The biggest comebacks in F1: Antonelli at Monza is the second biggest winning drive ever
F1 Legend Sim · Staff · 7 September 2026 · 7 min read
Kimi Antonelli won at Monza from nineteenth on the grid. We counted the places every driver has gained between the grid and the flag across the 1,150 Formula 1 races held since 1950, and that drive is the second biggest comeback in history among those that ended in a win. Only one sits above it, and it is from 1983. But the story in the data is not that Antonelli did something big: it is that this keeps happening less and less.
The ten biggest winning comebacks
Winning from the back is rare. In 1,150 Grands Prix there have been 25 wins with ten or more places gained, roughly one race in 46. These are the ten biggest:
| Driver | Starts | Gained | Grid | Classified |
|---|---|---|---|---|
| John Watson · 1983 · United States GP West | P22 | +21 | 26 | 13 |
| Kimi Antonelli · 2026 · Italian GP | P19 | +18 | 22 | 19 |
| Rubens Barrichello · 2000 · German GP | P18 | +17 | 22 | 12 |
| John Watson · 1982 · Detroit GP | P17 | +16 | 26 | 11 |
| Kimi Räikkönen · 2005 · Japanese GP | P17 | +16 | 20 | 16 |
| Max Verstappen · 2024 · São Paulo GP | P17 | +16 | 20 | 15 |
| Jackie Stewart · 1973 · South African GP | P16 | +15 | 25 | 11 |
| Michael Schumacher · 1995 · Belgian GP | P16 | +15 | 24 | 14 |
| Fernando Alonso · 2008 · Singapore GP | P15 | +14 | 20 | 15 |
| Olivier Panis · 1996 · Monaco GP | P14 | +13 | 22 | 7 |
| Max Verstappen · 2022 · Belgian GP | P14 | +13 | 18 | 18 |
“Grid” is how many took the start and “classified” how many were still ranked at the end. Both columns are there for a reason, explained below.
Those two right-hand columns are what make the table honest. Watson won in 1983 from P22 of a 26 car grid of which only 13 were classified. Antonelli gained 18 places in a race where 19 of 22 finished. The second is a great deal harder than the first, and the headline number hides it.
And yet the comeback is dying out
This is the part we did not expect when we started counting. We looked at the share of each decade’s races in which somebody gained ten places or more, and next to it, out of every hundred cars that took the start, how many were classified:
| Decade | Races | With a 10+ comeback | Biggest | Classified |
|---|---|---|---|---|
| 1950s | 74 | 59% | +26 | 51% |
| 1960s | 99 | 49% | +17 | 54% |
| 1970s | 144 | 88% | +22 | 54% |
| 1980s | 156 | 95% | +22 | 48% |
| 1990s | 162 | 77% | +20 | 53% |
| 2000s | 174 | 52% | +19 | 70% |
| 2010s | 198 | 46% | +21 | 83% |
| 2020s | 143 | 34% | +18 | 88% |
In the eighties, 95 races in every 100 featured a comeback of ten places or more. This decade the figure is 34. And the right-hand column says why: in the eighties 48% of starters were classified, and today 88% are.
Put another way: half the grid broke, and whoever survived inherited ten places without passing anybody. The great eighties comeback was in large part somebody else’s unreliability. Today almost everyone finishes, so places have to be taken on track, which is far harder and happens far less.
The relationship holds from 1970 to today, decade by decade, without a single exception. Where it does not fit is the fifties and sixties: just as few cars were classified and there were still fewer comebacks than in the seventies. We have no measured explanation for that and would rather say so than invent one, though the irregular size of those grids and the number of races per year suggest sample size matters more than the phenomenon.
Where this table could fool you
Two filters change the result completely, and both are applied.
First: the Indianapolis 500 is excluded. It counted towards the championship from 1950 to 1960, but it was a different race, with different cars, different drivers, an oval and 33 starters. With it included, eight of the fifteen biggest comebacks in Formula 1 history turn out to be races no Formula 1 driver entered. That is 11 editions set aside.
Second: if you did not take the start, you did not come back. Those rows carry a grid position of zero, which is not a pit lane start, it means the car never started. Without filtering them, the ranking fills up with comebacks from a grid slot that never existed.
And this is where our own simulator falls short
We have measured how far a good car works its way up from the back, circuit by circuit: we took the six best cars, put them last and counted the places they recovered, across 11 circuits with the starting order forced. The answer is that they recover 15 places at Monaco and 15.2 at Monza.
Two tenths of a place between the circuit where nobody overtakes and the easiest one on the calendar is, quite simply, wrong. Our engine computes pace and sorts, and it has neither walls inches away nor a single straight where the slipstream decides. It knows a fast car ends up in front, and it does not know that at Monaco it does not. It is the same limitation we wrote about in what is pole position worth?, seen from the other end of the grid.
That Antonelli’s comeback happened at Monza rather than Monaco is not a calendar detail. It is half the explanation. We measured the other half on Saturday, when we found that at that circuit the grid slot is worth two places and the car is worth five.