Exercises: Probability, Day 1

There are a lot of exercises today. Do not feel that you have to answer every single one.

P1.1 A pair of six-sided dice are rolled and their sum is recorded.
(a) What is the sample space in this experiment?

(b) If both dice are fair then what is the most likely sum?
(c) If both dice are fair and the first die rolls a 1 then what is the probability that the second die rolls a 1?

Solution: If we add two numbers between 1 and 6 together, then the total of those two numbers varies
between 2 and 12. Therefore, the sample space is

Ω = {2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}

P1.2 The heights of a football team are recorded, and their average is taken. What is the sample space of this
experiment?

Solution: The sample space for this experiment is the set of all possible averages we could get from
measuring people’s heights. This should therefore include all possible heights a person could be. This is
kind of open to interpretation – you might have to ask yourself what is the shortest/tallest a human could
possibly be? To be safe, you could always say your sample space is the set of non-negative numbers.

Ω = [0, ∞)

P1.3 The following game is played: A six sided dice is rolled, and a coin is flipped. If the coin shows heads, then
your final score is found by doubling the number shown on the dice. If the coin shows tails, then your final
score is found by halving the number shown on the dice. What is the sample space for the possible scores
you can achieve?

Solution: Every outcome is either half of a number between 1 and 6, or double a number between 1
and 6. Therefore, the sample space is

Ω = {0.5, 1, 1.5, 2, 2.5, 3} ∪ {2, 4, 6, 8, 10, 12}

= {0.5, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 10, 12}
P1.4 Brennan is playing a game which involves rolling fair twenty-sided dice (d20s). Rolling with advantage means
rolling a pair of d20s and keeping the higher result. Rolling with disadvantage means rolling a pair of d20s
and keeping the lower result.
(a) What is the probability that his result is a 20 when rolling with disadvantage?
(b) What is the probability that his result is a 20 when rolling with advantage?

Solution: In each part of this question we’re looking at a pair of d20 rolls. Note that we can represent
the sample space as a Cartesian Product

Ω = {1, 2, ..., 20} × {1, 2, ..., 20},

from which we can tell that there are 20 × 20 = 400 possible outcomes (|Ω| = 400). This will be useful
going forward.

(a) This is the easier part. To get a 20 with disadvantage we need to roll 20 on both dice, so this
event can be represented by the set D = {(20, 20)}. This has only one outcome (|D| = 1), so the
corresponding probability is
|D| 1
P (D) = =
|Ω| 400
.

(b) This is the harder part. To get a 20 with advantage we need to roll a 20 on die 1, or die 2. We can
represent this event as the following set

A = {(20, 1), (20, 2), ..., (20, 19), (20, 20), (19, 20), (18, 20), ..., (2, 20), (1, 20)}, (1)

This set has 20 + 19 = 39 elements (|A| = 39). Therefore the probability of this event is

|A| 39
P (A) = =
|Ω| 400

P1.5 A pair of fair four-sided dice are rolled. What is the probability that
(a) the largest number shown on either of them is a 1?

(b) the largest number shown on either of them is a 2?

(c) the largest number shown on either of them is a 3?
(d) the largest number shown on either of them is a 4?

Solution: Labelling the dice A and B, we can represent the possible outcomes from recording the largest
number shown on either of them with this table.
B 1 2 3 4
A
1 1 2 3 4
2 2 2 3 4
3 3 3 3 4
4 4 4 4 4

All outcomes are equally likely. Therefore, for each x ∈ {1, 2, 3, 4}, the probability that the largest

Page 2
 number shown on either of them is x is the number of times x appears in the table, divided by the total
number of entries in the table (which is 16). So we have

(a) 1/16 (b) 3/16 (c) 5/16 (d) 7/16

P1.6 You purchase three stocks at the start of the day. At the end day each stock has an equal chance of going
up by £10, or down by £20, (completely independently of what the other stocks do). What is probability
that your combined stocks are worth less, at the end of the day, than what you paid for them?
Hint: Think about you how might represent the different combinations of stocks going up or down, and thus
what the resulting change in value is.

Solution: We could represent each possible outcome of stocks going up or down by a sequence of three
U’s and D’s, where, for example if the first stock goes up and the other two stocks go down, we would
write UDD. The possible outcomes are then

Ω = {UUU, UUD, UDU, UDD, DUU, DUD, DDU, DDD}

The total change of value of the stocks at the end of the day corresponding to each outcome is then
represented by the table.

UUU UUD UDU UDD DUU DUD DDU DDD

+30 0 0 -30 0 -30 -30 -60

Each of our outcomes are equally likely, so the probability that our stocks are worth less at the end of
the day is the number of negative outcomes divided by the total number of possible outcomes.
4 1
=
8 2

P1.7 (a) A word is selected at random from this sentence.

i. What is the probability that the word is “from”?
ii. What is the probability that the word has more than 5 letters?
iii. What is the probability that the word contains the letter ‘a’ or ‘o’ ?
iv. What is the probability that the word is also in this sentence?

(b∗ ) Write a sentence so that when a word is selected from it at random:

• the probability it is an animal is 1/3; and
• the probability it contains the letters ‘a’ or ‘o’ is 1/2.

Solution:

(a) i. 1/9 ii. 1/3 iii. 4/9 iv. 4/9

Page 3
 (b) One possibility might be:
“The lion saw the huge elephant.”

P1.8 You are interested in the weather over a working week: Monday, Tuesday, Wednesday, Thursday, Friday.
Each day, it is equally likely that it will be rainy or be sunny. You are given that there 32 different possible
rain and sun combinations across the working week. What is probability that it will rain on:
(a) Monday and Tuesday only;
(b) Monday and Tuesday;

(c) Exactly two of the days.

Hint: For part (c), you know how many possible outcomes there are, you just need to find a way of counting
the outcomes that have rain on exactly two days.

Solution: We could represent each possible outcome for the weather over a working week by a sequence
of five R’s and S’s, where, for example if it was rainy on Monday and Tuesday but sunny on Wednesday,
Thursday and Friday, then we would write RRSSS. We are told in the question that there are 32 possible
sequences we could write like this, and we can see that each is equally likely (as sun and rain are equally
likely as each other).

(a) There is only one outcome where it rains on Monday and Tuesday only: RRSSS. So the probability
that it will rain on Monday and Tuesday only is 1/32.
(b) There are eight possible outcomes where it rains on Monday and Tuesday:

RRSSS, RRSSR, RRSRS, RRSRR, RRRSS, RRRSR, RRRRS, RRRRR

Therefore, the probability that it rains on Monday and Tuesday is

8 1
=
32 4

(c) We need a way to count the number of outcomes where it rains on exactly two days. Here’s one
way of doing this:
• There are 4 two-rainy-day outcomes for which the first rainy day is Monday: RRSSS, RSRSS,
RSSRS, RSSSR.
• There are 3 two-rainy-day outcomes for which the first rainy day is Tuesday: SRRSS, SRSRS,
SRSSR
• There are 2 two-rainy-day outcomes for which the first rainy day is Wednesday: SSRRS,
SSRSR
• There are 1 two-rainy-day outcomes for which the first rainy day is Thursday: SSSRR
• There are 0 two-rainy-day outcomes for which the first rainy day is Friday (if the first rainy
day is Friday, then there can only be one rainy day).
So the total number of outcomes that have rain on exactly two days is 4 + 3 + 2 + 1 + 0 = 10,
meaning the probability it rains on exactly two days is
10 5
=
32 16

Page 4
P1.9 Alice and Bob have cards with numbers on. Alice has three cards: 1, 3, and 4. Bob has four cards: 1, 2, 3,
and 5. They each pick one of their cards at random.
(a) What is the probability they both pick a 3?
(b) What is the probability they both pick odd numbers?

(c) What is the probability that the total of their two cards is at least 8?
(d) What is the probability that Bob’s card is higher than Alice’s card.

Solution: We can represent each of the possible outcomes as a pair of numbers (x, y), where x is the
number on Alice’s card and y is the number on Bob’s card, with each outcome being equally likely.
These outcomes are shown in the table below.
B 1 2 3 5
A
1 (1,1) (1,2) (1,3) (1,5)
3 (3,1) (3,2) (3,3) (3,5)
4 (4,1) (4,2) (4,3) (4,5)

We note that there are 12 possible outcomes in total.

(a) There is only one outcome where they both pick a 3: (3,3). So the probability they both pick a 3
is 1/12.
(b) There are six outcomes where they both pick odd numbers:

(1,1), (1,3), (1,5), (3,1), (3,3), (3,5)

Therefore, the probability they both pick odd numbers is

6 1
=
12 2

(c) There are only two outcomes where the total of the cards is at least 8: (3,5) and (4,5). Therefore,
the probability is
2 1
=
12 6
(d) There are five outcomes where Bob’s card is higher than Alice’s card: (1,2), (1,3), (1,5), (3,5), (4,5).
Therefore, the probability is 5/12.

Page 5
P1.10∗ Three coloured six-sided dice have different numbers on their faces. The numbers on the faces of each die
are as follows:
• Red die: 3, 3, 3, 3, 3, 6.
• Blue die: 2, 2, 2, 5, 5, 5.

• Yellow die: 1, 4, 4, 4, 4, 4.
You use these dice to play a game with a friend. Each player picks a die. The two dice are then rolled
together and whoever gets the highest value wins.
(a) i. What is the probability the red die beats the blue die?
ii. What is the probability the blue die beats the yellow die?
iii. What is the probability the yellow die beats the red die?
(b) When playing this game, is it better to choose your die first, or let your friend pick before you?

Solution: We shall represent each of the dice with letters according to their colours: R, B, and Y. The
table below shows the outcomes when each dice is rolled against each other, with the winner for each
outcome indicated, and all outcomes equally likely.
B 2 2 2 5 5 5
R
3 R R R B B B
3 R R R B B B
3 R R R B B B
3 R R R B B B
3 R R R B B B
6 R R R R R R
Y 1 4 4 4 4 4
B
2 B Y Y Y Y Y
2 B Y Y Y Y Y
2 B Y Y Y Y Y
5 B B B B B B
5 B B B B B B
5 B B B B B B
R 3 3 3 3 3 6
Y
1 R R R R R R
4 Y Y Y Y Y R
4 Y Y Y Y Y R
4 Y Y Y Y Y R
4 Y Y Y Y Y R
4 Y Y Y Y Y R

(a) i. From the first table, we can deduce that the probability the red die beats the blue die is
21 7
=
36 12
ii. From the second table, we can deduce that the probability the blue die beats the yellow die is
21 7
=
36 12

Page 6
 iii. From the third table, we can deduce that the probability the yellow die beats the red die is
25
36

(b) It’s better to let your friend pick first – no matter what die they pick, you can pick one of the others
so that your chance of winning is more than 50%.
This exercise describes non-transitive dice, which you can read more about here: https://singingban
ana.com/dice/article.htm

P1.11 Suppose we have a sample space Ω = {1, 2, Red, 5.5, π, 7, 8, 9}. Let A = {2, 5.5, π}, B = {Red, π} and
C = {1, 15, 2}.
(a) Which of the sets A, B and C are events for Ω?

(b) Find A ∪ B and A ∩ B.

(c) What is B c ?
(d) What is B c ∩ A?

Solution:

(a) A set E is an event for Ω if it is a subset of Ω.

• A ⊆ Ω, therefore A is an event for Ω.
• B ⊆ Ω, therefore B is an event for Ω.
• C ̸⊆ Ω (e.g. 15 ∈ C, but 15 ∈
/ Ω). Therefore, C is not an event for Ω.
(b)

A ∪ B = {2, 5.5, π, Red}

A ∩ B = {π}

(c) B c is the set of outcomes in Ω that are not in B.

B = Ω \ B = {1, 2, 5.5, 7, 8, 9}

(d) Using our answer to part (c),

B c ∩ A = {2, 5.5}

Page 7
P1.12 Suppose we flip three coins, using the sample space

Ω = {HHH, HHT, HTH, HTT, THH, THT, TTH, TTT}.

Describe the following events using set notation:

(a) We observe precisely two heads.
(b) The second coin is a tails.
(c) The first coin is a heads and the second is not a tails.
(d) There is at least one head, but the last coin is a tails.

Solution:

(a) {HHT, HTH, THH}

(b) {HTH, HTT, TTH, TTT}

(c) {HHH, HHT}

(d) {HHT, HTT, THT}

P1.13 Suppose Ω = {1, 2, 3, 4, 5, 6, 7} and let A = {2, 3}. Find an event B such that Ac ∩ B = {1, 5, 6, 7}.

Solution: To help us get to what we’re aiming for, let’s first write down what Ac is.

AC = Ω \ A = {1, 4, 5, 6, 7}

So we need to write down a set B such that

• B is an event, i.e. B ⊆ Ω
• {1, 4, 5, 6, 7} ∩ B = {1, 5, 6, 7}

One simple solution is for us to set

B = {1, 5, 6, 7}.
We should note that this is not the only choice for B that solves this question – there are several others
too!

P1.14 Consider the following sample space with events A and B. On a Venn diagram:
(a) Indicate the region (A ∪ B)c .

(b) Indicate the region Ac ∩ B c .

(c) What do you notice? What does this tell you about the events (A ∪ B)c , and Ac ∩ B c . (This is known
as De Morgan’s Law.)

Page 8
 Solution:

(a) To indicate the region (A ∪ B)c , we should shade all regions of the Venn diaram that are not in the
region A ∪ B (in words, “not in (A or B)”).

(b) To indicate the region Ac ∩ B c , we should shade any region of the Venn diagram that is not inside
A, and not inside B.

(c) We notice that the two diagrams are the same! This tells us

(A ∪ B)c = Ac ∩ B c

P1.15 At a library, books can be either paperback or hardback, and either fiction or non-fiction. Out of a total of
50 000 books: 30 000 books are fiction books, 15 000 books are hardbacks. In addition, 25 000 of the fiction
books are paperbacks. As I always do, I select a book from the library at random for me to read. What is
the probability that the book I read is a non-fictional hardback? Hint: Depending on how you formulate
this problem, it may be helpful to use De Morgan’s Law (see Question P1.14).

Solution: Let A be the event that the book selected is fiction book, and let B be an event that the
book is a hardback. From the information in the question, we have
30 000 3
P (A) = =
50 000 5
15 000 3
P (B) = =
50 000 10
25 000 1
P (A ∩ B c ) = =
50 000 2
We are interested in the event that the randomly selected book is a non-fictional hardback – so this is
the event Ac ∩ B. By the inclusion-exclusion principle, we have
P (Ac ∪ B) = P (Ac ) + P (B) − P (Ac ∩ B) .

Page 9
Rearranging this gives
P (Ac ∩ B) = P (Ac ) + P (B) − P (Ac ∪ B) . (2)
c
We know that P (A ) = 1 − P (A) and, by applying De Morgan’s Law, we have that

P (Ac ∪ B) = 1 − P ((Ac ∪ B)c ) = 1 − P ((Ac )c ∩ B c ) = 1 − P (A ∩ B c ) .

Therefore, we can substitute these facts into (2) and use our known probabilities to find P (Ac ∩ B).
3 3 1 1
P (Ac ∩ B) = 1 − P (A) + P (B) − (1 − P (A ∩ B c ) = 1 − + − (1 − ) =
5 10 2 5
Note that this question can also be solved by using the known probabilities to find the region indicated
in the Venn diagram below – following that method might be more intuitive than the calculations given
above.

Please email any typos, mistakes, corrections to Ben McCanna at b.d.p.mccanna@bham.ac.uk.

Page 10