Friday, 31 January 2014

CAN GREAT DRIVERS BE IDENTIFIED IN THEIR ROOKIE SEASON?

The story of how fast Michael Schumacher was on his debut race for Jordan-Cosworth in the 1991 Belgian GP at Spa is well-known. Instant and exceptional speed. The same went for Ayrton Senna for Toleman-Hart in 1984. These two future top-raters and champions were noticeably faster than their team-mates. They were obviously special. Lewis Hamilton debuted for McLaren-Mercedes in 2007 and performed at the same level as his established, twice-World Champion team-mate, Fernando Alonso, who was into his sixth season.
Senna 1984
Schumacher 1991
Vettel 2007

I collated the debut season driver-ratings as scored on my Rating System to determine how often this ‘instantly-fast’ rookie phenomenon occurred. My 1894-2013 spreadsheet and season tables showed that the majority of star-drivers shone in their first season.


'SAMPLE SIZE'

First I highlighted in my season tables those drivers who attained the ultimate driver rating of 100.0 or attained the highest rating in that season/era. I then went back to their rookie season to see what their rating was then. There were 48 ‘star drivers’ from 1894-2013, as decided by their racing achievements and acknowledged greatness and by my Rating System (which measures them throughout their career, not only their rookie season). Thirty-two of 48 great drivers (67%) started within the 100.5 to 100.8 bracket; 16 of the 48 (33%) were faster (between 100.0 and 100.4), and only one, Lauda, was much slower than all here (for example, in his first full season, in 1972, he driver-rated at 101.6).


METHODOLOGY

My workings and calculation method were not done retrospectively. Each season was treated in isolation; the drivers were measured and scored against their team-mates and peer rivals within each season. Of course, my driver selection now is done retrospectively with hindsight, knowing who the great drivers have been.

My rookie-season driver-rating stats have been surprisingly even over the 120 years of grand prix racing history from 1894 to 2013.


Caracciola 1926
Chiron 1927
Rosemeyer 1935

Bear in mind that the fastest, benchmark driver-rating is 100.0 on my System, and that 100.1 is 0.1% slower, 100.2 is 0.2% slower, 101.0 is 1.0% slower and so on. For 2013, my Rating System scored Alonso and Vettel equal top-rated at 100.0. Some Champion Drivers did not rate at the ultimate 100.0, such as Damon Hill, Jacques Villeneuve who at their peaks scored 100.5. Jack Brabham, Graham Hill and Emerson Fittipaldi also peak driver-rated at 100.5. Hawthorn, Scheckter and Jones peaked at 100.4. Hunt’s best was 100.3, while Mansell and Button peak driver-rated at 100.2, Hakkinen at 100.1. My System measures and scores drivers based on comparative speeds only, and takes no account whatsoever of the official championships, points or placings.

ROOKIE DRIVER-RATINGS FROM THE FORMULA ONE ERA (1950-2013)

Taking the greatest drivers of the Formula One era, 1950-2013, who debuted at similar driver-rating figures by my System:

· Alberto Ascari debuted in 1940 in a privately-owned Maserati to driver-rate at 100.7.
· Fangio’s first major Grand Prix appearance was at age 36 in 1948 driving for Gordini-Simca when he driver-rated at 100.7.
· Moss started aged 21 in 1951 to driver-rate at 101.0, but he only appeared twice; he then rated at 100.8 in 1952 when he competed more often.
· Hawthorn rated at 100.8 in 1952 when he drove a well-prepared Cooper-Bristol.
· Surtees scored 100.7 in 1960 driving a Lotus-Climax for just four races.
· Stewart started at 100.8 when debuting as team-mate to the experienced and competitive Graham Hill for BRM in 1965.
· Andretti on his first appearance equalled Lotus-Cosworth team-mate and twice-World Champion Graham Hill in 1968, to rate at 100.5.
· Gilles Villeneuve only raced once in 1977, but for his first full season with Ferrari in 1978 scored 100.8.
· Prost famously outpaced his fast and experienced McLaren-Cosworth team-mate Watson on his debut in 1980, rating at 100.5.
· Senna scored 100.7 in his debut season for Toleman-Hart in1984.
· Schumacher driver-rated at 100.8 in 1991 for Eddie Jordan’s new Jordan-Cosworth team.
· Hakkinen driver-rated at 100.7 for Lotus-Cosworth in 1992.
· Raikkonen driver-rated at 100.7 debuting for the Sauber-Ferrari team in 2001. He was slightly faster than his promising and more experienced team-mate Heidfeld in 2001.
· Alonso shone at 100.7 at the same driver-rating as Raikkonen. This despite driving a slow, backmarker Minardi-Cosworth in 2001.
· Hamilton amazed everyone at his 2007 debut by equal-rating at 100.2 with his illustrious McLaren-Mercedes team-mate Alonso. He was probably the best-prepared F1 driver, having been coached and supported by the McLaren team since his teenage years and spent many pre-season hours in the simulator. An exceptional performance.
· Vettel was first noticed by the media when he won the wet 2008 Italian GP for Toro Rosso-Ferrari. However my System driver-rated him a season earlier in 2007 at 100.7. This was despite him driving only the last seven races of the season.

These are 16 of the greatest Formula One drivers, all of whom went on to top, driver-rate and /or score World Championships. At the time many others driver-rated at about the same figures of 100.5 to 100.8, but none were rookies.

Clark 1960
Lauda 1971
Piquet 1978

The only other really great F1 drivers who do not appear on the above list are Clark, Lauda and Piquet. This is because their rookie-season driver-ratings were slower than the 100.5-100.8 range that is the norm for great drivers in their rookie seasons. Below I offer some possible explanations for these three surprising cases.

To compile the above list, I was working strictly to a narrow range of driver-ratings (of 100.5-100.8, as derived from the data) as scored on my tables and spreadsheet to within one-hundredths of a second differences. That is why these three were left out.

There are however explanations for two of these outliers which show that they did in fact conform to the fast-starting rookie driver-ratings.

· Clark debuted at a driver-rating of 101.0 for his six of nine appearances in 1960. He drove part-time for Lotus-Climax. Clark was occasionally cautious (or sensible) as a person, and was up against three faster and more experienced drivers in same cars for 1960: Moss, Surtees and Ireland. The next year 1961 Clark raced at 100.5.

· Lauda debuted in 1971 for just one race in an ill-prepared March-Cosworth and driver-rated at a very slow 104.9! He improved in 1972 after a full season with the March team to driver-rate at 101.6. Still no sign of greatness. In 1973, for BRM, Lauda’ technical and applicational skills started to show as he scored 100.7. This was a uniquely slow start for one who was to join the greats!

· Piquet raced just four times in 1978, three in an old-model McLaren-Cosworth and once in a Brabham-Alfa Romeo, to driver-rate at 101.8. In 1979 Piquet drove full-time for Brabham and driver-rated at the future-star ‘norm’ of 100.7.

None of these ratings was calculated or compared against drivers of other eras, nor against each other. I scored each season-by-season against their peers. Yet the results have been amazingly similar; including Clark and Piquet, 14 of the 16 debutants scored between 100.5 and 100.8.

This is a surprisingly close range considering the differences in fields, cars, technologies, races and formats over the 64 Formula One seasons.

The spread of 0.3% represents 0.3-second in a qualifying lap or 18-seconds in a 60-lap race of 100-minutes. Additionally these drivers were usually not front-runners and raced further back, some in the midfield.

What accounts for this level of accuracy is that my System measures and scores each driver against the fastest driver of each season, i.e., those that top-rate at or close to the stable, benchmark factor of 100.0. Only two of the above-listed 16 stand out in their rookie season performances: Hamilton faster at 100.2 and Lauda much slower.


ROOKIE DRIVER-RATINGS FROM THE PRE-FORMULA ONE ERA (1894-1949)

Going back to Grand Prix beginnings, in 1894, the same 100.5 to 100.8 pattern is found.

Charron 1898
Fournier 1901
Nazzaro 1904

Among the over 30 instantly-fast rookies are such greats as: Charron 1897, Levegh 1899, Nazzaro 1904,Wishart 1909, Anderson 1911, Andre Boillot 1919, Antonio Ascari1923, Benoist 1924, Caracciola 1926, Chrion1927, Nuvolari 1928 and Rosemeyer 1935. All debuted within this fast and very close 100.5 to 100.8 range.
         
Ralph de Palma 1908
David Bruce-Brown 1901
Antonio Ascari 1923

THE FASTEST DEBUTANT DRIVERS 1894-2013

Hamilton’s debut driver-rating of 100.2 stands out as the fastest among the 16 rookies of the Formula One era 1950-2013. However, in the earlier grand prix era, from 1894-1949, several debut drivers started at even faster rookie driver-ratings than Hamilton did: Henri Fournier in 1901, Théry, Albert Clement and Lancia in 1904, Baras and Salzer in 1906, de Palma in 1908, Sailer in 1914, Bordino in 1921 to Arcangeli in 1930.

Doubtless there are reasons for each, such as prior experience in highly-competitive prior grand prix-type races or classes below Grand Prix level, much as Fangio had done. As well of course as superior talent and aptitude.


SIMILAR DRIVER-RATINGS

The fact that two-thirds of these 48 greatest drivers (67%) started their careers at a similar ‘gap’ from the fastest, established stars demonstrates that my Rating System is more than ‘just personal opinion’. It indicates that this must be a fairly accurate method of driver-rating.


CONCLUSION

QUESTION:

mds wrote:

Patrick, as always, a great read and valuable information. Thank you for that.

However I don't think the starting question is answered in your blog post. The question that is being answered is more "Were the greats also fast as a rookie?". It takes the later periods of driver ratings as a starting point, then looks back at their starting seasons. To really answer the question "Can future stars be identified?" a correlation could be made starting from the rookie rating of each driver, and then go on to see if the fastest go on to become stars/greats.

Maybe an idea for a future blog post?

ANSWER:

Thanks MDS and Fiki, you are both right: I did not answer my title question because I can't.

Drivers such as Alesi and J Villeneuve came in at a high performance level, but did not go on to become 'great' or top-rated for various reasons.

What surprised me was the fair consistency of the rookie season driver-ratings for about two-thirds of the 'great' drivers.



© Patrick O’Brien. Nothing from this page can be used without the permission of Patrick E. O’Brien.

***

Sunday, 12 January 2014

MY RATING SYSTEM - AVAILABLE TO BUY!

After 12 years of development and refinement from its 2002 inception, my Rating System is being published.

I have started with the Formula One era, 1950-2013, and divided it into decades for practical purposes. The first book contains the 2000-2009 seasons, is 80 pages, soft-cover bound and available now online. A section entitled ‘Guidelines for interpreting my Rating System’ is included which briefly explains my System.

Each season is in three sections, Package, Driver and Car, with the whole field rated and tabulated, and text discussing mainly the winners and the front-runners. My own illustrations are used, and do not necessarily show the fastest, top ranked competitors, but rather a spread for each season. Of course it would be ideal to use the many excellent journalistic photo-images, but costs would be prohibitive.

I have divided the Formula One era into seven decades as follows:

Decade
Seasons
Winners and front-runners
1
1950
1959
Fangio vs Ascari to Hawthorn & Moss.
2
1960
1969
Clark vs Surtees, Graham Hill, Jack Brabham & Gurney to Stewart.
3
1970
1979
Stewart vs Rindt, Peterson & Fittipaldi to Andretti & Reutemann.
4
1980
1989
Jones vs Piquet through Lauda to Senna vs Prost.
5
1990
1999
Senna vs Schumacher to Schumacher vs Hakkinen.
6
2000
2009
Schumacher vs Montoya, Raikkonen & Alonso to Button.
7
2010
2013
Vettel vs Alonso, Hamilton, Raikkonen & Button.

Each ten-year period will be published separately. The sixth decade (2000-2009) is now available – see ‘Buy my Rating System’ above or click on the link:

For those who are interested further, additional explanations include:

Chapter 1: Introducing the problem and implications of my Rating System.
Chapter 2: Literature Review.
Chapter 3: Methodology.
Chapter 4: Findings (Season Summaries).
Chapter 5: Conclusion (replicability of my Rating System).

Chapters 1, 2, 3 and 5 may also be published.

© Patrick O’Brien. Nothing from this page can be used without the permission of Patrick E. O’Brien.

***