While most of the work I do on this blog is fairly serious analysis, this series has been an enjoyable, less serious side project. This is the final part of this five-part series, in which I apply the f1metrics model of driver and team performance to simulating historical hypothetical situations.
If you want to check out the previous articles in this series, they are linked here:
For each hypothetical scenario below, I simulated either the extension/comeback of a driver’s career or a change in teams during their career. Two types of outcomes are presented for the simulated scenarios:
(i) The adjusted driver performance rankings (in ppr) for each season, including the driver’s predicted hypothetical performance at that age and with that level of experience. This is a ranking (ranging 0-10) of absolute driver performances, taking teams out of the equation. In other words, it’s how the model predicts the drivers of that season would have performed in equal machinery. This is the same approach I use for my end of season driver rankings each year.
(ii) The predicted World Drivers’ Championship standings for each season. In cases where these tables are presented, I am including the effects of a driver’s hypothetical team on their points scored. In this simulation, I have either used the team’s average reliability that season to generate the number of mechanical DNFs (if it is a simulated comeback or career extension) or the target car’s reliability if it is a straight seat swap between two active drivers. To generate points tallies, I mapped the model’s scoring rate function to all other points scoring systems in F1 history so that I could convert accurately between ppr and points scored.
I couldn’t possibly finish this series without examining the career and potential of Gilles Villeneuve. Many fans consider Gilles their all-time greatest driver. He was a spectacularly gifted driver and a sheer entertainer behind the wheel. When his career was prematurely ended by a fatal crash in 1982, he was driving the season’s best car and possibly on course for a first title, despite friction with his teammate Didier Pironi and a less than ideal start to the season, How likely is it that he would have won the 1982 title? And what more could he have achieved beyond that?
To attempt to answer these questions, we can use the f1metrics model to make predictions. First, let’s take a look at how Gilles’ driving performances are rated in each season by the model, and where it predicts he would likely have ranked going forward, taking age and experience into account.
The model places Gilles among the top few drivers on the grid at the time he was killed. It doesn’t see him as likely to be the absolute best performer in any year, but between 1979 and 1984, he’s never far off. As he was 32 when he was killed, it is predicted that Villeneuve could have maintained a stable plateau of performance for several years to come, which could have included performances similar to 1979-1981, given the bounds of uncertainty.
Imagining that Villeneuve survived in 1982, we can use the model to simulate how he would most likely have performed in subsequent seasons, had he continued racing for Ferrari.
In 1982, Ferrari’s car was undoubtedly the class of the field. The team won the constructors’ championship despite losing both star drivers (Villeneuve to a fatal accident, Pironi to a career-ending injury), and as a result not even entering a car in 7 of the 32 available starts. With healthy drivers, Ferrari would clearly have taken the drivers’ title. Instead, Keke Rosberg squeaked to the title, winning only a single race along the way, in one of the most turbulent and bizarre seasons in F1 history.
At the time that he was seriously injured, Pironi was clearly leading the championship, having scored 39 points in 10 starts (his car was withdrawn at Zolder when Villeneuve was killed). John Watson was 2nd in the championship on 30 points, and Keke Rosberg was on only 23 points.
Villeneuve’s season, by contrast, had gotten off to a very poor start, with two DNFs, a disqualification, and a 2nd place when teammate Didier Pironi ignored team orders to overtake him for the win. He had scored only 6 points in those 4 starts when he had his fatal crash.
Applying the model, we can simulate the 1982 season with Gilles Villeneuve and Didier Pironi starting every race they missed, using the results they had already scored. In this case, Villeneuve is predicted to make up the deficit to Pironi across the season and narrowly win the drivers’ title. Keke Rosberg is a distant 3rd in the title race.
To simplify the hypothetical beyond 1982, I assumed that Pironi is no longer racing (although I note that Villeneuve would be predicted to outperform him as teammates in future seasons in any case). I simulated Villeneuve replacing Rene Arnoux (1983-1984) and Stefan Johansson (1985).
In 1983, Ferrari fielded Rene Arnoux and Patrick Tambay, finishing 3rd and 4th in the drivers’ championship, and winning the constructors’ title. The model rates the Ferrari the best car in 1983. Had Gilles continued racing for Ferrari, he would therefore have been in an excellent position to defend his hypothetical 1982 championship. Simulating the season, the model predicts Villeneuve as a likely 1983 title winner.
In 1984, McLaren had an unbeatable combination: the clear best car and the best overall driver pairing. Both Lauda and Prost doubled the points scored by Elio de Angelis, who finished 3rd in the championship. Racing for Ferrari, Villeneuve is predicted to perform better than either Alboreto or Arnoux managed with the car, but is still a clear 3rd in the championship.In 1985, Michele Alboreto mounted a serious title challenge in the first half of the season, which fell away with five mechanical failures in the last five races. The model predicts that Villeneuve could have extracted a bit more from the Ferrari than Alboreto, but still not quite enough to overcome the fundamental reliability issues to take the title from Prost.Beyond 1985, it would be difficult for Villeneuve to challenge the new elite, including Senna and Prost, especially if he remained at Ferrari, who were not seriously competitive again until 1990. In this alternate history, Villeneuve ends up with two championship titles (1982, 1983) and two near misses (1979, 1985). Ramifications of this timeline would include Keke Rosberg scoring no drivers’ titles, and Nelson Piquet scoring two drivers’ titles.
Michael Schumacher’s career is absolutely rife with potential alternative paths. I decided to look at three serious scenarios here, as well as one semi-serious scenario.
What if Schumacher had not retired in 2006?
This is probably the most often discussed hypothetical relating to Michael Schumacher. There are rumors that Schumacher was pushed into retirement before he was ready by Luca Montezemolo. While there were perhaps some early signs of age-related decline appearing in Schumacher’s driving, they were only noticeable relative to his near flawless performances earlier. In 2006, he was doubtless still one of the top drivers on the grid.
One theory about Schumacher’s retirement goes that Ferrari signed Kimi Raikkonen against Schumacher’s wishes and gave Schumacher the choice to continue racing alongside him, which Schumacher declined, allowing his protege Felipe Massa to retain a Ferrari seat. Had Schumacher continued at Ferrari into 2007-2008, he would have been 38-39 years old, but racing in a championship-contending team.
By the model’s estimation, Schumacher was still among the best-performing drivers on the grid, and would have remained so into 2007-2008, even as he began to experience age-related decline. If we simulate the 2007 championship with Schumacher in Felipe Massa’s place, the model sees Schumacher clearly leading the title race.
Continuing this hypothetical into 2008, when Massa narrowly lost the title to Hamilton, the model sees Schumacher as a slight championship favorite. The resulting Schumacher vs. Hamilton would have undoubtedly been epic.
In this hypothetical history, Schumacher would likely have taken his championship total from seven to nine, nearly doubling Fangio’s five titles!
What if Schumacher had continued racing into the hybrid era?
Schumacher’s comeback for Mercedes across 2010-2012 did not add to his 91 career wins. It could be viewed as unfortunately timed, considering Mercedes went on to become one of the most dominant teams ever just two years later. Part of Schumacher’s performance issue in 2010 was the lack of recent experience, especially in a climate with very limited testing compared to what he had become accustomed to during his F1 career. As he gained familiarity with the formula (especially the tyres), he appeared to close the gap to Nico Rosberg, with the two being closely matched in 2012. What if Schumacher had simply continued racing at Mercedes into the hybrid era in 2014 and beyond?
First, we have to discuss the issue of teammates. Schumacher leaving Mercedes in 2013 was a major catalyst for the team signing Lewis Hamilton. It is therefore plausible that Mercedes would have retained the Schumacher-Rosberg line-up into the hybrid era. On the other hand, Hamilton’s relationship with McLaren was clearly showing signs of strain, and might have been pushed beyond its limit by McLaren’s hopelessly uncompetitve MP4-28 in 2013. A move to Mercedes would certainly remain on the cards. I have therefore assumed in this scenario that Mercedes would field the dream team of Lewis Hamilton and Michael Schumacher into the 2014 hybrid era.
Second, we have to discuss the issue of age. Going into the 2014 season, Schumacher would have been 45 years old. Loss of performance at this age is significant and simply unavoidable. By the model’s estimation, Schumacher at this age would be no match for Hamilton in his prime, and would also be very likely to lose the championship if paired with Nico Rosberg.
Ultimately, there is no unwinding the hands of time, even for a driver who was as extraordinary as Michael Schumacher was at his peak.
What if Schumacher had raced at McLaren?
For a long time in the 1990s, Ron Dennis was eager to secure the services of Schumacher. Ultimately, Schumacher moved to Ferrari in 1996, along with many of the best technical talents from Benetton. But what if, instead, he had been lured to McLaren and spent the main segment of his career there rather than at Ferrari? This would potentially have set up a formidable combination of Michael Schumacher’s driving talents with Adrian Newey’s design talents. Newey moved to McLaren in 1997 after he fell out with Williams over lack of input into team decisions such as the hiring of Jacques Villeneuve and subsequent firing of Damon Hill. He stayed with McLaren until 2005.
Ferrari were in a massive rebuilding phase in the mid-1990s, led by Jean Todt. It is safe to assume that if they did not acquire Michael Schumacher, they would have gone after other top drivers, such as Damon Hill or Mika Hakkinen. In this hypothetical, I will assume that Hakkinen and Schumacher traded places, with Hakkinen moving to Ferrari in 1996 and Schumacher moving to McLaren.
In 1996, Schumacher famously dragged a disastrously ill-handling Ferrari chassis into an impressive 3rd in the championship. Irvine described the F310 as almost undriveable, scoring only 11 points. Overall, the model rates McLaren’s MP4/11 a slightly more competitive car than the Ferrari F310, in large part due to reliability — Ferrari had 13 mechanical failures in 32 starts, compared to 6 mechanical failures for McLaren. In a McLaren seat, the model thinks Schumacher may have been able to challenge Villeneuve for 2nd in the championship, but still could not have realistically beaten Hill. Hakkinen is predicted to score 18 points for Ferrari, dropping to 6th in the championship.
In 1997, Schumacher once again was fighting a dominant Williams team in an inferior car. This season, the model rates the Ferrari and McLaren as very closely matched, meaning Schumacher’s results are almost identical after a car swap.
In 1998, the first season that Newey had full input into car design, McLaren produced the dominant MP4/13, which Hakkinen took to the title in spite of some brilliant drives from Schumacher. With the best driver in the best car, the model sees the 1998 season as a walkover first title for Schumacher at McLaren.
The 1999 season was marred by Schumacher’s broken leg after a brake failure in the British Grand Prix, causing him to miss six races. It is difficult to know what to do with this case in a hypothetical history. Do we assume Schumacher still missed races at McLaren? Do we assume the brake failure happened to Hakkinen and caused him to miss races instead? Or do we assume it was a freak accident that would not repeat and give both drivers a full season? We are essentially free to invent whatever history we want here, so I have not run a hypothetical reconstruction of 1999.
Schumacher’s first Ferrari title came in 2000, when he narrowly beat Hakkinen to the title. The model considers McLaren to have had a competitive edge that season, meaning that when roles are reversed, Schumacher wins the title for McLaren with relative ease. Hakkinen’s performance that year is rated similar to Barrichello’s, meaning a close fight between the two Ferrari drivers.
By 2001, Ferrari had clearly the better performing car relative to McLaren. However, it was historically not Hakkinen’s strongest season either. With Schumacher in the McLaren, the model predicts that he would be the favorite to clinch the 2001 title.
At the end of 2001, Mika Hakkinen retired from F1, ceding his McLaren seat to the promising Kimi Raikkonen. In this timeline, Hakkinen would retire without having won the 1998 drivers’ title. Supposing that Raikkonen inherited Hakkinen’s seat at Ferrari for 2002, as he did at McLaren, he would be placed in the incredibly dominant F2002. Even Schumacher at the height of his powers could not have competed with this car. He is predicted to be only 3rd in the championship for McLaren in 2002, as Raikkonen takes his first title.
The 2003 season was a classic and controversial three-way fight between Ferrari, McLaren, and Williams. McLaren started the season very strong, while Ferrari uncharacteristically struggled until they introduced the new F2003-GA at San Marino. Ferrari then went on a winning streak until they began to seriously struggle with tyre wear as a heat wave struck Europe, which affected Bridgestone’s performance more than Michelin’s.
Williams-BMW had the most powerful engine, but suffered from excessive understeer until Michelin introduced a wider front tyre at Monaco. Williams were a clear front-runner from here until the final races, when this improvement was nullified by a protest claiming that Michelin’s tyres gained tread width as they became worn, forcing Michelin to introduce a new narrower tread tyre. Montoya’s championship challenge collapsed partly through this tyre change, but also through early and late season errors, and partly through an inability to beat teammate Ralf Schumacher sufficiently often. Ferrari’s strength in the final races allowed Michael Schumacher to narrowly clinch the title.
The model rates the three teams as closely matched in 2003, estimating that with equal drivers their ordering should have been Williams-McLaren-Ferrari. A seat swap between Schumacher and Raikkonen (all other factors playing out as they did) is therefore predicted to give Schumacher a clearer path to the 2003 title.
In 2004, the McLaren was unreliable and uncompetitive until the MP4-19B chassis upgrade in the second half of the season. Raikkonen historically managed only 45 points in the car. The model thinks Schumacher could only have done slightly better with it, while Raikkonen would romp to the title in the unbeatable F2004, in similar fashion to what Schumacher actually did that year.
In 2005, there were major changes to the tyre regulations, requiring tyres to last the entire duration of the race. This change was clearly intended to curb Ferrari’s dominance by hurting Bridgestone, and in that respect it was successful. The championship was made into a two-horse race between Michelin-runners McLaren and Renault. McLaren had developed the incredibly quick MP4-20 for the 2005 season. On pace, it was unstoppable in the hands of Raikkonen, but it suffered poor reliability. Alonso ultimately prevailed for Renault in the title fight. Schumacher finished a distant 3rd in the championship for Ferrari.
If given Raikkonen’s seat at McLaren, the model predicts that Schumacher would, like Raikkonen, have finished 2nd in the championship that year.
The first chapter of Schumacher’s career concluded in 2006, as he retired at the end of the year. It was a close championship battle with Fernando Alonso at Renault, with Renault starting the season with a clear advantage until their mass-damper solution was banned. The McLaren was the clear third-best team. Had Schumacher swapped seats with Raikkonen, the model predicts that he could have challenged Raikkonen for 2nd in the championship despite the slightly inferior machinery, but he would not have been in the title fight with Alonso.
Beyond 2006, we could imagine Schumacher continuing at McLaren into 2007-2008, as we did in the above hypothetical, but he could equally well have retired at this point, as he did in reality.
In the graphic below, I have compared the titles that Schumacher actually won for Ferrari with the titles that he could hypothetically have won for McLaren, had he spent the same 1996-2006 period there.
We can see from this hypothetical that a Schumacher-Newey alliance at McLaren would have had the potential to be nearly as successful as the Schumacher-Brawn-Byrne alliance was at Ferrari. In the hypothetical timeline, Schumacher would conclude 2006 with six drivers’ titles, while Raikkonen would already have two drivers’ titles.
What if Toyota had hired the other Schumacher?
The Toyota F1 team is the canonical example of money not necessarily buying success in F1. Due to a number of internal problems, they were one of the least effective manufacturer teams on a per-dollar basis in F1 history. As I showed in my previous economic analysis, Toyota spent over $3 billion in 8 years, with no wins to show for it.
One of the team’s most perplexing decisions was to hire Ralf Schumacher on a very high salary (reportedly around $19 million). It is often joked that the Toyota executives had accidentally signed the wrong Schumacher. What if Toyota had actually signed Michael Schumacher instead?
At the team’s best, in 2005, Toyota finished 4th in the constructors’ championship, with their drivers Ralf Schumacher and Jarno Trulli finishing 6th and 7th in the drivers’ championship, respectively. They scored 5 podiums and 0 wins. What could Michael Schumacher at his peak have realistically achieved with that car?
By the model’s estimation, the 2005 Toyota actually had a fair bit of promise with a top driver at the wheel. Notably, both Toyota drivers outscored the second Ferrari driver, Rubens Barrichello. The model rates the 2005 Toyota a better car than the 2005 Ferrari and predicts that Michael Schumacher would have scored more points in a Toyota seat than he did in his Ferrari seat that year, altough it would have required him adapting to Michelin rubber. To put it another way, the model thinks that Ralf’s 45 points scored for Toyota was closer to Michael’s 62 points scored for Ferrari than one would predict if the brothers were in equal machinery. Given the Ferrari seat in 2005, Ralf is predicted to score 23 points.
Where Toyota could have gone from there is unclear. In reality, they fell back to 6th in the constructors’ championship in 2006. But, with an all-time great driver to help galvanize the team, and 3rd place in the driver’s championship to show they seriously meant business, the entire team’s trajectory (at least up to the global financial crisis) could feasibly have been different.
Honorable mentions and notable omissions
Before concluding this series, I want to note that there there are a number of hypothetical cases I would have loved to include but didn’t for one reason or another. Most recently, I would have liked to include Jules Bianchi. He had a phenomenal junior driver career and I suspect he was destined for great things in Formula 1. Unfortunately, because he and Max Chilton only raced against each other as teammates in Formula 1, they form a disconnected pair, so the model cannot be used to estimate their performances against other drivers.
Going back to the 1950s, I was interested to explore the career of Stuart Lewis-Evans, who died aged 28 after an accident caused by mechanical failure. He was considered a very promising driver of the future. His qualifying records of 6-6 against Tony Brooks and 3-9 against the more experienced Stirling Moss indicate his impressive pace. But with only 6 counting races in his career (i.e., races without a non-driver DNF), it would be impossible to achieve any reliable predictions. The same problem applies to Ricardo Rodriguez, who has only 3 counting races. He debuted in Formula 1 at age 19 and likely had an excellent career ahead of him, were he not sadly killed aged 20.
There are of course many other drivers who had phenomenal results in junior single-seaters or other categories but never got a drive in F1. Those would be fascinating cases to explore, but at present I don’t have a reliable method for linking performances between series. That may be an interesting future project.
I hope you have enjoyed this series. I certainly learned a lot in putting it together!