Tag Archive: probability


(Submitted by Skepticality listener Katharine Shade)

Today during my 7 year old’s violin lesson, I was reading a “Mr. Men” book to my 4 year old. He had selected it randomly from the teacher’s complete set of the books.

I had just finished a page which mentioned “Mrs. Twinkle”, when my daughter started playing “Twinkle Twinkle Little Star”.

It wasn’t as a direct result of my reading – she was working her way backwards through her repertoire of pieces, so that had been set in motion before I’d even started reading the book. And there was nothing to specifically draw my son to this book.

I often notice that I’m reading a word at about the same time I hear somebody say it, but that easily make sense considering the number of words I read. But my daughter only knows about 6 violin pieces by heart!

So what are the odds?


Below are the extended notes provided by mathematician Brian Pasko for use in Skepticality Episode 267.  Brian is on the faculty at a university in the southwestern United States. His interests include scientific skepticism, popular science books and improbable coincidences that makes one wonder just what the fates are up to. Take a look and leave your comments below. Also, please be sure to listen to the podcast for our own hilarious commentary.

What a lovely vision: a young mother dramatically reading to the toddler in her lap while her daughter struggles through the elementary musical cannon. I can only hope that afternoon sunlight was streaming through the large windows of the studio…

How probable was your experience? I would say rather high! Of course, what questions we ask determines our estimate of the probability. Let’s start this way: the probability that a children’s book contains the word twinkle, fairly high; the probability that at some point during the lesson your seven year old daughter plays Twinkle, Twinkle Little Star I would say pretty close to one. (Probability is measured between zero and one with zero being cannot happen and one being cannot not happen.) In this case your experience is not that hard to believe.

Or, we could wonder what the probability is of reading a book with a character Miss Spider, Mr. Farmer, Ms. Wheels or, Madame Spaghetti (atop or otherwise) and noticing that your daughter plays one of the corresponding songs. In this case, your coincidence is even expected.

We can be a bit more specific, though we have to make some assumptions. There are numerous titles in the Mr. Men series but a box set of 50 books was issued in 2010, let’s suppose your son picked the book you read from one of these. I have been unable to determine the number of these books that include the character Mrs. Twinkle but let’s just assume two. (Yes, a bit arbitrary but it seems unlikely that she appears in only one title; if she is a regular character in the series, I expect a Google search would turn up some mention of her.) Twinkle, Twinkle Little Star takes about 30 seconds to play. Reading a Mr. Men book takes about 10 minutes and if the words ‘Mrs. Twinkle’ appears on three pages there is a 2 minute window that you could read the words ‘Mrs. Twinkle’ while your daughter is playing the song.

I wrote a short computer program to model the situation assuming the lesson was thirty minutes long. It turns out that the probability that you read the words “Mrs. Twinkle” while your daughter was playing Twinkle, Twinkle Little Star is about 0.27. Since two of the 50 books your son could have chosen include this character, I estimate the probability of your experience around 1.1%.

So, I think your improbable event is not all that improbable. Imagine the alternative: that one chooses a children’s book that does not have some similarity to a common children’s song.

Littlewood’s Law

(Submitted by friend of the blog Jonathan MS Pearce)

I recently reviewed Randal Rauser and John Loftus’ debate book entitled God or Godless. I have also responded to Randal’s post on why I am an atheist as well as posting an article critiquing Randal on why he is a Christian. During my review, I noted that I was particularly frustrated at Randal’s prayer chapter.

Randal’s chapter recounted an anecdote involving prayer. In simple terms, this is it:

  1. Pastor Kent Sparks, living in North Carolina with wife, were pursuing adoption with House of Ruth
  2. No luck in year and a half, so pursued private adoption in Georgia
  3. After adoption of daughter Emily went through, Kent called House of Ruth to leave message to suspend their file
  4. Staff were in meeting to discuss with a client who had chosen Sparks for a child
  5. Meeting ended, staff called Cheryl Sparks to tell her good: news – another child for adoption
  6. Cheryl called a friend to ask her for prayer
  7. Kent returned from work and Cheryl asked him to conduct a devotion without telling him of the news
  8. Kent opened Bible and read Proverbs 3:27: “Do not withhold good from those to whom it is due,
    When it is in your power to do it.”
  9. Cheryl’s friend later phoned with a Bible verse, Proverbs 3:27
  10. They adopted their second child, Cara

And this was the main evidence to support his argument that prayer works.

Kent and Cheryl’s case evinces these same hallmarks of contingency, complexity and specification. While these events are obviously contingent, they are also complex since they involved multiple factors timed together (e.g., Kent’s call concurrent with the adoption meeting), and they included specified information (e.g., two independently confirmed referenced to Proverbs 3:27). Consequently, Kent and Cheryl (and we) are fully justified in drawing the conclusion of divine action in confirmation of the adoption. (p. 142)

When I read this account and the rationalisation of it thereof, I was staggered. Randal is an intelligent guy who claims he is conversant with cognitive biases and suchlike (I think he has written a book about them). This example is so very easy to dismiss. Let me refer you to the chapter on prayer in my book The Little Book of Unholy Questions:

The subject of prayer provides several problems for the believer, if thorough critical questioning is followed through. Part of the issue of perceived success of prayer is down to religious people interpreting coincidence as divinely purposed, and this is very common. I am aware of this, and am constantly amazed at the amount of seemingly dauntingly huge coincidences that I go through on a daily basis. Most of these are so innocuous as not to even stick in the memory. Usually, this will entail reading a book, and a certain word that you haven’t heard for ages, and then hearing it five seconds later on the television in the background. Wow! Who would have believed it? The problem is, we see things as much bigger coincidences than they really are because we are unaware of the frequency involved in calculating the probability. For example, buying a lottery ticket might mean that the probability of winning the lottery is staggeringly small, say one in fourteen million. However, if you bought fifteen million tickets, then it becomes likely. Also, if you look at the frequency of tickets bought as a whole, then someone winning is a statistical certainty. To translate this across to the word scenario, then the number of words I read or use per year, and the amount of words I hear in the background per year, means that the occurrence of these weird coincidences actually becomes a statistical certainty too. Don’t just look at the incident in isolation, but in the greater context of everything around it.

Now, as mentioned, these are innocuous cases. However, let’s look at something that happened to me the other day. I am the proud father of newly born twin boys. These two delights give us great joy, and yet they can also be a great challenge. When we introduced them to solids recently, they had a week of screaming the house down at night. This led my partner and me to have some degree of sleep deprivation, as they were waking every two hours to be breastfed. We sat down one Sunday afternoon and discussed this for about four hours. We had all the books out, and were scouring the internet for different routines, opinions and helpful tips. We were fairly stressed, and this was really important for us, especially as the boys were pretty stressed too. After all the talk and worry, we simply couldn’t conclude what to do – there were so many options. It was at this point that, had we been praying people, we would almost certainly have joined hands and prayed for strength and insight; for an answer.

Giving up, I walked myself down to the local shop for some milk, as we had some surprise guests over for a cup of tea. Just walking out of my local shop as I got there, on a random Sunday afternoon, was a woman we knew from Twins Club. I had never seen her on this road before, or even outside of Twins Club. And there she was. I stood and talked to her for half an hour. She had had exactly the same problem with her twins, gave us a routine and some ideas, and hey presto, we were sorted and so much happier. What were the chances!

Of course, had I prayed, this would have been bona fide proof that prayer works, that God listens to me, that my faith works. Imagine the joy in God’s works that I would have experienced, and imagine the evangelising I would have done at the church in telling my Christian friends of the ‘miracle’. I didn’t pray, and don’t hold that faith. What to a Christian in exactly the same sort of situation, and who has a real spiritual moment of transcendent evidence of prayer and faith, becomes just another funny coincidence to someone like me. For someone who prays frequently every day, the chances of a ‘successful prayer’ are very high.

These coincidences happen all the time. But when they happen to a religious person, they take on a whole different religious meaning derived from the religious context. Prayer works for a lot of people who follow a lot of different religions. At least most of those gods don’t exist, so something must be up. “My God and my prayers work, but yours are just coincidences,” seems like special pleading to me. The chances are, in my opinion, that most (if not all) incidences of prayer working can be put down to coincidence. We do and say an awful lot of things every day, and we wish for an awful lot of things every day. Some of them are bound to actually happen.

Besides, I’ve never seen an amputee grow back their limb after prayer. I have seen evidence of cancer naturally go into remission without prayer. Enough cancer patients get prayed for, for there to eventually be a correlation. Not, may I add, a causal relationship.

Let me now refer you to Littlewood’s Law:

Littlewood defines a miracle as an exceptional event of special significance occurring at a frequency of one in a million. He assumes that during the hours in which a human is awake and alert, a human will see or hear one “event” per second, which may be either exceptional or unexceptional. Additionally, Littlewood supposes that a human is alert for about eight hours per day.

As a result a human will in 35 days have experienced under these suppositions about one million events. Accepting this definition of a miracle, one can expect to observe one miraculous event for every 35 days’ time, on average – and therefore, according to this reasoning, seemingly miraculous events are actually commonplace.

Ever since learning about Littlewood’s Law I have been cognisant of coincidences and ‘wow’ moments and I have to admit, I have bloody loads.

Archbishop of Canterbury William Temple once observed, “When I pray, coincidences happen, when I do not pray coincidences do not happen.” Many Christians can resonate with Temple’s wry reference to God’s providence. But atheists demur, charging that such experiences only evince a selection bias that counts the hits and ignores the misses.

And I would say that Randal’s example simply does not represent a specified complexity which would prove God. Cheryl’s friend is likely to find some such relevant passage, and Kent would have such issues of the adoption at the forefront of his mind whilst choosing passages. As in my own case, things like this happen all of the time to people who don’t believe and don’t pray. They get forgotten, or not even seen as significant in any way.

Here is an excerpt which I posted on my previous blog to illustrate the point further:

I have an analogy which I hope will illustrate why at least a lot of examples of alleged successful prayer or interventions of God take place.

Transportation

Yesterday I was pumping up the tyres to my twins’ buggy. I have an old bicycle pump which I bought probably seven years ago. I bought it for £3 – peanuts. This pump has been very hard working – two bicycles and a buggy at regular intervals (the buggy particularly often needing pumping up). The pump has worked tirelessly (pun intended).

For the first time ever, whilst pumping the tyres up to the buggy in the kitchen, I wanted to talk about this pump, and laud its efficiency, reliability and value for money to my partner.

“This pump is brilliant. I’ve had it for seven years now, and it’s never let me down. I only paid three quid for it and it has been such a good bargain. Basically, it’s genius.”

And like a Greek tragedy, surprise, surprise. What amazed me was the timing. No sooner had I finished the ‘us’ of ‘genius’ than the mechanism of the pump twanged and it broke in my hands. The two of us burst out laughing at the sheer amazing coincidence of it. The first time, after very regular use for seven years, that I had ever even mentioned the pump, after singing its praises in my over-exuberant manner, it broke in my hands. Really, what were the chances!? It was like there was some supernatural force making that happen.

It was like there was some supernatural force making that happen… And that made me think.

Let me now change the analogy around – shift the paradigm. Let me now put myself in the position of being a praying Christian.

I am said Christian. I am on my way to work, and am late for an important meeting for the first time. The level crossing that I cross very often is always down. As I approach, I fear it is down. But suddenly, I see it is UP! I race through it thanking God for doing that! Woo Hoo! Now imagine, just before I approach it, I give a little prayer. When it is up, and I race though, I think to myself, “God listened! I won’t be late for that crucial meeting! Thank you God!”

Now imagine that same crossroad which is always down, is open after a little prayer with my critically ill partner on the way to the hospital. That small amount of time could be the difference between life and death. That same prayer has a massive consequence. Now God really is listening and I will remember that for the rest of my life.

But let us return to the original event. The pump breaks after an amazingly coincidental exuberant display of affection for the pump. Hey-ho, I forget about it after a week.

If I was a fervent believer, I would be praying multiple times a day, asking for things very often. The sheer volume of prayer means that many of them, by the laws of statistics, will be ‘successfully acted upon’.

The sheer volume of things we do every day, every week and every year (considering we are often doing many thing simultaneously – driving to work whilst listening to the radio and thinking of my twins) means that, statistically, HUGE coincidences will happen remarkably often. If you attach a prayer prior to that, a remarkable event will seem to happen at the will of God in answering your prayer.

And just in case you aren’t convinced, here is an example of me comparing my experience further above to my Christian friend who produced a very similar example and argument to Randal. This is an email I sent a couple of years ago using the same twins example used above:

With regards to last night’s session in the pub talking of miracles, we used a miracle claim of Colin to steer the talk. Colin has claimed a miracle of answered prayer occurred whose specified complexity points towards it being a miracle. It went something like this (apologies if I misrepresent you here, Colin):

  1. Colin had a specific problem which was affecting him badly with regards to a biblical passage.
  2. He was going away for the weekend to a Christian retreat / party
  3. He, the next day or two, had an image in his mind of a golden sword.
  4. The next day he was in a book shop and the second or third book he pulled out had an image of a golden sword on the front. He opened it to a page in the book which answered all his worries.
  5. He claims this had such a specified complexity as to be best explained by it being a second-order miracle (one that does not violate natural laws).

Andy and I both came back with some ‘incredible coincidence’ stories. Colin claimed these did not have the same level of specified complexity. I will now attempt to show you that he was wrong.

Here is the quote from my last book to explain the scenario:

[I use the quote above to give the case involving the twins.]

So what we have here is this:

  1. We had a problem that was affecting us which we sought the answer to.
  2. Some surprise guests turned up unannounced
  3. We had run out of milk and I had, at that particular moment, to go to the shop to buy some.
  4. At the shop I met a mother of twins who I have never seen before or since on my road.
  5. She gave us all the answers we needed to our massive relief as she had been through EXACTLY the same issues.

Now let’s compare these two stories for probability. At the end of the day, miracles deal in probability and specified complexity is merely a reflection of probability.

First of all, we have the problem. Colin is a Christian, there are many Christians and many have issues with passages in the bible. This problem we had involved not one, but four people, thus the probabilities that must exist to conspire to all of us being there to have that problem are higher. However, as an individual starting point, these probabilities are less relevant.

The catalyst: Colin had an image over a 2-3 day period which coincided with the cover of the book. We had a situation where we were discussing the problem at length and right afterwards some unannounced guests arrived. The actions of two other people must now be calculated such that the chances of them coming to our door, from living in London, are very low indeed. Suddenly they are there. AND THEN we had to have run out of milk in order for me to need to go to the shop. Just on the catalyst front, my story appears far more improbable, statistically.

Next, Colin is in a book shop and picks out a book which corresponds to his image. I walk to the shop and find not just anyone, but the EXACT person who had experienced THE EXACT SAME THING, there with her twins. I had and have never seen her there before. Had we prayed, she literally would have been the answer to our prayers. The probability of her being in that exact place at that exact time, of being a mother of twins with exactly the same problem (and for me to need to go to the shops at that time due to milk running out and unexpected guests) is astronomically more improbable that a book detailing information on a biblical passage being in a Christian bookshop full of other Christian books.

As I was pointing out to Colin , I don’t think there is often an understanding of the massive improbability of coincidences like mine, and there is often a desire to make the calculations for probabilities which seem to involve purpose much lower due to intuitive belief that the events are purposes. At the end of the day, If Helen and I had held hands and prayed before my friends came to the door, that chain of events would have seemed more powerful, I posit, than Colin’s miracle claim. Heck, I would have been praising the Lord!

Using Littlewood’s Law, of course, we know that highly improbable events take place with alarming regularity since the frequency of things we do and experience is phenomenal. Littlewood calculated you would experience a ‘miracle’ once a month.

Thus I hope to have shown that massive coincidences happen regularly and have just a low probability, and often lower (as in this case), than many religious miracle claims. Just because there is no perceived purpose does not mean the probability is any higher.

So, given these points, I think that Randal’s case is exceptionally weak. It certainly does not evidence God. Think of all the ways in which prayer could work which would leave one with no doubt. The complexity which Randal invokes is simply not strong enough or specified enough to do what he wants it to do. Only if you overload it with copious amounts of cognitive bias. Again, we could talk of growing back limbs and what have you. What do we have instead? Events which look extraordinarily like coincidences.


A Tippling Philosopher is a blog dedicated to the philosophy of religion, with a popular, easy to digest approach. The name comes from the casual philosophy and theology group that author and blogger Jonathan MS Pearce frequents in Hampshire, UK. This blog is an extension of that, with guest posts by other thinkers with the same questioning vein from around the world. What started with Socrates, in challenging the legitimacy of religious beliefs of his time, will hopefully be continued several thousand years later with the lively community of critical thinkers in the Skeptic Ink Network.

As an author, Pearce writes about the subjects which fascinate him hugely. His first book “Free Will?” is a work dedicated to investigating free will and determinism, presenting a wealth of evidence to support a deterministic worldview. His second book “The Little Book of Unholy Questions” is a cumulative case against the existence of God written in the form of a set of questions asked directly to God. His last book “The Nativity: A Critical Examination” is a synthesis of the work detailing the analysis of the infancy narratives in the New Testament, showing that the two Gospel accounts are clearly a-historical.

Visit JP’s blog here.

(Submitted by Skepticality listener Michael O’Dea

Hi there,

I enjoy the show and want to tell you my against-the-odds-story.

I am from Dublin, Ireland and I was on vacation in Boston, visiting my cousin about 20 years ago.

There was a free public concert in the Boston Common park. (It was Kid Creole and the Coconuts, not that that is relevant!)

I was with an American friend who was a server in a Boston restaurant (Legal’s) at the time. As we enjoyed the music he met a colleague from the restaurant who was with a companion and they chatted for a few minutes as we watched the gig. My friend then went to introduce me, when the companion turned around it was my next-door-neighbour from Dublin!

We had not seen each other for years and had no other connection of any kind other than growing up in adjacent houses.

What do you think?


Below are the extended notes provided by cognitive psychologist and statistician Barbara Drescher for use in Skepticality Episode 239.  Take a look and leave your comments below. Also, please be sure to listen to the podcast for our own hilarious commentary. Also, visit Barbara’s blog.

I think this is an interesting coincidence! Normally, I would talk about the factors that would increase the probability of this happening, so I will, but there are really very few. People living next door to one another are much, much more alike than two people chosen at random from the global population. They are more likely to be close together in S.E.S. (socioeconomic status), for example. They are more likely to be exposed to similar cultural icons (such as music genres). Factors such as these may exponentially increase the probability of running into each other at just such an event.

However, given the astronomically small base probability (e.g., given all of the people in the world, the probability of any two people, chosen at random, would meet), this is still a story with crazy odds.

Consider the factors that don’t really come into play here, but have in similar stories we have encountered. For example it is unlikely both been inspired to visit by the same event (e.g., hearing a mutual friend talk about visiting Boston). They may have been inspired to visit (assuming the companion was also visiting and not living there) by cheap airfare to the U.S., but then why choose Boston? The probability that they all met each other through mutual friends is greatly reduced by the fact that the Americans know each other because they work together (unless, of course, they knew each other before working there).

So we must rely on the mathematical rule that we should expect at least some low- and even astronomical-probability events to occur in our lives, given the large number of events that occur.

Monty Hall

Monty Hall

I have always been amused and intrigued by responses to “The Monty Hall Problem”, especially when I talk about it to audiences with a high concentration of engineers and mathematicians. If you are familiar with it, but you’ve always struggled with an unsettled feeling of “this can’t be right”, read further and let me know if my explanation of the solution helps to alleviate the discomfort. If you are not familiar, I guarantee you will give your brain a workout by reading on.

First posed to statisticians in 1975, “The Monty Hall Problem” is well-known among academics because it still sparks debate. Many seem to think that disagreements about its solution stem from issues in the clarity of the problem, but I contend that it really stems from human flaws in the way that we process information.

I often discuss this problem in statistics and cognitive psychology courses for several reasons. It is a great exercise in probability calculation and it can be used to teach basic mathematical modeling (and its purpose). An added benefit, since almost all of my students were psychology majors, is that it also illustrates a flaw in human cognition as well as a pattern of problem solving.  Even a knowledgeable statistician feels the need to run simulations to see the solution in action. Even then, fully grasping the mechanisms behind the answer often requires brute force cognition.

In general, human beings have a very difficult time wrapping their brains around concepts of probability. It is much like a visual illusion; we know that the lines are parallel/the circles are the same size/there is no motion, but we can’t make our brains process it in a way that represents that reality. It’s just not how our visual system works. I hypothesize that one of the reasons that probability is such a difficult field for most people is that it involves theory and models, which are distinct from observations and we must represent them differently in our minds to properly deal with them. Applications of probability often involve switching gears from the realm of models to data or vice versa and this is where I think most mathematicians get side-swiped in The Monty Hall Problem.

The Poser

In essence, here’s the problem:

View full article »

Kindergarten Koincidence

(Submitted by reader Daniel L)

When I was in kindergarten some 20 years ago, I became close to one of the ladies working there and she was my favorite teacher of them all.

When I was in the 8th grade (about 9 years later) my favorite teacher showed up in my dream for the first and only time since I went to kindergarten. Later that very same day when I was riding my bike home from school, who do you think I run into on the sidewalk, if not, my old kindergarten teacher.

To this day, I’ve carried this story with me as a perfect example of a “one in a million event”. Although I am a statistician by education, and can point to several high-probability factors in this story to increase the odds a bit, I still think it’s a really cool story.

[EDITOR: I’ll refrain from speculation as to what happened in the dream.]