Adjusting for the number of races in a season

The number of races in a season increased fairly sharply from 7 in 1950 to 17 in 1977 before stabilising for the next quarter of a century.  The last decade or so has seen a further increase in the number of races as F1 management attempts to maximise revenues from the sport.  There is ultimately a ceiling on the number of races given the need to physically transport the teams and equipment around the world to compete, although there are 52 weeks in a year and it is not clear exactly where the end point will be.

Initially we decided it was only fair to adjust our stats for the number of races in a season in order to standardise the results.  For example it’s hard for 1950 to compete with 1982’s 11 unique race winners when there were only 7 races held in the season.  In the end we decided that the raw stats were preferable for a number of reasons.  Firstly, they’re just plain simpler – and in general simple is good.  As Einstein said, an explanation should be as simple as possible, but no simpler.  Secondly, if we decided to normalise the results we would probably have needed to introduce another factor based on the number of races in the season on the assumption that for most fans, more races is better than less.  So rather than adding complication by normalising and then reducing the impact of the normalisation by adding another factor that worked in the opposite direction, we decided to keep it clean and let the raw numbers tell the story.