31 August 2008

Some Chess Players Fib About Their Birthdays

Continuing with More About Chess Players' Birthdays, the following diagram, besides revealing my subpar skills charting Excel data, tells us something about the distribution of chess players' birthdays. The data is taken from the 2005 FIDE rating list, the last list I have that included date of birth. Subsequent lists had only year of birth.

Taking the 47661 players with birthdays on the 2005 rating list, and using 365.25 days in a year, I calculate that an average of 130.5 players were born on any random day. The chart is a histogram showing how the days of the year, grouped by the number of players born on a particular day, cluster together.

For example, the tallest bar on the chart is the black line that peaks at 52. You'll have to take my word for it, but it represents the value 130 on the horizontal axis. It means that there were 52 days in the year where more than 125 players, but less than 131, were born on that day (125 < x < 131, where x is the number of players born on a particular day). The blue bar to the right of it peaks at 49, and means that there were 49 days where more than 130 players, but less than 136, were born on a certain day. My own birthday falls on 14 November. On the FIDE list there were 109 other players having the same birthday and our common day clusters into the second blue bar from the left.

Now let's consider the extremes on the chart. If you look carefully, you'll see that there is a small blip at greater than 320. That bar, at 325, represents the number of players who have given 1 January as their birthday. Since it's very unlikely that so many players were born on the first day of the year, with that day so far off the chart, it's safe to conclude that these players either don't know their real birthday or don't want to reveal it.

Looking at the leftmost blip on the chart, it stands for 29 February, when 22 players were born. I find it curious that, even after multiplying that number by four, it is still less than the second least frequent day, 7 October when 97 players were born. It is more likely that some people prefer not to give their birthday as 29 February, but give instead 1 March, which is indeed one of the top-10 most frequent days on which chess players claim to be born.

Also curious is the second blip from the right, representing 1 July with 187 players. It means that the second most frequent birthday -- significantly higher than the third most frequent day, 2 February with 169 players -- falls exactly midyear. I have a non-chess playing friend who was born in India. His exact date of birth wasn't recorded, but his mother remembers that it was 'during the monsoon season'. He has adopted a birthday according to his mother's memory. I suspect that a non-trivial number of players have given their birthdays as 1 July, because they only know it was around midyear.

The top-10 most frequent birthdays include 1 January, 1 March, and 1 July. What about the other first days of the month? Three of them also fall in the top-10 most frequent birthdays and another misses it by a whisker. I find this all very curious and conclude that some players, for whatever reason, simply fib about their birthdays.

What does all this have to with chess? Not much, I admit, except to show once again how the small world of chess reflects the larger world around us.


Going back to my post Chess Players Prefer to Be Born in February,
a consequence of this current post is that the earlier number of chess players born in January is overstated. The number for 1 January skews the result.


Note: Listed under 'CHESS CULTURE' on Chess Blog Carnival 1/09.

No comments: