Quote:
Originally Posted by somigosaden
There are six questions, 1 through 6, and six answers, A through F. I have no idea what the questions or the answers are; I just have to hope to match them up.
I'm guessing there are 6! ways to pair the numbers with the letters.
I'm not a mathematician. I appreciate mathematicians. I love mathematicians.
But anyhow, 6! seems right to me. And 6!=720
Quote:
So my chance of picking all of them right is 1/720.
Yes. (I think so too...
1/720 for all of then right).
Quote:
Somehow my chance of getting at least five right must also be 1/720 because I can't get five right without getting six right.
There's no way to get just five of them right.
Thus there are 0/720 ways to get exactly five right.
Quote:
My chance of getting at least four right is 1/360
No.
Quote:
because half the time when I get four right I get the fifth right, which means I get the sixth right. I suppose then that the chance of getting exactly four right is also 1/720, since half the time when I get four, I also get six.
I don't think that logic works. I try to just use logic I know works.
To get four right, you must have two wrong. It's like having a row of six marbles, four of them white and two of them black. A white marble represents correct placement while a black marble represents incorrect placement.
To have two black,
if the first marble is black, then there are five ways to have another marble be black. And that's it for the first marble.
if the second marble is black, then there are four ways to have another marble (aside from the first) be black. And that's it for the second marble.
if the third marble is black, then there are three ways to have another marble (aside from the first and second) be black. And that's it for the third marble.
if the fourth marble is black, then there are two ways to have another marble (aside from the first, second and third) be black. And that's it for the fourth marble.
if the fifth marble is black, then there is one ways to have another marble (aside from the first, second, third, and fourth) be black. And that's it for the fifth marble.
We've already counted all the ways the sixth marble can be black.
5+4+3+2+1=15. That's the number of ways 2 marbles can be black. Thus that's the number of ways two answers can be exchanged and thus the number of ways exactly four answers can be correct
Quote:
If I get the first three right, I have a 1/3 chance of getting at least four right, so maybe it stands to reason that I have a 1/120 chance of getting at least three right. When I get the first three right, there is a 1/3 chance I'll miss all of the next three, so my chance of getting exactly three right is 1/360.
I don't think that logic works.
For exactly three answers to be wrong, first we choose the three answers that will be wrong, then choose how many ways they can be wrong. Let's do the second part first. If the order is ABCDEF, and we choose ABC to be wrong, there are six (3!) ways to arrange ABC (ABC, ACB, BCA, BAC, CAB, CBA) but it's immediately obvious there are only two arrangements with all three out of place).
Now we just have to decide how many ways we can arrange three white and three black marbles in a row.
000111
001011
001101
001110
010011
010101
010110
011001
011010
011100
100011
100101
100110
101001
101010
101100
110001
110010
110100
111000
If I counted correctly, there are 20 of these. And each one has exactly three misplaced twice.
So 20*2=40. There are 40 ways to have exactly three misplaced (and thus exactly three correctly placed) answers.
To have two correctly placed, consider again black and white marbles. There are 15 ways to arrange 2 white and 4 black marbles. But now when we look at the four black marbles that are misplaced, think of ABCD. 4!=24 possibilities. Here they all are:
ABCD no
ABDC no
ACBD no
ACDB no
ADBC no
ADCB no
BACD no
BADC yes
BCAD no
BCDA yes
BDAC yes
BDCA no
CABD no
CADB yes
CBAD no
CBDA no
CDAB yes
CDBA yes
DABC yes
DACB no
DBAC no
DBCA no
DCAB yes
DCBA yes
I think that's it. 24, (4!), of those, but only 9 (those marked yes) have all four letters out of sequence.
So again, if I have counted correctly,
15*9=135. That's how many ways to have exactly two correct answers.
There are 720 arrangements possible and we have only accounted for 1+0+15+40+135=191 of them. There are 529 more arrangements. Some of those arrangements have one answer correctly placed and the rest have no answers correctly placed. If we find one, we can subtract from 529 to get the other.
I'll come back to this problem tomorrow. Maybe I'll think of something in the meanwhile. Almost sixty years ago as a junior in college I took a graduate math class for fun. I couldn't solve all the problems on the final exam. Several months after the class was finished, I was digging a ditch and not thinking about the problems on that final exam... and the solution to one of the problems popped into my head! Perhaps I had been subconsciously thinking about the problem all that time. Anyhow, maybe I'll subconsciously think about how to put either exactly one or exactly no answers in the correct place.
Quote:
If I get two right, my chance of getting the next one right is 1/4, so my chance of getting at least two right is 1/30.
I don't think that works.
Quote:
At this point I can't count in my head the combos of getting all the next ones wrong, so I'm not sure about my chance of getting exactly two right. But I'll continue my pattern to say that my chance of getting at least one right is 1/6, and my chance of getting at least zero right is 6/6. It also follows that my chance of missing them all is 5/6, which can't possibly be right.
I agree.
Quote:
Obviously there's some cascading error in my thinking, but I don't know where it is.
I don't know what a cascading error is.
Quote:
I'd love to know how to solve this for all cases, but at the very least for the cases of getting at least one right and getting exactly one right.
nickthegeek got the answers for 6, 5, 4, 3, and 2 correctly placed right. (At least we got the same answers). My guess is he also got 1 and 0 right... but I haven't figured out how to do either yet... maybe tomorrow.
It's a fun problem for me. Thanks.
Buzz