How smart is Sup Forums?

How smart is Sup Forums?

Other urls found in this thread:

youtube.com/watch?v=f00L6OBYJP4
en.wikipedia.org/wiki/Conditional_probability
twitter.com/SFWRedditGifs

Thread theme:

youtube.com/watch?v=f00L6OBYJP4

this is the thread beam:

50%

So one of them is garunteed to land heads? In that case it's 50%

Anyone who says 50%, congratulations, you're fucking retarded.

Nevermind 33%!

Two possible outcomes:
1. Head + Tail
2. Head + Head

Probability is 1/2

33%

Possible options: both heads, one heads one tails, one heads one tails.

1/3.

Wrong. Try again, guys.

First, ask yourself what all the permutations are.

Then ask yourself how many of those permutations contain at least 1 heads.

Then ask yourself how many of those contain both heads.

Protip: Conditional probability, Bayes' theorem, answer is NOT 50%, even though most people will say it is.

You're wrong
He's right

what about Tail + Head?

1/2 * 1/2 = 0.25 = 25%

>1. Head + Tail
>2. Head + Head
>3. Tail + Head

1/3

You're retarded too.

Answer = 1/3

There are 4 possible outcomes total. the answer is .5 times .5

0%? If they both landed heads how the fuck could one land on tails?

The hard part of this question is that it says "at least one of them will be heads", that doesn't imply it will always be one of the coins. It only means that one of the coins will be heads and it could be either.

This means that there are two possible outcomes in which it is heads and tails, as well as another one in which it is heads and heads. Giving three outcomes in total making it 33%

That would be true if order mattered, which it doesn't. Tail + Head is the same as Head + Tail

In order to answer any question about probability, we must first consider our sample space. In the case of a coin flip, it is as follows: (H-heads T-tails)

HH
HT
TH
TT

Now that we have our full sample space for flipping two coins, we can narrow the scope to fit the question: At least one landed heads.

Here is the new sample space:

HH
HT
TH

Now we simply compute the probability. 1/3 of the options are "both landed heads", therefore the probability is 33.3%

>re-reads problem
>both lands heads if one lands heads
>im retarded

it's not overall probability; it's probably given one landed on heads.

Are you mentally retarded?

Already got at least 1 hands. Means you subtract the TT possibility, so instead of 25% (1/4) the probability becomes 33% (1/3)

But you see my friend, the order does matter when you are computing probability. The end result is that at least one coin shows heads. There are three possible outcomes where this is true, but only one where both are heads.

Precisely.

conditional probability:

1. Probability that one lands head AND both land heads = 1/2
2. Probability of at least one head = 3/4
3. (1/2) / (3/4) = 2/3 = 66.666666...%

>Tail + Head is the same as Head + Tail

Dubs are wasted on you.

they are NOT the same.

Think of it like this:

Take two different coins. Let's say, a penny, and a quarter. Each coin has a 50% chance of landing either heads or tails. You flip both coins. What are the possible outcomes? Well, let's see:

penny = heads & quarter = heads
penny = heads & quarter = tails
penny = tails & quarter = heads
penny = tails & quarter = tails

4 possible outcomes. Each of them equally likely to occur (25% or 1/4)

Now, surely we can all see how the results,
penny = heads & quarter = tails
and
penny = tails & quarter = heads
are different, right? Surely you can see how these are two distinct, separate and equally probable outcomes, yes?

Answer = 1/3

I love these threads because they remind me that I'm intelligent. It's 1/3

en.wikipedia.org/wiki/Conditional_probability

>1. Probability that one lands head AND both land heads = 1/2

that's 1/4, fam.

>2. Probability of at least one head = 3/4
>3. (1/4) / (3/4) = 1/3 = 33.3333333...%

FTFY

correct.

Er, yes it does. Because there are two possible ways to get a 1 head, 1 tail outcome. Ignoring the "one already landed heads" bit, if you flip two coins repeatedly you tend towards 25% HH, 25% TT, and 50% H+T in whatever order. OP's question loses the 25% for TT, leaving the 50% H+T and 25% HH - scaling to 100%, that's 2/3 and 1/3.

It's not correct.

This is 1/3

Odds are a human construct and have nothing to do with nature

This guy is an idiot.

1. Probability that both are head and at least one head=1/4
2. Probability of at least one head= 3/4

3. (1/4) / (3/4) = 1/3

Why does 25 / 0.7 = a diffrent answer to 25 x 1.3? In theroy they should bofh work ou 30%?

No, *0.7 would be a 30% change. Why would you divide?

Because that's what you do. It's how I work margins out.