How Common is Your Birthday? This Visualization Might Surprise You

By Matt Stiles | | Topics: Demographics, News

It’s baby season in America, with September the busiest month for births on average in the last two decades. So it seemed like the right time to remix this blog’s most-popular post: How Common is Your Birthday?

That old heatmap, which highlighted specific dates for popularity, has been viewed more than 500,000 times here and published across the web. But it was flawed, namely that it used ordinal data (birthday ranks by date) rather than continuous data (actual births counts by date). This graphic finally addresses that problem:

This new version of the heatmap, which is interactive on larger screens, uses births by day from 1994 to 2014, thanks for the fine folks at FiveThirtyEight, whose reporters posted the tables on Github for an unrelated project. It lists the average number of births by day, the rank (number one is most common) and an estimated — if slightly irresponsible — conception date.

Some highlights:

  • Holidays: People generally seem to have time for baby making during their time off. Several of the most-common birth dates, in September, correspond with average conception periods around Christmas. Sept. 9 is most-common in this dataset, though other days in that month are close. Sept. 19 is second. Following a customary gestation period, many of these babies would, in theory, have been conceived on Dec. 17 and 27, respectively.
  • Choice: Clearly, some people choose when they have their children. While they’re making babies during the holidays, many people aren’t really having them then. The least-common birthdays in this dataset were Christmas Eve, Christmas and New Year’s Day. Dates around Thanksgiving aren’t as common. July 4 is also at the bottom of the list. Conversely, Valentine’s Day ranks relatively high, as you can see in the graphic, as are the days just before a new tax year begins.
  • Skewed: There are some fun patterns in this data, but the difference between your birthday — unless it’s on a truly rare day — isn’t that much different than a top date in September. There a left-tailed skewness to the data, which ranges from 6,500 births per day to more than 12,000. The median number of births per day, though, is around 11,000. The most-common day had 12,300 births, on average. More on the data distribution soon.

You can have a look at the data for yourself in this table:

Chances are you came here looking for the legacy birthday graphic. If you must, find it here, along with a related post about the flawed methodology.

Meanwhile, check out your birthday, share your thoughts in the comments — and tell the Internet to do the same.

Credits: Thanks, Cynthia Brewer, for your wonderful color guide. After trying several of your sample palettes, I couldn’t decide. So my three-year-old daughter — who celebrates a birthday Sept. 22 — chose this one. It looked the most like a birthday cake, she said. Also, thanks to for the cake image.

4 thoughts on “How Common is Your Birthday? This Visualization Might Surprise You

  1. This is accurate. I was unfortunate enough to have been told the actual day of my conception, and the chart says I should have been conceived just the day before. Given all the variables in play, one day off is incredibly accurate.

  2. The data is still somewhat biased because of the irregular sample size. In order to have completely unbiased data you need to have a number of samples such that each day falls on a particular day of the week the same number of times as any other day.

    By looking at the graph there are clear spikes on April 4, 11, and 18. We know this is a cause of this bias because:

    For example, the day April 11th, from 1994-2014, falls on:
    Monday 3 times (1994, 2005, 2011)
    Tuesday 3 times (1995, 2000, 2006)
    Wednesday 3 times (2001, 2007, 2012)
    Thursday 3 times (1996, 2002, 2013)
    Friday 3 times (1997, 2003, 2008, 2014)
    Saturday 2 times (1998, 2009)
    Sunday 3 times (1999, 2004, 2010)

    It has more occurrences on a Friday and less occurrences on a Saturday compared to the other days of the week. Because of other studies showing that weekend birthdays are less common than weekday ones, we can conclude that the lowered frequency of weekend dates for the days April 4, 11, and 18 are causing a spike.

    In order to have unbiased data you need a sample size of a 28-year span (e.g: 1989-2016) in which each day of the year falls on each day of the week exactly 4 times, expect Feb. 29 in which it falls on each day of the week exactly once.

Leave a Reply

Your email address will not be published. Required fields are marked *