What Makes The Formula One Champion. Regression Analysis
1 like1,269 views
The document analyzes the factors contributing to success in Formula One racing through regression analysis. Key variables include driver salary, team dynamics, and engineering experience, with findings indicating the influence of these factors on race outcomes. The model achieved limited predictive accuracy for team standings, correctly identifying only some positions in the final rankings.
What Makes The Formula One Champion. Regression Analysis
- 1. What Makes The Formula One Champion? Regression Analysis Wojciech Latocha
- 2. WHAT IS FORMULA ONE? • the highest class of single-seater auto racing • the season is compounded of a series of races, known as Grand Prixes • races are held on purpose-built circuits and public roads • results of each race are added to define two annual World Champions: • individual • constructor
- 3. THE IMPORTANCE OF THE F1 ANALYSIS • Social factors: • Worldwide popularity • each F1 Grand Prix attracts a global audience of 600 million people • 2001 F1 season: 54 billion TV viewers, 200 countries • Economic factors: • Perfect example for Snob Effect and Veblen Good concept • Formula One revenues are equal to GDP of Mongolia or Kyrgyzstan and almost two times higher than GDP of Faroe Islands
- 4. LITERATURE REVIEW • Numerous general nature book publications and articles: • Explaining rules • Presenting history • Describing former World Champions • Liu Xiao: Formula One’s Financial Crisis, 2009 • A few scientific resources: • Eichenberger and Stadelmann, Who Is The Best Formula 1 Driver? An Economic Approach to Evaluating Talent, 2009 • Mastromarco and Runkel, Rule changes and competitive balance in Formula One motor racing, 2009
- 5. DATA • The official website of the Formula One • Official websites of Formula One teams • Official websites of drivers • Formula Money
- 6. METHODOLOGY • Regression Analysis • Ordinary Least Squares Methodology • STATA software
- 8. SIMPLE DATA SERIES • Was the individual world champion racing for the team? – a dummy variable • Number of years that a main engineer spent with a team
- 9. RATIOS • Percentage spread between drivers' points • First driver salary / First driver podiums • First driver salary / First driver positions awarded with points (excluding podiums) • Second driver salary / Second driver podiums • Second driver salary / Second driver positions awarded with points (excluding podiums) • Teams' testing laps / The most active tester's laps • Pit-stop spread
- 10. REASONS OF THE DISTINCTION • Changes in regulations • Forecasting disability of some simple series (no ability of their value's control) • Driver’s transfers • Number of competing team’s changes
- 11. SUMMARY OF STATISTICS Standard Variable Mean Deviation final position 6.018182 3.223128 was the individual world champion racing for the team? 0.090909 0.290129 percentage spread between drivers' points 0.692449 0.341747 number of years main engineer spent with the team 2.927273 2.417946 teams' testing laps to the most active tester's laps 0.69628 0.270442 pit stop spread 0.187378 0.193195 1st salary/podiums 4.589197 6.190581 1st salary/points 2.642526 3.218502 2nd salary/podiums 3.351749 5.292386 2nd salary/points 1.857905 2.539029
- 12. RESULTS OF THE REGRESSION final position Coefficient Standard Error t P>t was the individual world champion racing for the team? -1.86251 1.232371 -1.51 0.138 percentage spread between drivers' points 2.287651 .9805094 2.33 0.024 number of years main engineer spent with the team -0.27952 .138966 2.01 0.05 teams' testing laps to the most active tester's laps -1.88072 1.584256 -1.19 0.241 pit stop spread 2.643283 1.948494 1.36 0.182 1st salary/podiums -0.07556 .0665366 -1.14 0.262 1st salary/points -0.1668 .1104518 -1.51 0.138 2nd salary/podiums 0.195983 .0838879 2.34 0.024 2nd salary/points -0.43651 .1611728 -2.71 0.01 constant 7.177511 1.498906 4.79 0
- 13. DISCUSSION • 7.18 - an initial position of any team competing in The Formula One • 1.86 - a position improvement due to employing a former individual World Champion • 0.23 - a result depreciation caused by each 10% of a difference between points collected by each of two drivers • 0.28 - a position increase through prolonging team’s cooperation with its main engineer • 0.44 - a position improvement due to each $1 million paid to the second driver for scoring a position points-awarded • 0.2 - a position drop caused by each $1 million paid to the second driver for scoring a podium
- 14. FORECAST FOR 2011 Virgin-Cosworth HRT-Cosworth Lotus-Renault Williams-Cosworth STR-Ferrari Sauber-Ferrari Force India-Mercedes Renault Mercedes Ferrari McLaren-Mercedes RBR-Renault 0 2 4 6 8 10 12 actual final position predicted position
- 15. CONCLUSION • The model was able to predict an exact final position of one out of twelve teams • Numerical values returned by the equation indicated four out of twelve positions in the final standing, including: • The World Champion • All three teams on the podium • The forecast indicated a fierce battle for a few positions – what was proved on the track • The model underestimated a result of the first seven teams and overestimated the result of the others
- 16. Thank you for your attention!