Discussion»Statements»Rosie's Corner» I always wonder what the odds are of related things occurring simultaneously. What makes some odds easier to calculate while others aren't?
The closer the relationship and the more common or expected the events, and the wider the spread over which the events occur, the easier it is to assess such odds; but it still depends heavily on such events being amenable to objective statistical analysis.
Also, you have to beware the 3-C trap: Co-incidence, Correlation, Cause.
Co-incidence at its simplest is a matter of chronology.
"It was pouring with rain in London and New York today", is a co-incidence of common events. The two cities often have rain, but are too far apart and in very different physical-geographical settings with an ocean in between, to be affected in the same way at the same time by the same weather system. In fact I don't think it would even be the same system.
Correlation is a match of similar things or events.
"It was pouring with rain all day in London, Southampton and Milton Keynes". Yes - because a depression is many hundreds of miles in diameter, so the two other towns, less than 100 miles from London, are well within the same depression; and by correlating the local weather effects in them over a few days you can map that weather system's behaviour.
Geologists use correlation to help build a picture of past geographical and climate conditions, by the evidence given by similar rock types and fossils.
Correlation is used in some electronic measuring systems to pinpoint a particular event in space or time. Water-supply companies use the principle to find a hidden leak in a mile or so of buried pipe, by an instrument that calculates it from the time taken for the leak's noise in the water at any instant to reach sensors on or in the pipe each side of the suspected area.
Cause is what is says: the event in question is due to the observed, correlated influences or events; and muddling correlation or mere co-incidence with cause can be a very dangerous thing to do.
The link between smoking and lung-cancer is a classic example of how it should work. Doctors began to notice many sufferers were smokers, so started to examine the correlation. Actual cause came much later, when sufficient cases had been examined to sift alternative causes such as industrial lung-diseases or pure chance from smoking; and very importantly, to calculate the risk of developing cancer from the identified hazard, the tobacco tar.
I have personal examples that illustrate this. My Dad smoked for many years, but it was bone, not lung, cancer that killed him (the smoking can't have helped him though, might have been a factor but might just be co-incidence). "Everyone" knows cirrhosis of the liver is "caused" by alcohol - yes, long-term excessive drinking certainly can cause it, but not in my mother's case as she did not drink alcohol.
Unfortunately, we humans have an ingrained wish to see patterns even where none exist: ask any astrologer. This does not always help us sift the Thee Cs from each other; to see risk from hazard, or to use statistics properly; and so understand which patterns are real, and whether apparent relationships are genuine and or merely co-incidental.
Sometimes I wonder if the Old Testament prophets actually did understand it, empirically at least, far better than many of today's soothsayers. After all it is easy to prophesy to a largely-uneducated population, an Act of God natural disaster of a type fairly rare but still known in your region, if you are careful not to suggest when it might happen!
Once again thank you for your thoughtful, informative and very thorough reply Durdle. I am one who is ALWAYS looking for connections/correlations/meaning. Please bear with me and I shall give you two examples. A long-time very good internet friend used the exact same word to describe a situation that I intended to use but decided against as possibly being "a bridge too far". The exact word. It is a little-used word. EUNUCH. She lives in New York. I live in California. What are the ODDS of that? Are they calculable? It made my day frankly.
A couple of days ago we watched a Food Network episode of a guy who travels around visiting different restaurants in the US. It always includes 3-4 locations per show. The show we watched featured a restaurant in VIRGINIA BEACH right after the Virginia Beach Massacre occurred. The show was taped in 2011, 8 years before. I KNOW there cannot be any connection whatsoever. But there are so many hundreds of thousands and thousands and thousands of restaurants and so many hundreds of thousands and thousands and thousands and thousands of cities I couldn't help but wonder WHAT ARE THE ODDS of watching a show 8 years old part of which took place in a city where a massacre just occurred in the NOW? Can it be calculated or would that be impossible to do? My brain is always seeking connections between disparate things. You know about quantum entanglement I'm sure. Odd strange weird inexplicable but there it is! Thank you for your reply Durdle! :)
Eunuch? A castrated human male? Unusual topic of conversation, so yes, quite some co-incidence! The distance has no real bearing on it, but as long-time friends it could well be that one of you picked the word up from the other.
I don't think you could calculate the odds on that, as it's too random.
; Nor could you calculate the odds of the co-incidence of a crime being carried out in a restaurant that happened to one of many reviewed in a perfectly straightforward way some years before.
However, I wonder though if the repeat was pure co-incidence or to give some sense of normality to the scene. The show's run of repeats would have been planned some time ago, and either the producers had not spotted that particular episode was about to be aired when it was, or they knew it was but chose to let it go ahead rather then re-schedule it.
The BBC has sometimes re-schedule a programme when a particular atrocity or disaster would made the intended one, untimely.
What does seem to happen is that we spot co-incidences in things that are not actually related, because they make an impression on us. The more uncanny the co-incidence the more we latch onto it, and remember it.
I found this in the aftermath of the attack on the World Trade Center, I happened to look at the time on a few mornings right on 11:11. It was of course pure, incalculable chance; but the attack itself such a major event that glancing at my watch at that particular time, it clicked in my mind. Any of the other many times I looked at my watch, would gone by un-noticed except at the instant for that instant's own reasons.
Durdle 