Points on the Same Semicircle

The probability that all 6 6 randomly chosen points on the circumference of a circle can be covered by the same semicircle has the form a b \frac {a}{b} where a , b a, b are coprime integers. What is a + b a+b ?


The answer is 19.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

5 solutions

Jatin Yadav
Nov 19, 2013

Let us generalize this problem:

Say we have to choose n n points on the circumference such the lie within the same semicircle.

We can try fixing any of the n n points and then multiply by n n as total no. of fixings could be n n .

All of remaining n 1 n-1 points must lie within a semicircle around the fixed point,

the probability for a given point to lie within a semicircle = 1 2 = \frac{1}{2} ,

Hence, for n 1 n-1 points , probability = 1 2 n 1 \frac{1}{2^{n-1}} .

Finally, multiplying by n n ,

Required probability = n 2 n 1 \frac{n}{2^{n-1}}

Here n = 6 n = 6 , hence P = 3 16 P = \boxed{\frac{3}{16}}

Mark Hennings
Nov 18, 2013

Suppose that there are n n points to be placed on the circle. Let p j p_j be the probability that all n n points lie in the same semicircle, with the j j th point the "first" point occurring in that semicircle (going around the circle in an anticlockwise direction). This is just the probability that the other n 1 n-1 points lie in that semicircle defined with the j j th point at its start (going anticlockwise), and hence p j = 2 1 n p_j \; = \; 2^{1-n} It is clear that the desired probability is just (since the probabilities p 1 , p 2 , , p n p_1,p_2,\ldots,p_n represent disjoint events) p 1 + p 2 + + p n = n 2 1 n p_1 + p_2 + \cdots + p_n \; =\; n2^{1-n} which equals 3 16 \tfrac{3}{16} when n = 6 n=6 , gving the answer 3 + 16 = 19 3+16=19 .

Seriously, don't you think this problem was pretty straightforward for a 300 points problem? :P

Sreejato Bhattacharya - 7 years, 6 months ago

Log in to reply

Maybe, but it was not as bad as the SHM problem in Mechanics last week! That was 300 marks for one line.

Mark Hennings - 7 years, 6 months ago

Log in to reply

Agreed!

Sreejato Bhattacharya - 7 years, 6 months ago

Yep, the level of 300-pt. problems is decreasing! :(

Sometimes , the point distribution goes unfair mostly in physics, as 300-pt. problem seems easier than 230-pt. or even 160pt. one , you take example from previous week!

jatin yadav - 7 years, 6 months ago

Let S = { P 1 , P 2 , , P 6 } S= \{P_1, P_2, \cdots, P_6\} be a random selection of six points on the circumference of a circle. Let O O be the center of this circle. For convention we take P 1 O P 1 \angle P_1OP_1 to be 0 0 . Then any other point P i P_i on the circumference will be uniquely described by the angle P i O P 1 \angle P_iOP_1 . We then assign a coordinate θ i \theta_i to each of the P i P_i s, which measures the angle P i O P 1 \angle P_iOP_1 .

Note: For convention, we take the anticlockwise angles to be positive, and clockwise angles to be negative.

Note that each of the θ i \theta_i 's are either positive, or 0 0 , or non-negative, but are pairwise distinct. Now, note that if θ 2 > 0 \theta_2>0 , the necessary and sufficient condition for the six points to lie in the same semicircle is θ 3 , θ 4 , θ 5 , θ 6 > 0 \theta_3, \theta_4, \theta_5, \theta_6>0 . Since the probability that each θ i \theta_i is positive is equal to the probability that each θ i \theta_i is negative (from symmetry), the probability of each θ i \theta_i being positive is 1 2 \frac{1}{2} . Thus, this case gives the probabiity 1 2 × 1 2 × 1 2 × 1 2 × 1 2 × 1 2 = ( 1 2 ) 6 = 1 64 \frac{1}{2}\times\frac{1}{2}\times\frac{1}{2}\times\frac{1}{2}\times\frac{1}{2} \times \frac{1}{2}= \left ( \frac{1}{2} \right ) ^6= \frac{1}{64} Similarly, if θ 2 < 0 \theta_2<0 , we get the same probability. Summing them up, we get the probability 1 64 + 1 64 = 1 32 \frac{1}{64} + \frac{1}{64}= \frac{1}{32} However, note that this is not the answer. We are taking the point P 1 P_1 as the measuring point. To count the total probability, we need to count the probability by taking each of the other P i P_i s as the measuring points. By symmetry, this will just increase the answer 6 6 times, so our desired answer is 6 × 1 32 = 3 16 6 \times \frac{1}{32}= \frac{3}{16} Which gives the sum 3 + 16 = 19 3+16=\boxed{19} .

Shubham Raj
Nov 17, 2013

Consider a simple circle with n random points on its circumference. Lets take any reference point. We can do this in C(n,1) ways. Now we are left with n-1 points. Let us, without loss of generality consider that these n-1 points lie on the semi-circle in clockwise sense to out reference points. Each point has a probability of 1/2 to lie in the clockwise circle or not, yielding us a probability 1/2^(n-1). Since we have a choice of C(n,1) reference points, our required answer is C(n,1)/2^(n-1) We shall substitute n=6 for this question.

Michael Mendrin
Mar 27, 2014

I should come back and do this in Latex. But the proof is somewhat straightforward. For large n, given 2n points on a circle, so that any semicircle contains n points, there are (2n)^6 ways 6 points can fall on the circumference. Given an arc containing x points, and given that 2 points fall on its endpoints, there are x^4 ways for 4 points to fall on it, and thus (2n)(6)(5)x^4 ways for 6 points to fall on an arc containing x points. We add up all the ways 6 points can fall on an arc containing 0 to n points, and divide it by (2n)^6.

(2n)^(-6) (2n)(6)(5) ∫ (x = 0 to n) x^4 dx = 3/16

We can use this method for the general case of k points, the probability being k 2^(1-k) for all points falling on a semicircle.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...