arabia | 18 | 0.8841622203517878 |
| 3 | Australia | 17 | 0.7559289460184544 |
| 4 | Japan | 17 | 0.9338761969006686 |
| 5 | China | 17 | 0.8617790203492923 |
| 6 | Miami | 19 | 0.8020489029654562 |
| 7 | Emilia Romagna | 19 | 0.9119657140188276 |
| 8 | Monaco | 16 | 0.9739766932676583 |
| 9 | Canada | 15 | 0.8234034387173558 |
| 10 | Spain | 20 | 0.9473684210526315 |
| 11 | Austria | 20 | 0.6406015037593985 |
| 12 | Great Britain | 18 | 0.9277605779153767 |
| 13 | Hungary | 19 | 0.787719298245614 |
| 14 | Belgium | 18 | 0.867448219855279 |
| 15 | Netherlands | 20 | 0.8466165413533835 |
| 16 | Italy | 19 | 0.8805666454439663 |
| 17 | Azerbaijan | 19 | 0.4093909502160067 |
| 18 | Singapore | 19 | 0.8837645909424906 |
| 19 | United States | 19 | 0.5834611516331515 |
| 20 | Mexico | 17 | 0.8523493779409855 |
| 21 | São Paulo | 15 | 0.37066175167122767 |
| 22 | Las Vegas | 18 | 0.8246282244697245 |
| 23 | Qatar | 15 | 0.7460470969529908 |
| 24 | Abu Dhabi | 17 | 0.44832193324804603 |
For a preview, the practical use is modest: the grid gives a measured starting baseline, and the past 2024 distribution shows how much that baseline shifted. A driver-specific forecast would need a separate model and assumptions tested on races beyond its training sample. This season-wide recount does not provide that validation. Compare it with our single-weekend practice-to-qualifying study, browse the Race Previews archive, or see how score rules change championship arithmetic in our points-system recount.
Method and sources
The source ledger contains 479 driver-event entries from official starting-grid and race-result tables. We excluded 48 entries without both a numbered grid position and numbered final classification, leaving n = 431. Position movement is grid position minus classified finish; mean absolute movement is 1/431 × Σ|grid − finish| = 2.58 places. Pearson r measures linear association in the pooled integer positions; R² is from a simple one-variable linear fit. Repeated observations from the same drivers and teams are not independent, and no causal or validated forecasting conclusion follows. Source records link each race’s exact starting grid and classification; see the Monaco starting grid, Monaco race result, São Paulo starting grid, and São Paulo race result. The official 2024 results archive indexes all events.