Pick a Number, Any Number

In The Drunkard’s Walk: How Randomness Rules our Lives by Leonard Mlodinow, he introduces a pattern: 2,4,6. He says that you need to find out what the pattern is, but that he cannot tell you if you’re correct about the pattern, but he will tell you if you give him a pattern if that pattern matches the original pattern (1). Naturally, my brain turned to 8, 10, 12 as a guess for a pattern that would fit the original pattern, which he writes is correct, as is 20, 24, 30.

If you’re like me, you’re probably thinking the pattern is increasing even numbers. However, as he points out, did we ask about any patterns that are not even numbers? Or were we just confirming our initial hypothesis about even numbers? Did we even think to try to refute our initial hypothesis instead of confirming it? In fact, he writes, 1,2,3 would also fit the pattern, but how many people would have thought to ask that? (I didn’t)

So what does this have to do with genealogy? In genealogy we start with a research question, and then we set out to find information that will give us the answer. For example, Is John Smith the father of Robert Smith? We will look everywhere for evidence that John is the father of Robert. In the interests of reasonably exhaustive research (2), looking at all these sources that might point us toward an answer is important, but we must also consider the reverse question: are there sources that can show me that John is NOT the father of Robert?

Confirmation bias is deeply ingrained in us. While it helps to simplify our lives, we need to be aware of it since it can mislead us. This is especially true in the world of genealogy, where having our own pet hypothesis can get in the way of drawing correct conclusions. It’s important to remember that we need to work as hard to try to disprove our hypotheses as we do to prove them.


  1. Leonard Mlodinow, The Drunkard’s Walk: How Randomness Rules our Lives (New York, New York : Pantheon, 2008), 238.
  2. Jennifer Wiebe, “How many is enough?” Jennealogie (https://maltsoda.wordpress.com/2021/08/04/how-many-is-enough/: accessed 1 June 2022).

One response to “Pick a Number, Any Number”

  1. […] be foolish to ignore the existence of bias” (28) is a hallmark of both forensic science and of genealogy […]

Leave a reply to The Sydney Declaration for Genealogists: The Genealogist as Scientist | Jennealogie Cancel reply