In May of this year I attended a conference of humanist organizations in Atlanta, Georgia where I had a conversation with one of the local organizers. She told me she had a brother who was living in Hawaii but considering a move to Los Angeles, where I live, sometime later in the year and asked if she could pass my phone number on to him.
I had forgotten about the exchange until last week when I got a call from an unknown 808 area code number. The young man on the other end of the line explained who he was and how he had my phone number. We chatted briefly and I found out he and his wife had arrived in LA, they were looking for a place to rent and we made a date for lunch with a couple days later.
As we got to know each other over lunch, I learned that they knew nothing of the organization his sister and I are both affiliated with, so I told him how it was I came to meet her. Then they asked about my family and I explained that I had two children, one not much different in age from them, who had graduated from a small college in Minnesota in 2014 with a degree in Classics.
The young woman interjected, “Your daughter didn’t happen to go to Carleton College, did she?” Which, if you’re listening to this podcast, you can already guess what my answer was. Listeners should understand that Carleton is a college of 2,000 students in rural Minnesota. This young woman explained that her childhood best friend from growing up in Houston, TX had graduated from Carleton the year before in 2013, also in Classics, a department of about a dozen students.
I texted my daughter and my lunch companion texted her childhood friend to ask if they knew each other only to find out that the two of them had been study buddies through ancient Greek language for the 3 years they overlapped and are still close friends.
And by this last weekend, they had found a place to live: they will be renting from my wife and me starting in a couple weeks.
Seriously? The odds of this must be crazy!
Below are the extended notes provided by statistician and podcaster Kyle Polich for use in Skepticality Episode 264. Take a look and leave your comments below. Also, please be sure to listen to the podcast for our own hilarious commentary.
(Kyle studied computer science followed by artificial intelligence in grad school with a focus in probabilistic reasoning and planning. His general interests range from obvious areas like statistics, machine learning, data viz, and optimization to data provenance, data governance, econometrics, and metrology. He enjoys exploring the intersection of statistics and skepticism and sharing related insights with others including through his podcast Data Skeptic. Visit Kyle’s blog Data Skeptic, and give the podcast a listen.)
So this story covers a series of seemingly unlikely events. Let’s try and break them down and isolate the parts that are not surprising from the parts that are eyebrow raising.
One of the important lessons here is around *conditional* probability. What is the probability that a person can play “Mary Had a Little Lamb” on the bassoon? Pretty low! What about the probability of that given the fact that they’re a professional bassoon player – very high!
To begin with, our listener is out of town chatting with a conference organizer who mentions her brother is moving to the listener’s city. There are almost 40k municipalities in the United States, so shall we say the odds of this are 1 in 40k or 0.0025%? Not quite.
Let’s consider the complement of this situation. Imagine you meet someone and proudly announce “I’m from Los Angeles”, to which they reply, “Cool! I have a good friend that lives in Gainsville, FL!” I mean, that’s nice, but I’m from LA. I think it’s fair to say our investigation only starts *conditioned* on the fact that a common city comes up in conversation.
Moving ahead to the part of the story in which the listener meets the relocating young brother and wife, and mentions having a daughter who attended a small college in Minnesota in 2014 with a degree in classics. The young women mentions having a close childhood friend who studied the same subject in a Minnesota school, and asks if they might perhaps have attended the same school and know each other. There are almost 200 colleges and universities in Minnesota. I’m not sure what qualifies as small, but if half of them are considered small, we can call those odds about a 1% chance.
Setting aside how many childhood friends the young woman had and how many universities they spread out into, maybe we call these odds 1 in 100 chance. That’s like betting on a specific number for rulet and winning. Unlikely, but not extraordinary.
But now we get into *conditional* probability. What are the odds that two students at a small school in a small department of about a dozen students know each other? I should hope pretty high!
So all in all, I find this one noteworthy, but not excessively surprising, and if I had to put a firm number on it, I’d say in the neighborhood of 1% likelihood.
The US Census tracks state to state movement. Kyle put together a fun, interactive data visualization that allows people to select a state and see the percentage of people that leave that state and what other states they migrate to.