Jump to content
Join the Unexplained Mysteries community today! It's free and setting up an account only takes a moment.
- Sign In or Create Account -

Strange Question Concerning Probability


Simatong

Recommended Posts

So, seeing as how I have nothing to do (My clients have been rather quiet as of late, for the most part.), I've been thinking about probability when it comes to certain types of psi phenomena, specifically about the likelihood of someone accurately predicting the certain semi-specific things in terms of natural phenomena. Since I have no understanding of mathematical calculations pertaining to probabilities, I was wondering if someone might be able to just provide some sort of calculation. Again, I have little to do right now, and the question has been intriguing me anyway.

So here are a few scenarios.

What is the likelihood of someone correctly predicting:

*the date that a large volcanic eruption commences in a North American location after years of dormancy (assuming that the location was merely stated to be North America at least a month before it happens? (This excludes any and all locations in Hawaii, as there are persistent eruption there.)

*the exact date and duration of a coronal mass ejection during solar minimum? (I am using solar minimum in the scenario, as I have read they occur less frequently during this period and are thus even harder to predict.)

*the day and intensity of a thunderstorm at least four days in advance when a meteorologist might have originally concluded that said day would be without storms?

*a semi-specific intensity prediction (i.e. "significantly weaker than what had been predicted") for a hurricane season that might be predicted to be extremely active, as well as accurately predicting some of the phenomena influencing the weakness of a hurricane season (i.e. El Nino-like energy offsetting storm-causing conditions, the mentioning of dust, the occurrence of which actually weakens the storm-inducing effects)?

*accurate wind speed predictions for days for which no wind speeds had been predicted?

* and finally, the probability of providing all this information and being accurate with all the aforementioned predictions.

Thanks in advance and sorry if this seems like an eyebrow-raiser lol.

Link to comment
Share on other sites

 

Zero likelihood.

Link to comment
Share on other sites

Zero likelihood.

Wow! That was fast! lol

To be honest, I had thought that it might be slightly likely to occur by way of chance (like 1 in 500,000 or 1 in 1,000,000), but, at least from your perspective, it isn't. Thanks for the input, and sorry if the question seems to come out of left field. :)

Link to comment
Share on other sites

What is the likelihood of someone correctly predicting:

*the date that a large volcanic eruption commences in a North American location after years of dormancy (assuming that the location was merely stated to be North America at least a month before it happens? (This excludes any and all locations in Hawaii, as there are persistent eruption there.)

Location dependent. There are some active volcanoes in North America. Most of these have not erupted for the past 5 million years: http://en.wikipedia.org/wiki/List_of_volcanoes_in_the_United_States

There's about 40 in Alaska that have erupted in the past 200 years (so one eruption every 5 years or thereabouts... the chance of getting the date right is 1 in 1826. Not impossible but unlikely. Now, if you're looking at "dormant" volcanoes, then you have to define "dormant." But the chances of predicting an eruption of the Yellowstone caldera by psychic powers are nearly zero. When the ground becomes active, you will see all sorts of predictions from psychics -- but you wouldn't see this prediction during a time when the Yellowstone Caldera is relatively quiet.

*the exact date and duration of a coronal mass ejection during solar minimum? (I am using solar minimum in the scenario, as I have read they occur less frequently during this period and are thus even harder to predict.)

90% -- with a disclaimer that you could also "psychically predict" them by using a random number table or any method you care to generate a date and a time -- like a coin flip. Focus on the correct ones and your coin will look like The Amazing Kreskin. The reason for this is that they're so darn frequent. http://nopr.niscair.res.in/bitstream/123456789/11192/1/IJRSP%2040%281%29%207-10.pdf

So even if you knew nothing about them, you could predict them if you allowed a window of a few hours or a few days (which most do for psychic predictions.)

*the day and intensity of a thunderstorm at least four days in advance when a meteorologist might have originally concluded that said day would be without storms?

25% -- 50% that a thunderstorm is predicted and 50% that you get the strength approximately right (about the same you'd get from flipping a coin twice.) Weather is influenced by many things-- it's a chaotic system and there are too many inputs to calculate. So it's a "flip of the coin" One guess wouldn't be unusual ... but a sequence of 15 or more with no misses would be significant.

*a semi-specific intensity prediction (i.e. "significantly weaker than what had been predicted") for a hurricane season that might be predicted to be extremely active, as well as accurately predicting some of the phenomena influencing the weakness of a hurricane season (i.e. El Nino-like energy offsetting storm-causing conditions, the mentioning of dust, the occurrence of which actually weakens the storm-inducing effects)?

Depends on how you define it. "extremely active" is one of those words that could be anything from 10 hurricanes to 40 (and depends if you count tropical storms, hurricanes in the Pacific and elsewhere, etc. Now, getting the EXACT number of hurricanes -- that's a pretty high chance even if you use a coin toss to calculate the number (use binary, where heads = 0 and tails = 1) because you're not going to have an Uncountable Number of Hurricanes during a season. "Hurricane Season" is June 1 to 30 November -- 167 days. The chance of one or more hurricanes on any particular day is 1 in 167. You also have to ask "When is it declared a hurricane"?

*accurate wind speed predictions for days for which no wind speeds had been predicted?

You'd have to ask "when and where and who's measuring?" If the weather service predicts no winds, there may be winds at my house some 5 miles away that I can actively measure (think of the atmosphere as being formed of a lot of cells that are 1/2 mile in diameter. This is a gross oversimplification but gives you an idea of why you can have rain down the block while there's no rain at your house.)

One correct guess would not be astonishing. A sequence of 25 correct with no misses WOULD Be unusual.

* and finally, the probability of providing all this information and being accurate with all the aforementioned predictions.

Again, you could get ANYTHING right just once by sheer chance. Getting all of that exactly right 25 times in a row would indicate something very unusual.

  • Like 2
Link to comment
Share on other sites

Location dependent. There are some active volcanoes in North America. Most of these have not erupted for the past 5 million years: http://en.wikipedia....e_United_States

There's about 40 in Alaska that have erupted in the past 200 years (so one eruption every 5 years or thereabouts... the chance of getting the date right is 1 in 1826. Not impossible but unlikely. Now, if you're looking at "dormant" volcanoes, then you have to define "dormant." But the chances of predicting an eruption of the Yellowstone caldera by psychic powers are nearly zero. When the ground becomes active, you will see all sorts of predictions from psychics -- but you wouldn't see this prediction during a time when the Yellowstone Caldera is relatively quiet.

One should also take into account such things as the VEI (Volcanic Explosivity Index). While a level 2 VEI occurs weekly (somewhere on the globe), combining that with the continental location and date of eruptive commencement, I wonder if the probabilities would be affected much. And by "dormant", I mean active volcanoes that are not currently erupting, but are expected to sometime in the future.

90% -- with a disclaimer that you could also "psychically predict" them by using a random number table or any method you care to generate a date and a time -- like a coin flip. Focus on the correct ones and your coin will look like The Amazing Kreskin. The reason for this is that they're so darn frequent. http://nopr.niscair.... 40(1) 7-10.pdf

Actually, during solar minimum, you get about 1 CME every 5 days or so, meaning that, while that is still not the rarest of occurrences, they still happen infrequently enough during solar minimum to to make predicting when one will occur rather challenging, even without psychic intervention. Now, if we are talking solar maximum, then yes, I agree with you on the frequency of such occurrences; one could argue that accurately predicting a CME in October of '07 would not be as easy as it would in say maybe July of '14.

So even if you knew nothing about them, you could predict them if you allowed a window of a few hours or a few days (which most do for psychic predictions.)

By the date, I mean to say if a psychic were to predict that a CME would occur between January 11th and January 14th and a CME occurs during those exact dates.

25% -- 50% that a thunderstorm is predicted and 50% that you get the strength approximately right (about the same you'd get from flipping a coin twice.) Weather is influenced by many things-- it's a chaotic system and there are too many inputs to calculate. So it's a "flip of the coin" One guess wouldn't be unusual ... but a sequence of 15 or more with no misses would be significant.

Then let me provide some more specifics: predicting not only a thunderstorm for that day (or days), but that there would be severe storm warnings/watches as well, and just severe weather overall, with two to three storms and/or potential storms being predicted accurately (or at least the eventual and official declaration of the watches and warning)

Depends on how you define it. "extremely active" is one of those words that could be anything from 10 hurricanes to 40 (and depends if you count tropical storms, hurricanes in the Pacific and elsewhere, etc. Now, getting the EXACT number of hurricanes -- that's a pretty high chance even if you use a coin toss to calculate the number (use binary, where heads = 0 and tails = 1) because you're not going to have an Uncountable Number of Hurricanes during a season. "Hurricane Season" is June 1 to 30 November -- 167 days. The chance of one or more hurricanes on any particular day is 1 in 167. You also have to ask "When is it declared a hurricane"?

What is declared as a hurricane is pretty simple to define, fortunately, with normal parameters calling "hurricane" once the storm clocks in at at least 75 mph. By extremely active, I do not mean the number of storms, but rather their ACE (Accumulated Cyclone Energy); you could have a season with 17 storms and, in terms of energy output. it could be the same as, or lower than, that of an average hurricane season. An extremely active hurricane season based on the aforementioned parameters would therefore need to have an ACE of at least 111 (120% of the 1981–2010 median). Therefore, if The likely range of the ACE index is, say 120% to 200% of the median (based on non-psychic predictions), this would be the definition of "extremely active".

You'd have to ask "when and where and who's measuring?" If the weather service predicts no winds, there may be winds at my house some 5 miles away that I can actively measure (think of the atmosphere as being formed of a lot of cells that are 1/2 mile in diameter. This is a gross oversimplification but gives you an idea of why you can have rain down the block while there's no rain at your house.)

In hindsight, I should have been more specific with my scenarios than I actually was, and for that I apologize. By unpredicted wind speeds, I meant that (at least in the United States) most forecasters do not give wind speeds more than two to three days ahead it seems. If a psychic were to make predictions about wind speeds on those days WITHOUT wind speed predictions (mundanely and meteorologically gained predictions, that is), and if they were to sense winds with speeds of say 20-40 mph multiple times during a one-week period (that week starting on the first day for which the meteorologist did not predict wind speeds--with such high wind speeds being rather rare inland), what would the odds of them being accurate, say, 5 out of 7 times be?

Again, you could get ANYTHING right just once by sheer chance. Getting all of that exactly right 25 times in a row would indicate something very unusual.

I asked about only one time guesses simply because I wanted to know the likelihood of a correct guess happening that one time; if 25 guesses were made and all (or nearly all) were correct, than calculating probability would almost be unnecessary, as the likelihood of that happening would approach 0 anyway.

Thanks for the detailed response! Am eager to hear more about your insights. :)

Edited by Simatong
  • Like 1
Link to comment
Share on other sites

One should also take into account such things as the VEI (Volcanic Explosivity Index). While a level 2 VEI occurs weekly (somewhere on the globe), combining that with the continental location and date of eruptive commencement, I wonder if the probabilities would be affected much. And by "dormant", I mean active volcanoes that are not currently erupting, but are expected to sometime in the future.

Well, vulcanologists seem to think that "dormant" could be up to 30,000 years or longer -- Yellowstone's recharge period seems to be 700,000 years (wikipedia http://en.wikipedia.org/wiki/Volcano#Dormant.) -- I just picked 200 years to make the data easier to count.

Actually, during solar minimum, you get about 1 CME every 5 days or so, meaning that, while that is still not the rarest of occurrences, they still happen infrequently enough during solar minimum to to make predicting when one will occur rather challenging, even without psychic intervention. Now, if we are talking solar maximum, then yes, I agree with you on the frequency of such occurrences; one could argue that accurately predicting a CME in October of '07 would not be as easy as it would in say maybe July of '14.

Quite true. I did, however, find a paper showing a fairly high level of activity one month during a solar minimum, although I did note the 1 every 5 days figure. However, here you have to define what a "hit" would be -- simply stating the date? If so, the GMT date or local date? Predict to within 24 hours of the ejection (essentially a 3 day window)?

By the date, I mean to say if a psychic were to predict that a CME would occur between January 11th and January 14th and a CME occurs during those exact dates.

That 4 day window (say, start of January 11 GMT to end of January 14) has an almost 80% chance of getting a hit on something that occurs an average of every 5 days.

Then let me provide some more specifics: predicting not only a thunderstorm for that day (or days), but that there would be severe storm warnings/watches as well, and just severe weather overall, with two to three storms and/or potential storms being predicted accurately (or at least the eventual and official declaration of the watches and warning)

Freakonomics has some interesting statistics on the reliability of weather forecasters. So... it seems about 50%-60% for a 3-4 day window. (8 years ago.) I'd say that the chances of out-predicting the weatherman was still one in two. One hit wouldn't astonish me. Two years of 90% accuracy would.

What is declared as a hurricane is pretty simple to define, fortunately, with normal parameters calling "hurricane" once the storm clocks in at at least 75 mph. By extremely active, I do not mean the number of storms, but rather their ACE (Accumulated Cyclone Energy); you could have a season with 17 storms and, in terms of energy output. it could be the same as, or lower than, that of an average hurricane season. An extremely active hurricane season based on the aforementioned parameters would therefore need to have an ACE of at least 111 (120% of the 1981–2010 median). Therefore, if The likely range of the ACE index is, say 120% to 200% of the median (based on non-psychic predictions), this would be the definition of "extremely active".

There's a huge amount of information on el Nino and la Nina on the internet, so a prediction done of the upcoming hurricane season -- 50-50 chance. Predicting a fairly specific range 7-9 years out and getting (say) 80% accuracy for a decade would be interesting. To check it, we'd have to do a blind test on past data to see how well chance can match with the results.

In hindsight, I should have been more specific with my scenarios than I actually was, and for that I apologize. By unpredicted wind speeds, I meant that (at least in the United States) most forecasters do not give wind speeds more than two to three days ahead it seems. If a psychic were to make predictions about wind speeds on those days WITHOUT wind speed predictions (mundanely and meteorologically gained predictions, that is), and if they were to sense winds with speeds of say 20-40 mph multiple times during a one-week period (that week starting on the first day for which the meteorologist did not predict wind speeds--with such high wind speeds being rather rare inland), what would the odds of them being accurate, say, 5 out of 7 times be?

Measured at the meterological station? Or elsewhere? Wind speeds vary at different places even in the same city.

Link to comment
Share on other sites

Well, vulcanologists seem to think that "dormant" could be up to 30,000 years or longer -- Yellowstone's recharge period seems to be 700,000 years (wikipedia http://en.wikipedia....olcano#Dormant.) -- I just picked 200 years to make the data easier to count.

Quite true. I did, however, find a paper showing a fairly high level of activity one month during a solar minimum, although I did note the 1 every 5 days figure. However, here you have to define what a "hit" would be -- simply stating the date? If so, the GMT date or local date? Predict to within 24 hours of the ejection (essentially a 3 day window)?

That 4 day window (say, start of January 11 GMT to end of January 14) has an almost 80% chance of getting a hit on something that occurs an average of every 5 days.

Freakonomics has some interesting statistics on the reliability of weather forecasters. So... it seems about 50%-60% for a 3-4 day window. (8 years ago.) I'd say that the chances of out-predicting the weatherman was still one in two. One hit wouldn't astonish me. Two years of 90% accuracy would.

There's a huge amount of information on el Nino and la Nina on the internet, so a prediction done of the upcoming hurricane season -- 50-50 chance. Predicting a fairly specific range 7-9 years out and getting (say) 80% accuracy for a decade would be interesting. To check it, we'd have to do a blind test on past data to see how well chance can match with the results.

Measured at the meterological station? Or elsewhere? Wind speeds vary at different places even in the same city.

1, By "those exact dates", I meant that the psychic would have claimed that a CME would begin on the 11th and end on the 14th, not that it would merely happen somewhere in between those time frames. It is one thing to just say that a CME would be occurring on a specific date; it is another to say exactly when one begins, and even more so to say when one would end.

2. Good point about the weather. However, meteorologists rely on computer models and the like to figure out what is going to happen with the weather; I am talking about someone, with no understanding of meteorological modelling whatsoever, stating that rain and severe weather would happen on a date that was forecast to be quiet and without rain.

3. The occurrence of El Ninos and La Ninas are themselves rather tricky to forecast. For instance, it was projected that La Nina would help boost storm activity in the Atlantic during the '07 season, but it was unexpectedly weaker than what had been projected. Also, I stated it would be an El- Nino-like phenomenon for this scenario, not an actual El Nino, with the former sometimes having as much of an impact as the latter. Additionally, one might think that a purported psychic's prediction of hurricane activity being significantly weaker than what was expected, with storm-causing factors being offset by interfering factors, would affect the probability outcome in some way

4. Wind speeds do vary from place to place, but again, winds as high as 20-25 miles per hour are rare in the United States. However, even when taking the difference in wind speeds into consideration, the aforementioned statement still stands and is augmented by another truth: It does not matter where in the city the wind speed occurs when said speeds are has high as 20-25 mph given their rarity, only that they occurred at all, especially if it were to be in a city that does not have a common occurrence of such high wind speeds. But even if we were to take location into consideration, then all that would need to be said was that the meteorologists' gauge or what-have-you would say that he speeds were such and such.

5. Again, a single occurrence may not mean much in the grand scheme of things, but I am asking about the possibility of one occurrence of each of these correct predictions just for interest's sake, as after multiple hits, the probability reaches zero, and thus probability becomes meaningless at that point.

Thanks again for taking the time to answer.

Link to comment
Share on other sites

Actually, there's an excellent way to check the predictions -- something called 'catastrophe bonds' ('cat bonds'.)

They're bonds that insurance companies use against disasters. I haven't looked into the specifics, but they're bought and sold like any commodity and they fluctuate depending on whatever disaster hits. So if someone thinks they have a good psychic prediction ability for this, they should open an account on finance.yahoo.com (or any board where you can pretend to buy and sell stock market commodities) and prove their excellence with the catastrophe bonds (buy and sell based on their psychic predictions about catastrophes.)

I'm thinking that on the average (if they followed, for instance, some 10-15 bonds) they won't do much better than the finance gurus, and might actually do worse than most financial gurus who have been in this for awhile.

The advantage of "prove it by bond market trades" (on Yahoo, when I was doing it, you didn't actually have to OWN the stock. You just pretended to buy shares and sell shares.) is that the person who's making the predictions can't fudge time windows. It has to be on the mark or your investment fails big time. They can't come back and "correct" a prediction, either.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.