Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Coincidentally or is it?

  1. May 16, 2003 #1

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    How about some stories, ideas, or critique [please be nice I am trying to be objective here]? One question that I like to consider is: Are coincidences always coincidences? Now I know all kind of ideas exist about synchronicity and what not, but these are unproven philosophies or theories, I am interested in events and odds.

    Sometimes it seems that the odds against some coincidences are extremely high. So, at times I have tried to consider the odds of such things. For example, and this is just as an example, I don't remember the actual word now, but a rather rare word that I had never heard came up in a conversation one day. That night I heard the same word used on a TV program. The two events seemed completely unrelated by subject or context. So, what are the chances of such an event? Then, how often should we expect such events? Finally, do these events happen more frequently than we might expect?

    After a little consideration, it seems that some coincidences happen much more often than is statistically predicted. Now I know this is really soft, hence the PS post rather than the physics or math forums, but understand also that I have thought about this for many years, and a pattern does appear to be repeating. Coincidences seem to happen too often. For this reason I am asking for your thoughts on this approach. I make no claims of proof.

    Now someone can probably crucify my on this point, but one approach considered for the example above is that x number of words exist in the English language. Some words are more likely than others, but to a zeroth order approximation, there is a chance of 1:x that the next word that I hear will be "the". We can quickly consider nouns, adverbs, pronouns etc etc the relative proportion of each. Using this reasoning we can consider the average number of words heard each day, then the chance for any one particular type of word, and finally an estimate of the chances for a class of words such as "rare animals", or "common items found in common places","pronouns", "common verbs" or "figures from history", "common references such as the bible" etc etc. The point is not to designate all orders of words found, rather it is to develop a framework for judging any particular word, phrase, or event. I am using word thing as the example, but typically I really mean events.

    I will stop here for whatever comments you may have. Also, I am not selling a theory. These are just honest questions and thoughts about this approach – in very broad terms - for your most valued consideration.
  2. jcsd
  3. May 16, 2003 #2
    a thought

    It seems to me that trying to calculate the coincidences of many things is just far to complex and there is to many indeterminate variables. But im not very good at math strange seeming im a computer programmer.
    It's not a bug it's a feature :)
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?