Jump to content
  • entries
    260
  • comments
    800
  • views
    178,015

Statistics are weird


~Shockwave~

739 views

So for some reason my mind started wandering (again) and this simple sounding topic came into my brain. Statistics. Percentages. It's really odd how we use them. we use them in weather, we say there's a 39.54% chance of rain. but in reality, it's either going to rain. Or it isn't. in science, if you do an experiment, than do it again, the exact same way it was done before, it should have the same results. of course you would have to assume nothing changed. which is probably never safe to assume. but still, when a rain cloud comes in it doesn't just say "Eh, not today." (though it feels like that where I live sometimes.) there has to be a reason that it didn't rain. But I can only assume that we use percentages because we don't have the means to gather all the data we need to make an accurate prediction. But that doesn't make a percentage a good way of representing what's going to happen. Percentages work fine for information that's already been gathered, like test scores or votes or whatever. but when used to predict future events, it just doesn't work.

 

Let's conduct a little experiment. I like Transformers, let's use that. Ok, so lets say I have a box that contains a single transformer figure. I have long since forgotten which one or what form it's in. I might proceed to guess what form it's in, therefore giving each one a value as to how likely it is that it's in. I tend to keep them in robot mode, but if it's in storage It's liable to be in either mode. To make things even more confusing, I'm not sure where my Astrotrain figure is. (He's a triple changer for the unknowing.) So right now it could be one of at least 2 figures, one of which is in 1 of 2 modes, the other 1 of 3. But since I'm not sure I can remember every figure in my collection, he could be any number of different figures, in a variety of modes. So I can make a prediction who, and in what mode, is in that box. Now it isn't a complete shot in the dark but it's mostly a random guess. Now who is in the box doesn't matter. what matters is that my prediction doesn't change what's in the box. so why do I bother predicting anyway? Or am I just ranting and this makes no sense and is a bunch of jibberish... Whatever... I don't know. I feel like I have something here...

 

 

7 Comments


Recommended Comments

Statistics are fun. :D

If you haven't before, look up the Monty Hall Problem.

That's some weird percentage stuff.

Link to comment

There's a 39% probability of it raining. That doesn't mean you're only going get 39% of the rain, it means that there is a 39 out of a 100 chance that you're going get some amount of rain. =P

 

Meteorologists have these percentages not because rain clouds can walk up and decide "eh not today", but because we literally cannot predict their movements fast enough. The weather system is an extremely complex system that, while we can simulate, we can't simulate fast enough. So we have to take out some variables to speed it up and make guesses. If we see a giant blob on the radar we can say "huh, yup going get some rain." If we see sporadic clouds we say "eeeh there's so and so chance of it raining here."

 

It's going rain or it isn't going rain, but weathermen are not all knowing, and thus rely on probabilities for their job. =P Saying that you can't use probabilities in this sense would also completely invalidate the field of quantum mechanics, and we all know how well that'd go over.

Link to comment

"Percentages work fine for information that's already been gathered, like test scores or votes or whatever. but when used to predict future events, it just doesn't work."

 

This would imply that the meteorologist is in the wrong, seeing as you say it doesn't work, not "I don't get it." The meteorologist is very much correct. =P

Link to comment

Well... Those in AZ would beg to differ... Unless the ones here just don't have it figured out. Maybe I should have used a different example. like the airplane vs car crash ratio or something. But I blame freakonomics for that one.

Link to comment

Well, obviously, our understanding of the weather still remains imperfect. My point being, the terminology is a-ok, the methodology, less so.

Link to comment
Guest
Add a comment...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...