Lesson Library> Probability

Probability

You have been chosen to participate in a lottery show on TV. The host shows you three doors. Behind one of them is a brand new Ferrari, but behind the other two doors is a goat. You get to choose one of the doors and keep the object behind it. 

You decide to choose the door to the left. The host then opens the door on the far right to show that there was a goat behind it. You are now given two options: stick with your first choice or switch to the door in the middle. What should you do?

Most people would think the probability is 50%-50% of getting the Ferrari no matter if you switch or not. But your chance of getting the Ferrari is actually twice as high if you choose to switch. How could this be? 

In this lesson you’ll learn everything you need to know about probabilities to solve this question!

Introduction to Probability

Probability is a topic that is closely linked to combinatorics. The probability that an event $X$ will happen is simply the number of ways that $X$ could happen divided by the total number of possible outcomes. Let us take the following problem as an example.

Daniella has two bags. One contains three green and two blue marbles and the other contains two green marbles and one blue marble. She chooses one marble from each bag. What is the probability that both marbles are blue?

We can choose a blue marble from the first bag in two ways and a blue marble from the second bag in one way. Daniella can thus choose two blue marble in $2$ ways. The total number of ways she can choose two marbles is $5 \cdot3 = 15$. The probability that she chooses two blue marbles is thus $\pmb{\dfrac{2}{15}}$.

Even though the two blue marbles in the first bag look identical they are not the same marble. That is why there are two ways to choose one blue marble from each bag. This becomes clear if we number all the marbles in both of the bags in the following way: 

We can then display all $15$ ways of choosing one marble from both bags with this $5 \times 3$ matrix:


Both marbles are blue in $2$ of the $15$ possible outcomes. So the probability of this happening must therefore be $\dfrac{2}{15}$. 

The probability that an event $X$ occurs is written as $P(X)$. We can thus express the probability that Daniella chooses two blue marbles as $P(\text{two blue marbles})$ or $P(\text{blue-blue})$. In principle, all probability problems can be solved with the combinatorial approach: 

$$P(X) \ = \ \dfrac{\text{#Ways that X could happen}}{\text{#Total outcomes}}$$

However, this method is not always the most efficient. Another method is to multiply the probabilities with each other.

Multiplying Probabilities

Let's say that Daniella chooses a blue marble from the first bag. Her chances of now picking a blue marble from the second bag is $\dfrac{1}{3}$. The probability of both marbles being are blue marbles is therefore $\dfrac{1}{3}$ of the probability that the first marble was blue $\Big(\dfrac{2}{5} \Big)$. 

To get one third of anything we simply multiply by $\dfrac{1}{3}$. The probability that Daniella picks two blue marbles is therefore:

$$\dfrac{2}{5} \ \cdot \ \dfrac{1}{3}=\mathbf{\dfrac{2}{15}}$$

This method is generally applicable. If you have two or more events, the probability that all events will occur is the product of all the individual probabilities. This is often called the probability multiplication rule: 

$$P(A) \ \cdot \ P(B) \ = \ P(A \ \text{and} \ B)$$

$$P(A) \ \cdot \ P(B) \ \cdot \ P(C) = \ P(A \ \text{and} \ B \ \text{and} \ C)$$


Exercise 1

What is the probability of getting three heads on three coin flips?

The probability to get a head once is $\frac{1}{2}$ so the probability that she gets it three times in a row is $\frac{1}{2} \cdot \frac{1}{ 2} \cdot \frac{1}{2} = \pmb{\frac{1}{8}}$


Exercise 2

Sofie practices basketball. The probability that she scores the ball in the basket at any given throw is 60%. What is the probability that she first scores, then misses and then scores again?

The probability that she scores the ball is $\frac{3}{5}$ and the probability that she misses is $\frac{2}{5}$. So the probability that she scores, misses and then scores is $\frac{3}{5} \cdot \frac{2}{5} \cdot \frac{3}{5} = \pmb{\frac{18}{125}}$.


Exercise 3

Jacob picks two marbles from a bag of $6$ green and $4$ yellow marbles. What is the probability that both marbles are green?

The probability that the first marble is green is $\frac{6}{10}$. The second time Jacob picks a marble there are only $9$ left. So the probability that the second marble is green is $\frac{5}{9}$. The probability that Jacob picks two green marbles is therefore $\frac{6}{10} \cdot \frac{5}{9} = \frac{30}{90} = \pmb{\frac{1}{3}}$.


Dependent and Independent Events

In the previous section we showed that the probability of both $A$ and $B$ occurring is equal to the product of both individual probabilities: 

$$P(A) \ \cdot \ P(B) \ = \ P(A \ \text{and} \ B)$$

However, this is not entirely true. Multiplying individual probabilities is only possible if the events are independent, i.e the probability that $B$ occurs is the same regardless of whether $A$ occurred or not. Flipping a coin twice and getting heads on both flips is an example of two independent events. The outcome of the first flip has no effect on the likelihood of getting heads one the second flip. 

But sometimes events are dependent, i.e the probability that $B$ occurs changes depending on whether $A$ occurred or not. For example, if the weather is foggy there is a higher likelihood of ending up in a car accident compared to if the weather is not foggy. When events are dependent we can’t take the product of the individual probabilities $P(A)$ and $ P(B)$ as they are. We must first adjust $P(B)$ depending on whether $A$ occurred or not. 

This is what we did in Exercise 3 on the previous page. The probability of choosing a green marble was $\frac{6}{10}$. But after the first green marble had been chosen the likelihood of choosing another green marble was $\frac{5}{9}$. The probability of picking two green marble was therefore not $\frac{6}{10} \cdot \frac{6}{10}$ but $\frac{6}{10} \cdot \frac{5}{9}$. When we have adjusted the probability of $B$ depending on whether $A$ occurred, we can multiply the individual probabilities to determine the likelihood of both events occurring.

The probability that $B$ occurs given that $\pmb{A}$ has already occurred is written as $P(B \ | \ A)$. 

$$P(B \ | \ A) \ = \ P(B \ \text{given that} \ A \ \text{has occurred})$$ 

The probability of picking two green marbles in Exercise 3 could therefore have been written as: 

$$P(\text{(green}) \ \cdot \ P(\text{(green | green}) \ = $$

$$\dfrac{6}{10} \ \cdot \ \frac{5}{9} = \frac{1}{3}$$ 

The general probability multiplication rule is written as:

$$\pmb{P(A) \ \cdot \ P(B \ | \ A) \ = \ P(A \ \text{and} \ B)}$$ 

or

$$\pmb{P(B) \ \cdot \ P(A \ | \ B) \ = \ P(A \ \text{and} \ B)}$$

We can also rearrange this equation to solve for the individual probabilities:

$$P(A) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(B \ | \ A)} \ \ \ \ \ \ \ P(B) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(A \ | \ B)}$$

The general multiplication rule always holds true. If $A$ and $B$ are independent then $P(A) = P(A \ | \ B)$ and $P(B) = P(B \ | \ A)$, so the equations can be simplified to:

$$P(A) \ \cdot \ P(B) \ = \ P(A \ and \ B)$$ 

$$P(A) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(B)} \ \ \ \ \ \ \ P(B) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(A)}$$

For independent events we can therefore take the product of the individual probabilities without having to adjust them. If the events are dependent, however, the probability that $B$ occurs is now adjusted thanks to the expression $P(B \ | \ A)$.


Exercise 1

Which of the following are examples of independent events?

a) That it's raining and that it's cloudy.

b) To roll a die and get an even number three times in a row.

c) To make a left and a right turn at the same time.

d) To rob a bank and go to jail.

e) Getting stuck in traffic on the way home and winning the lottery.

a) It’s impossible for it to rain unless it is cloudy. The events are dependent.

b) The outcome of one die roll doesn’t affect the probability of getting an even number on a second die roll. The events are independent.

c) Making a left and a right turn cannot occur at the same time. The events are dependent.

d) The probability of going to jail is obviously much higher if we rob a bank compared to if we do not. The events are dependent.

e) The events of getting stuck in traffic and winning the lottery obviously have no impact on one another. The events are independent.


Example 1

If $P(A) = P(A \ | \ B)$, is it true that $P(B) = P(B \ | \ A)$?

Solution

The general multiplication rule states that: 

$$P(A) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(B \ | \ A)} \ \ \ \ \ \ \ P(B) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(A \ | \ B)}$$

If $P(A) = P(A \ | \ B)$, the right version of the rule above can be simplified to: 

$$P(B) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(A)}$$

Which can be rearranged to:

$$P(A) \ \cdot \ P(B) \ = \ P(A \ \text{and} \ B)$$

$$P(A) \ = \ \dfrac{P(A \ \text{and} \ B)}{P(B)}$$

If we compare this equation with the left version of the general multiplication rule above we see that $P(B) = P(B \ | \ A)$. It is therefore true that: 

$$P(A) = P(A \ | \ B) \ \ \iff \ \ P(B) = P(B \ | \ A)$$


Exercise 2

For two events $A$ and $B$ we know that:

$$P(A) \ \cdot \ P(B) \ = \ P(A \ \text{and} \ B)$$ 

Can we be $100$% certain that the events are independent?

Two events are independent if and only if $P(A) = P(B \ | \ A)$ or $P(B) = P(B \ | \ A)$. The general multiplication rule states that:

$$P(A) \ \cdot \ P(B \ | \ A) \ = \ P(A \ \text{and} \ B)$$

If we compare this with the equation given in the question, we can see that $P(B) = P(B \ | \ A)$. So if: 

$$P(A) \ \cdot \ P(B) \ = \ P(A \ \text{and} \ B)$$ 

we can be $100$% certain that the events $A$ and $B$ are independent. 


Exercise 3

In New Orleans, the weather is sunny two out of five days and windy five out of twelve days on average. The likelihood of the weather being both sunny and windy on any given day is $\frac{1}{7}$. Are the events sunny and windy weather dependent or independent?

The events of sunny and windy weather are independent if and only if $P(\text{sunny}) \cdot P(\text{windy}) =$ $P(\text{sunny and windy})$. But from the question we know that $P(\text{sunny}) \cdot P(\text{windy}) =$ $\frac{2}{5} \cdot \frac{5}{12} = \frac{1}{6}$ while $P(\text{sunny and windy}) \ =$ $\frac{1}{7}$. The events of sunny and windy weather are therefore dependent.


Exercise 4

A preschool class consists of eight girls and nine boys. If three students are chosen at random, what is the probability that the first is a girl and the other two are boys?

The probability that the first student is a girl is $\frac{8}{17}$. Next time we select a student there are only $16$ children left, so the chance of that student being a boy is $\frac{9}{16}$. When we choose the third student there are $8$ boys left from a total of $15$ students, so the probability that we choose another boy is $\frac{8}{15}$. It therefore follows that $P(\text{girl-boy-boy}) = $ $\frac{8}{17} \cdot \frac{9}{16} \cdot \frac{8}{15} \ =$ $\frac{1}{17} \cdot \frac{3}{1} \cdot \frac{4}{5} \ =$ $\pmb{\frac{12}{85}}$.


Probability and Venn Diagrams

One way to illustrate outcomes and their probability of occurrence is by using venn diagrams. (If you are not familiar with this topic, you can visit the lesson Sets & Venn Diagrams).

Let's say that there are two possible events, $A$ and $B$, with the following probabilities: 

  • $P(\text{only} \ A) = \frac{15}{60}$
  • $P(\text{only} \ B) = \frac{1}{60}$
  • $P(A \ \text{and} \ B) = \frac{9}{60}$
  • $P(\text{neither} \ A \ \text{nor} \ B) = \frac{35}{60}$

We can illustrate this with the following venn diagram: 


This venn diagram is drawn to scale, which means that the area of ​​each region in relation to the whole rectangle is the probability that this outcome will occur. You may have noticed that the area where both circles overlap covers a very large proportion of circle $B$. This means that $A$ is more likely to occur given that $B$ has already happened, so these events are dependent.


Exercise 1

What is $P(A)$ and $P(B)$ in the diagram above?

$$ P(A) = P(\text{only} \ A) + P(A \ \text{and} \ B) \\ P(A) = \frac{15}{60} + \frac{9}{60} = \pmb{\frac{2}{5}}$$

$$ P(B) = P(\text{only} \ B) + P(A \ \text{and} \ B) \\ P(B) = \frac{1}{60} + \frac{9}{60} = \pmb{\frac{1}{6}}$$


Exercise 2

What is $P(A \ | \ B)$ and $P(B \ | \ A)$? 

We can use the general multiplication rule to help us solve this question: 

$$P(A) \ \cdot \ P(B \ | \ A) \ = \ P(A \ \text{and} \ B) \\ P(B) \ \cdot \ P(A \ | \ B) \ = \ P(A \ \text{and} \ B)$$

Using the given probabilities we get:

$$\Big( \dfrac{15}{60} + \dfrac{9}{60} \Big) \ \cdot \ P(B \ | \ A) \ = \ \dfrac{9}{60}$$

$$P(B \ | \ A) \ = \ \dfrac{\Big(\dfrac{9}{60}\Big)}{\Big(\dfrac{24}{60}\Big)}$$

$$P(B \ | \ A) \ = \ \pmb{\dfrac{3}{8}}$$

And

$$\Big( \dfrac{1}{60} + \dfrac{9}{60} \Big) \ \cdot \ P(A \ | \ B) \ = \ \dfrac{9}{60}$$

$$P(A \ | \ B) \ = \ \dfrac{\Big(\dfrac{9}{60}\Big)}{\Big(\dfrac{10}{60}\Big)}$$

$$P(A \ | \ B) \ = \ \pmb{\dfrac{9}{10}}$$


Exercise 3

Are the events $A$ and $B$ dependent or independent?

Two events $A$ and $B$ are independent if and only if:

$$P(A) \ \cdot P(A) \ = \ P(A \ \text{and} \ B)$$

In this case we have: 

$$P(A) \ = \ \Big( \dfrac{15}{60} + \dfrac{9}{60} \Big) \ = \ \dfrac{3}{8}$$

$$P(B) \ = \ \Big( \dfrac{1}{60} + \dfrac{9}{60} \Big) \ = \ \dfrac{1}{6}$$

$$P(A \ \text{and} \ B) = \frac{9}{60} \ = \ \dfrac{3}{20}$$

And since:

$$\dfrac{3}{8} \ \cdot \ \dfrac{1}{6} \ = \ \dfrac{1}{16} \ \neq \ \frac{3}{20}$$

the events are dependent


Dependent events are easiest to illustrate with circles because it gives us a lot of flexibility. For example, if we want to keep $P(A)$ and $P(B)$ the same but change $P(A \ \text{and} \ B)$, we only need to increase or decrease the overlap between the two circles. If we have two events $A$ and $B$ where both can't occur simultaneously, or if $B$ can only occur given that $A$ has already happened, we can also describe it with the following diagrams:

When two events can't happen at the same time they are called mutually exclusive. The left diagram above shows two mutually exclusive events.

When two events are independent, however, it’s better to use rectangles instead of circles. This way we can more easily illustrate the probability of the different outcomes in relation to the area of ​​each region.


Now for some tougher practise! About three events $A$, $B$ and $C$ we know the following:

  • $P(A) = \frac{7}{15}$
  • $P(B) = \frac{23}{60}$
  • $P(C) = \frac{7}{15}$
  • $P(\text{only} \ A) = \frac{1}{12}$
  • $P(A \ \text{and} \ C) = \frac{13}{60}$
  • $P(A, B \ \text{and} \ C) = \frac{1}{15}$
  • $P(\text{neither} \ A, B \ \text{nor} \ C) = \frac{1}{6}$


Exercise 4

Draw a venn diagram (does not have to be scale) and fill in the following probabilities in the corresponding region:

  • $P(\text{only} \ A) = \frac{1}{12}$
  • $P(A, B \ \text{and} \ C) = \frac{1}{15}$
  • $P(\text{neither} \ A, B \ \text{nor} \ C) = \frac{1}{6}$

What is $P(A \ \text{and} \ C \ \text{but not} \ B)$? 

All fractions above have $60$ as a common denominator. We are therefore going to express all probabilities with denominator $60$ to make our calculations easier to follow:

$P(\text{only} \ A) \ = \ \frac{1}{12} \ = \ \frac{5}{60}$ 

$P(A, B \ \text{and} \ C) \ = \ \frac{1}{15} \ = \ \frac{4}{60}$

$P(\text{neither} \ A, B \ \text{or} \ C) = \frac{1}{6} \ = \frac{10}{60}$ 

All the three regions above can be filled in immediately as none of them overlap with any other regions. 

From the information given in the question we know that $P(A \ \text{and} \ C) = \frac{13}{60}$. Since $P(A, B \ \text{and} \ C) = \frac{4}{60}$ it follows that: 

$$P(A \ \text{and} \ C \ \text{but not} \ B) = \frac{13}{60} - \frac{4}{60} = \pmb{\frac{9}{60}}$$



Exercise 5

What is $P(A \ \text{and} \ B \ \text{but not} \ C)$?

Let $x = P(A \ \text{and} \ B \ \text{but not} \ C)$. We can then fill in the venn diagram as follows:

It now follows that:

$$ P(A) = \frac{5}{60} + \frac{9}{60} + \frac{4}{60} + x = \frac{18}{60} + x $$

But since we know that $P(A) = \frac{7}{15} = \frac{28}{60}$ we have:

$$\frac{18}{60} + x \ = \ \frac{28}{60}$$

$$ x \ = \ \pmb{\frac{10}{60}}$$



Exercise 6

Determine and fill in the probabilities for the remaining regions of the venn diagram. 

Let $y = P(B \ \text{and} \ C \ \text{but not} \ A)$. We can now fill in the venn diagram as follows:

If we now add the circles $B$ and $C$, subtract the region where both circles overlap (because we counted it twice) and then add all remaining regions outside of circles $B$ and $C$ we added every region in the entire diagram. Because the diagram represents every possible outcome that can occur, the sum of all these region must equal $1$ or $\frac{60}{60}$. This gives us:

$$ P(B) + P(C) - \Big( \frac{4}{60} + y \Big) + \frac{5}{60} + \frac{10}{60} \ = \ \frac{60}{60}$$

$$ P(B) + P(C) - y \ = \ \frac{49}{60}$$

We know that $P(B) = \frac{23}{60}$ and that $P(C) = \frac{7}{15} = \frac{28}{60}$. If we input this into the equation we get:

$$\frac{23}{60} + \frac{28}{60} - y \ = \ \frac{49}{60}$$

$$ y = \frac{2}{60}$$

It then follows that:

$$ P(B) \ = \ \frac{23}{60} - \frac{10}{60} - \frac{4}{60} - \frac{2}{60} \ = \ \frac{7}{60}$$

$$ P(C) \ = \ \frac{28}{60} - \frac{9}{60} - \frac{4}{ 60} - \frac{2}{60} \ = \ \frac{13}{60}$$

So the complete venn diagram looks like this:



Tree Diagrams

Let's return to Daniella and the green and blue marbles again. One way to illustrate these types of problems is by using so-called tree diagrams. All possible outcomes are represented as branches. Whenever two or more possible outcomes can occur, the diagram is further split into more branches. Along each branch we write the probability of that event occurring, and at the end of every branch we write the combined probability of that particular outcome. The problem with picking the two marbles can be described with the following tree diagram:

This diagram illustrates the four different possible outcomes that can occur and the probability for of each of them. For example, $P(\text{green-blue})$ $= \frac{3}{5} \cdot \frac{1}{3} = \frac{3}{15}$. As expected, we multiply the probabilities together as we work our way down the diagram.

Note that the sum of all four outcomes, $\frac{6}{15}$ + $\frac{3}{15}$ + $\frac{3}{15}$ + $\frac{2}{15}$, is equal to $1$. This is true for all tree diagrams. Since the diagram shows all possible outcomes that can occur the sum of those outcomes must be $1$ or $100$%. This may seem like trivial information without much practical value, but that the sum of all outcomes is always equal to $1$ can be used as a shortcut in many problems. As you may remember from the lesson on combinatorics, it’s sometimes easier to count something by counting what we don’t want and subtracting that from the total. We can use the same method when we work with probability.


Exercise 1

You pick one marble from each bag. What is the probability that at least one of the marbles is green? 

We can solve this by determining $P(\text{green-green})$, $P(\text{green-blue})$ and $P(\text{blue-green})$ and then add the probability of each outcome. But we can also determine the probability that it will not happen, $P(\text{blue-blue})$, and then subtract this from $1$. Not choosing two blue marbles is the same as choosing at least one green marble. The probability that we choose two blue marbles is $\frac{2}{15}$, so the probability that we choose at least one green marble is therefore $1 - \frac{2}{15} = \pmb{\frac{13}{15}}$.


In probability there are always two methods we can use: 1) the direct addition method where we add all the outcomes we want, or 2) the indirect subtraction method where we count everything we don’t want and subtract this from $1$. We should always determine which method will require the least amount of calculations. When problems contain the phrases:

  • "At least"
  • "At most
  • "No more than" 

it’s a clue that the subtraction method will be the fastest approach. 

Same Outcome In Different Ways

Let us examine a new problem.

Sandra is driving to work. On the way she needs to cross three traffic lights. The light is red on average 75% of the time. What is the probability that Sandra needs to wait at at most one traffic light?

We can illustrate this problem using the following tree diagram:


The tree diagram shows each possible scenario that may occur and their respective probabilities. These scenarios can be summarized in four unique outcomes:

(1) Sandra gets $\pmb{3}$ red lights

(2) Sandra gets $\pmb{2}$ red lights

(3) Sandra gets $\pmb{1}$ red light

(4) Sandra gets $\pmb{0}$ red lights

Since there are no other possible outcomes, the sum of these four cases must equal to $1$. We want to know the probability that Sandra will have to wait at either one or zero red traffic lights:  the sum of cases (3) and (4). Both the addition and subtraction method require us to determine two of the four outcomes, so we will use the addition method.


Exercise 2

What is the probability that Sandra gets no red lights? 

The probability that Sandra gets no red lights is $\dfrac{1}{4} \cdot \dfrac{1}{4} \cdot \dfrac{1}{4} =$$\pmb{\dfrac{1}{64}}$. 


Exercise 3

What is the probability that Sandra gets exactly $1$ red light?

Case (3), that Sandra gets exactly $1$ red light, is a little more tricky. If you look at the tree diagram you’ll see that there are three ways in which Sandra could get exactly $1$ red light: 

(green-green-red) $=\dfrac{1}{4} \cdot \dfrac{1}{4} \cdot \dfrac{3}{4} = \dfrac{3}{64}$

(green-red-green) $=\dfrac{1}{4} \cdot \dfrac{3}{4} \cdot \dfrac{1}{4} = \dfrac{3}{64}$

(red-green-green) $=\dfrac{3}{4} \cdot \dfrac{1}{4} \cdot \dfrac{1}{4} = \dfrac{3}{64}$

The probability that she gets exactly $1$ red light is the sum of all these three sub-cases. Notice that the likelihood of each case is the same since we are multiplying the same factors, only in a different order. This means that we can just multiply the by $3$ instead of adding all sub-cases one by one. 

The probability that Sandra gets exactly $1$ red light is therefore $3 \ \cdot $$\Big(\dfrac{1}{4} \cdot \dfrac{1}{4} \cdot \dfrac{3}{4} \Big) =$$ \pmb{\dfrac{9}{64}}$. 


You may have noticed that we only listed four outcomes when there were eight different outcomes in the tree diagram. How come?

The four cases we listed only considers the actual outcome, not the which order they might occur. That Sandra gets $1$ red light is one unique outcome, but that outcome can happen in three different ways. If we list the four unique outcomes and the number of ways in which they can occur, we get the following result:

(1) Sandra gets $\pmb{3}$ red lights  (can happen in $1$ way)

(2) Sandra gets $\pmb{2}$ red lights  (can happen in $3$ ways)

(3) Sandra gets $\pmb{1}$ red light  (can happen in $3$ ways)

(4) Sandra gets $\pmb{0}$ red lights  (can happen in $1$ way)

Notice how the eight outcomes in the tree diagram here are distributed over the four unique outcomes. 


Exercise 4

Why can Sandra get $1$ red light in exactly three different ways? Why not four or five different ways?

How many ways can we arrange $1$ red and $2$ green traffic lights? The answer is $\dfrac{3!}{2!}=\dfrac{3 \cdot 2 \cdot 1}{2 \cdot 1}=3$ ways! Three unique objects can be arranged in $3\cdot2\cdot1} ways, and since two of them are identical in this case we need to divide by $2 \cdot 1$ because it doesn't matter in which order they come in. 

When you need to determine an outcome that can occur in several different ways, you can always use your knowledge to make your calculations faster. If you are not familiar with these methods then check out the lesson on lesson on combinatorics. 


Exercise 5

Claudia rolls a die five times. What is the probability that she gets exactly three even and two odd numbers?

The probability of getting either an even or odd number is $\frac{1}{2}$. This means that $P(\text{even-even-even-odd-odd}) = $$\frac{1}{2} \cdot \frac{1}{2} \cdot \frac{1}{2} \cdot \frac{1}{2} \cdot \frac{1}{2} =$$ \Big( \frac{1}{2} \Big)^5 =$$ \frac{1}{64}$. However, this outcome can occur in $\frac{5!}{3! \ \cdot \ 2!} = 10$ different ways. The answer is therefore $\frac{1}{64} \cdot 10 = \pmb{\frac{5}{32}}$.


Exercise 6

A coin is flipped three times. Patrick claims that it is three times as likely to get two heads and one tail compared to three heads. Tobias flips the coin twice and gets heads on both flips. He gets excited and exclaims: "Now I have three times the likelihood of getting tails on my next throw compared to heads!". 

Are Patrick's and Tobias' statements true?

The probability of getting two heads and one tails is $\big(\frac{1}{2} \Big)^3 \cdot \frac{3!}{2!} =$$ \frac{3}{8}$ and the probability of getting three heads is $\big(\frac{1}{2} \Big)^3 = $$\frac{1}{8}$. Patrick's statement is true.

Now let’s turn to Tobias' statement. Coin flips are independent events, so the probability of getting tails on the third flip is still $\frac{1}{2}$. Before we began flipping the coin, there were three ways in which we could get two heads and one tails: 

(heads-heads-tails)

(heads-tails-heads)

(tails-heads-heads). 

However, when we have flipped the coin twice and gotten two heads, there is only one way to get the desired outcome: (heads-heads-tails). This is the difference between Patrick's and Tobias' reasoning. 

If you flip a coin and get heads again and again, it’s common to think "I’ve already gotten so many heads. The next flip must be tails!". But in fact, it is just as likely to get heads as tails on the next flip no matter how many heads you already have. When events are independent, past outcomes have no effect on future outcomes. Tobias' statement is false


A Final Word On Tree Diagrams

One last tip. I would highly recommend that you don’t use tree diagrams when solving problems. This may seem contradictory given how much time we have devoted to them in this lesson. But we only used them here to better illustrate the different aspects of probability. Drawing tree diagrams is messy in practise and requires a lot more work than they are worth. You should know the principles behind tree diagrams (how they work and why) but you should never have to draw them when solving problems.

What you need to know about tree diagrams:

1) We multiply the probabilities as we work our way down the diagram.

2) The sum of all possible outcomes is $1$. When you need to determine the sum of several outcomes, you can use both the addition or subtraction method, whichever requires the least calculations.

3) Some outcomes can occur in more than one way. Use your knowledge in combinatorics to speed up your calculations. For example, if a unique outcome can occur in $12$ different ways, and the probability that any of them occurs is $\frac{1}{50}$, the probability that the unique outcome will happen is $\frac{1}{50} \cdot 12 =$$ \pmb{\frac{6}{25}}$.

Easy but Not Intuitive

Many probability problems require very little work to solve. Despite this, we can easily end up on the wrong path because the right answer doesn’t seem correct. In these cases our intuition becomes a problem rather than an asset, so it’s extra important that we trust our calculations.


Example 1

We assume that giving birth to a boy is as likely as a girl.

a) Inge van Gogh has two children. The oldest of them is called Albert. What is the probability that both are boys?

b) Vincent De Bruijn also has two children. One of them is called Erik. What is the probability that both are boys?

Solution

There are four possible ways that a family could have two children: (girl-girl), (girl-boy), (boy-girl) or (boy-boy). It is just as likely that any of these outcomes will occur.

a) When the older child is a boy there are two possible outcomes: (boy-boy) or (boy-girl). The probability that both children are boys is $\pmb{\frac{1}{2}}$.

b) When one of the children is a boy, but we don’t know if it’s the older child who is a boy or not, there are three possible outcomes: (boy-girl), (girl-boy) or (boy-boy). The probability that both children are boys is therefore $\pmb{\frac{1}{3}}$.


At first, question b) seems to be exactly the same problem as a). But they are actually different. The best way to solve this kind of problem is not to go on our initial gut feeling, but to carefully think through what possible outcomes there are given the conditions. A tip is to use the following equation:

$$P(X) \ = \ \frac{\text{#Ways that X could happen}}{\text{#Total outcomes given the conditions}}$$

Using this formula, we notice there are $2$ possible outcomes in a) given that the older child is a boy. Both children are boys in only $1$ of these outcomes so the answer to a) is $\frac{1}{2}$. 

In b), however, there are $3$ possible outcomes given that one of the children is a boy. Both children are boys in only $1$ of these outcomes so the answer to b) is $\frac{1}{3}$.


Exercise 1

You are given three bags. One of the bags contains two black marbles, one contains two white marbles and one contains one black and one white marble. You randomly take one marble out of one of the bags and see that it’s white. What is the probability that the other marble in the bag you chose is also white?

There are three possible outcomes where the first marble is white: 

(white marble #$1$ from the bag with two white marbles) 

(white marble #$2$ from the bag with two white marbles)

(white marble from the bag with one white and one black marble)

It is equally likely that any of these three cases will occur. If either of the first two cases happens, the second marble will also be white. The probability that we choose two white balls is therefore $\pmb{\frac{2}{3}}$.


Let us now return to the question we posed at the introduction of this lesson: 

You have been chosen to participate in a lottery show on TV. The host shows you three doors. Behind one of them is a brand new Ferrari, but behind the other two doors is a goat. You get to choose one of the doors and keep the object behind it. 

You decide to choose the door to the left. The host then opens the door on the far right to show that there was a goat behind it. You are now given two options: stick with your first choice or switch to the door in the middle. 


Exercise 2

What should you do to maximize your chances of getting the Ferrari? 

We make two decisions in this problem. First, we decide which of the three doors we want to choose. Second, we decide whether we want to keep our initial choice or not. There are therefore six possible outcomes in this problem:

(Goat #$ 1 $ - keep)    $\Longrightarrow $   Goat

(Goat #$ 1 $ - switch)    $\Longrightarrow $   Ferrari

(Goat #$ 2 $ - keep)    $\Longrightarrow $   Goat

(Goat #$ 2 $ - switch)    $\Longrightarrow $   Ferrari

(Ferrari - keep)    $\Longrightarrow $   Ferrari

(Ferrari - switch)     $\Longrightarrow $   Goat

If we decide to keep our initial choice we will get a goat in two of the three possible cases $\Big( \dfrac{1}{3} \Big)$. But if we switch doors we get a Ferrari in two out of three cases $\Big( \dfrac{2}{3} \Big)$. The best decision is therefore to switch doors.

Another way to think about this is that we have a $\frac{2}{3}$ chance of picking a goat in the first step. So if we decide to keep our choice the probability is $\frac{2}{3}$ that we get a goat and $\frac{1}{3}$ that we get the Ferrari. Look at the different outcomes outlined above so you really understand this. If we switch doors, however, we turn these probabilities around so that the likelihood of getting the Ferrari is $\frac{2}{3}$ while the probability of getting a goat is $\frac{1}{3}$.


Problems to Solve

You now have all the tools you need to solve the majority of probability problems on your own. Try putting what you've learnt into practise by solving these questions. 


Exercise 1

The wheels below rotate around their centers. A nail at the top outside each wheel indicates the value we get when the wheels stop. What is the probability that the sum of the two numbers is even?

For the sum of the two numbers to be even, both must be either odd or even. We have:

$P(\text{both even} = \frac{1}{4} \cdot \frac{2}{3} =$$ \frac{1}{6}$ 

$P(\text{both odd} =$\frac{3}{4} \cdot \frac{1}{4} =$$ \frac{1}{4}$

So the probability that the sum is even is $\frac{1}{4} \cdot \frac{1}{6} = $$\pmb{\frac{5}{12}}$.


Exercise 2

Four cards, two red and two black, are placed face down on a table. We select two of these cards. What is the probability of getting a red card?

We have three possible outcomes in this problem:

(1) $\pmb{0}$ red cards

(2) $\pmb{1}$ red card

(3) $\pmb{2}$ red cards 

The probability of getting a red card is the sum of (2) and (3). In this situation it’s simplest to use the subtraction method, so we will determine the probability of not getting a red card. 

The odds of getting $0$ red cards are $\frac{2}{4} \cdot \frac{1}{3} = \frac{1}{6}$. The probability of getting a red card is therefore $ 1 - \frac{1}{6} = \pmb{\frac{5}{6}}$.


Exercise 3

In a forest glade there are four bird nests. Each nest has a bird mother who spends $12$ minutes every hour at the nest feeding her younglings. The bird mother spends the rest of the time out in the woods looking for food. What is the probability that there is at least one bird mother in the forest glade at any given moment?

At any given moment there are either:

(1) $\pmb{0}$ bird mothers

(2) $\pmb{1}$ bird mother

(3) $\pmb{2}$ bird mothers

(4) $\pmb{3}$ bird mothers

(5) $\pmb{4}$ bird mothers 

in the forest clearing.

The probability that there is at least one bird mother the forest clearing is the sum of cases (2), (3), (4) and (5). In this situation it’s obviously easier to use the subtraction method. We will therefore determine case (1) and subtract this from $1$. 

A bird mother spends $48$ minutes every hour away from the nest. So the likelihood of a specific bird mother not being in the forest clearing is $\frac{48}{60} = \frac{4}{5}$. The probability that none of the four bird mothers are present is $ \Big(\frac{4}{5} \Big)^5 = \frac{256}{625}$. 

The probability that at least one bird mother is in the clearing at a given moment is therefore $ 1 - \frac{256}{625} = \pmb{\frac{369}{625}}$.


Exercise 4

Five ants are placed on the corners of a regular pentagon, one ant on each corner. The ants can only walk clockwise or counter clockwise along the sides of the pentagon. If the ants all start walking in a random direction at the same time and at the same speed, what is the probability that the ants never collide?

he first ant can choose to go either clockwise or counter clockwise. The other four ants must all choose to go in the same direction as the first ant to avoid colliding. Each of these four ants has two directions to choose from so the probability that they never collide is $ \Big(\frac{1}{2} \Big)^4 = \pmb{\frac{1}{16}}$.


Exercise 5

You toss a coin three times. What is the probability of getting at least two tails in a row?

If we toss a coin three times we can get $ 2^3 = 8 $ possible outcomes. In three of these outcomes (heads-tails-tails, tails-tails-heads, tails-tails-tails) we get at least two tails in a row. The probability of getting at least two tails in a row is thus $ \pmb{\frac{3}{8}}$.


Exercise 6

In an urn there are $9$ balls numbered $1$ to $9$. You pick out three balls at random and place them in numbered order with the greatest number to the right. What is the probability that the difference between the first and the second ball is equal to the difference between the second and the third ball?

The difference between two consecutive balls can be either $ 1 $, $ 2 $, $ 3 $ or $ 4 $. That the difference is $ 1 $ can happen in seven ways, that the difference is $ 2 $ can happen in five ways, that the difference is $ 3 $ can happen in three ways and that the difference is $ 4 $ can happen in one way. In total, there are $ 7 + 5 + 3 + 1 = 16 $ ways that the differences can be equal. One way to illustrate why we end up with $7,5,3$ and $1$ ways for the differences to be equal is to draw nine empty spaces next to each other. How many ways you could place three objects in these empty spaces if the objects on the right and left both must be either one, two, three or four spaces away from the center object? 

We can choose three balls from an urn with nine balls in $\displaystyle{\binom{9}{3}}=$$\frac{9 \ \cdot \ 8 \ \cdot \ 7}{3!} = 84 $ different ways. The probability that the difference between consecutive balls is equal is thus $\frac{16}{84} = \pmb{\frac{4}{21}}$.


Exercise 7

Rachel plants five seeds in her flower bed. The probability of a seed taking root and growing into a flower is $\frac{3}{4}$. How likely is Rachel to get:

a) exactly one flower?

b) exactly two flowers?

c) at most three flowers?

a) That exactly one flower sprouts can happen in five different ways: it’s either flower #$1$, #$2$, #$3$, #$4$ or #$5$. The probability that any particular one of these five outcomes will occur is $\frac{3}{4} \cdot \frac{1}{4} \cdot \frac{1}{4} \cdot \frac{1}{4} \cdot \frac{1}{4} =$$ \frac{3}{1024}$. The probability that exactly one flower sprouts is therefore $ \pmb{\frac{15}{1024}}$.


b) That exactly two seeds sprout can happen in $\frac{5 \ \cdot \ 4}{2!} = 10 $ different ways. The probability that any particular of these ten outcomes will occur is $\frac{3}{4} \cdot \frac{3}{4} \cdot \frac{1}{4} \cdot \frac{1}{4} \cdot \frac{1}{4} =$$ \frac{9}{1024}$, so we have to multiply this by $ 10 $. The probability that exactly two of the five flowers sprout is thus $\frac{90}{1024} = \pmb{\frac{45}{512}}$.


c) We can divide this problem into the following outcomes:

(1) $\pmb{0}$ flowers sprouts

(2) $\pmb{1}$ flower sprout

(3) $\pmb{2}$ flowers sprouts

(4) $\pmb{3}$ flowers sprouts

(5) $\pmb{4}$ flowers sprouts

(6) $\pmb{5}$ flowers sprouts

The probability that at most three flowers sprout is the sum of cases (1), (2), (3) and (4). In this situation, it’s simpler to use the subtraction method and instead determine the sum of cases (5) and (6) and subtract the result from $1$. 

That exactly four flowers that sprout can happen in $\frac{5 \cdot 4 \cdot 3 \cdot 2}{4!} = 5$ different ways. The probability that any particular of these five outcomes occurs is $\frac{3}{4} \cdot \frac{3}{4} \cdot \frac{3}{4} \cdot \frac{3}{4} \cdot \frac{1}{4} = $$ \frac{81}{1024}$. The probability that exactly four flowers sprout is therefore $\frac{81}{1024} \cdot 5 = \frac{405}{1024}$.

That all flowers sprout can only happen in only one way. The probability of this outcome is therefore $ \Big(\frac{3}{4} \Big)^5 = \frac{243}{1024}$.

The sum of these two outcomes is $\frac{405}{1024} + \frac{243}{1024} = \frac{648}{1024}$. The probability that we get at most three flowers is thus $ 1 - \frac{405}{1024} - $$\frac{243}{1024} - =$$ \frac{376}{1024} =$$ \pmb{\frac{47}{128}}$.


Exercise 8

It is known that $3$ out of $250$ births are twin births and that the population of Sweden is approximately $ 9.6 $ million. How many twins are there in Sweden? Determine your answer with the accuracy of two value digits. 

$ 247 $ normal births and $ 3 $ twin births means $247 \cdot 1 + 3 \cdot 2 =$$ 253 $ children. $ 6 $ of these children are twins. This means that there are $\frac{6}{253} \cdot 9.6 \cdot 10^6 \ approx \pmb{230 \ 000}$ twins in Sweden.


Exercise 9

You randomly select two positive integers.

a) What is the probability that the sum of the numbers is divisible by $3$?

b) What is the probability that the product of the numbers is divisible by $3$?

a) When a number is divided by $ 3 $ we get one of three possible remainders: $ 0 $, $ 1 $ or $ 2 $. In order for the sum of the numbers to be divisible by $ 3 $, the sum of the remainders of those numbers when divided by $3$ must be divisible by $ 3 $. Once we have selected the first number, there is only one remainder which the second number may have for the sum to divisible by $ 3 $. So the answer is $ \pmb{\frac{1}{3}}$.


b) Since every third number is divisible by $ 3 $, the probability that we choose such a number is $\frac{1}{3}$. For the product of the numbers to be divisible by $ 3 $, at least one of the numbers must be divisible by $ 3 $. We can divide this problem into three possible outcomes:

(1) None of the numbers are divisible by $\pmb{3}$

(2) Exactly one of the numbers are divisible by $\pmb{3}$ 

(3) Both of the numbers are divisible by $\pmb{3}$

We want to find the sum of cases (2) and (3). The simplest approach here is to use the subtraction method and instead determine (1): the probability that none of the numbers are divisible by $3$. The answer is therefore:

$$ P(\text{at least one number divisible by} \ 3) \ = $$

$$ 1 - P(\text{no number divisible by} \ 3) \ = $$

$$ 1 - \frac{2}{3} \cdot \frac{2}{3} \ =$$ 

$$\pmb{\frac{5}{9}}$$


Exercise 10

You are lost in the rainforest and have happened to eat a poisonous mushroom. To save your life, you need to lick the antidote secreted by a specific frog species. Unfortunately, only the females produce the antidote. In the rainforest there are as many female and male frogs. Female and male frogs look identical, but the male has a special croak that distinguishes them. You have seconds left to live when you suddenly see such a frog to your right on a stump. To your left you suddenly hear the characteristic male croak, you turn shortly after the sound stops and see two identical frogs close together in a clearing.

In which direction should you run to have the highest chance of licking a female and surviving? (You may lick two frogs at the same time).

The frog sitting on the stump can be either a female or male frog. Since there are as many frogs of each sex, the probability is that that frog is a female is $\frac{1}{2}$. 

There are three possible cases for the frogs sitting in the clearing: (female-male), (male-female) or (male-male). Because each fall is equally likely to occur, the probability that one of the frogs is a female is $\frac{2}{3}$. So you should run left to maximize your chance of survival.


Exercise 11

The probability of a baseball player hitting the ball is $\frac{3}{10}$. What is the probability that she will get exactly two hits on four attempts?

The baseball player can get two hits and two misses in $\frac{4!}{2! \ \cdot \ 2!} = 6 $ different ways and the probability of each of these outcomes is $\frac{3}{10} \cdot \frac{3}{10} \cdot \frac{7}{10 } \cdot \frac{7}{10} = $$\frac{441}{10000}$. The probability that she gets exactly two hits and two misses is therefore $ 6 \cdot \frac{441}{10000} = $$\pmb{\frac{1323}{5000}}$.


Exercise 12

Your friend makes the following statement: "You are more likely to get at most one six on a three-dice roll than you are to get at most two sixes on a four-dice roll." Is your friend's statement true?

If we roll three dice we can get the following four outcomes:

(1) $\pmb{0}$ sixes

(2) $\pmb{1}$ six

(3) $\pmb{2}$ sixes

(4) $\pmb{3}$ sixes

The probability of getting at most one six is equal to the probabilities of cases (1) and (2). The probability of (1) is $ \Big(\frac{5}{6} \Big)^3 = \frac{125}{216}$ and the probability of (2) is $\frac{1}{6} \cdot \Big(\frac{5}{6} \Big)^2 \cdot \frac{3!}{2!} = \frac{75}{216}$. (We multiply with $\frac{3!}{2!}$ becasue that is how many ways that one six and two non-sixes can be arranged). We know have:

$$ P(\text{at most one six on three throws}) = $$

$$\frac{125}{216} + \frac{75}{216} = \frac{200}{216}$$ 

If we roll four dice, we can get the following five outcomes:

(1) $\pmb{0}$ sixes

(2) $\pmb{1}$ six

(3) $\pmb{2}$ sixes

(4) $\pmb{3}$ sixes

(5) $\pmb{4}$ sixes

The probability of getting at most two sixes is equal to $1 - P(3 \ \text{sixes}) - $$P(4 \ \text{sixes})$. 

$P(3 \ \text{sixes}) \ =$$ \Big(\frac{1}{6} \Big)^3 \cdot \frac{5}{6} \cdot \frac{4!}{3!} =$$ \frac { 20}{1296}$ 

$P(4 \ \text{sixes} \ =$$ \Big(\frac{1}{6} \Big)^4 = \frac{1}{1296}$. 

We then have:

$$ P(\text{at most two sixes on three throws}) = $$

$$1 - \Big(\frac{21}{1296} + \frac{1}{1296} \Big) = $$

$$\frac{1274}{1296}$$

In comparison, $ P(\text{at most one six on three throws}) = $$\frac{200}{216} = $$\frac{1200}{1296}$. It is therefore more likely to get at most two sixes on a four-dice roll than to get at most one six on a three-dice roll. Your friend's statement is false.


Exercise 13

David owns an unevenly loaded die. If it is thrown twice, the probability of getting at least an even number is $\frac{1}{3}$. What is the probability of getting an odd number if you roll the dice once?

Since the probability of getting at least one even number on two dice rolls is $\frac{1}{3}$, the probability of getting two odd numbers must be $\frac{2}{3}$. Die rolls are independent events, so this means means that:

$$ P(\text{odd}) \cdot P(\text{odd}) = \frac{2}{3}$$

$$ P(\text{odd}) = \pmb{\sqrt{\frac{2}{3}}}$$


Exercise 14

What is the probability of getting exactly two aces if you choose five randomly selected cards from a deck of cards? (A standard deck of cards has $52$ cards, of which four are aces). 

We can solve this problem by determining number of five-card shuffles that contain exactly two aces and dividing it by the total number of possible five-card shuffles.

There are $\displaystyle{\binom{4}{2}}=$$\frac{4 \ \cdot \ 3}{2!} = 6 $ ways to choose two of the four aces. There are $48$ other non-ace cards and we can choose three of these in $\displaystyle{\binom{48}{3}}=$$\frac{48 \ \cdot \ 47 \ \cdot \ 46}{3!} = $$(16 \cdot 47 \cdot 23) $ ways . The number of five-card shuffles with exactly two aces is thus $ 6 \cdot (16 \cdot 47 \cdot 23) $. We will wait with multiplying this out because we will also divide the result by the total number of five-card shuffles. We will likely be able to eliminate several common factors when we perform that division. It’s therefore better to wait with multiplying the numbers until we have been able to simplify the fraction as far as possible. 

The total number of possible five-card shuffles is:

$$displaystyle{\binom{52}{5}} = \dfrac{52 \cdot 51 \cdot 50 \cdot 49 \cdot 48}{5!} = $$

$$\frac{52}{2} \cdot \frac{51}{ 3} \cdot \frac{50}{5} \cdot \frac{49}{1} \cdot \frac{48}{4} = $$

$$ 26 \cdot 17 \cdot 10 \cdot 49 \cdot 12 $$

Notice how we also simplified this fraction as far as possible before multiplying out the numerator and the denominator.

The probability that a five-card deck contains exactly two aces is therefore:

$$\frac{6 \cdot 16 \cdot 47 \cdot 23}{26 \cdot 17 \cdot 10 \cdot 49 \cdot 12} = $$

$$\frac{6 \cdot 47 \cdot 23}{13 \cdot 17 \cdot 5 \cdot 49 \cdot 3} = $$

$$\frac{2 \cdot 47 \cdot 23}{13 \cdot 17 \cdot 49} = $$

$$ \pmb{\frac{2162}{10829}}$$

Here you can see the benefit of keeping the numerator and denominator as a product of separate numbers rather than multiplying them the first thing we do. Had we multiplied the fraction to $\frac{103 \ 776}{2 \ 598 \ 960}$ instead of $\frac{6 \ \cdot \ 16 \cdot \ 47 \ \cdot \ 23}{26 \ \cdot \ 17 \ \cdot \ 10 \ \cdot \ 49 \ \cdot \ 12}$ it would have been much harder to see that both the numerator and the denominator were divisible by $48$. It’s almost always easiest to keep your results as a product until the end .

In this problem, it's ok to leave $\frac{6 \ \cdot \ 16 \cdot \ 47 \ \cdot \ 23}{26 \ \cdot \ 17 \ \cdot \ 10 \ \cdot \ 49 \ \cdot \ 12}$ as your final answer. The most important thing is that you show how you arrived at the answer and that you understand how to solve the problem. Not that you can multiply with large numbers. If the numbers become large, it’s perfectly ok to leave the answer as an incomplete calculation.


Exercise 15

Linda is $31$ years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations. Which is more probable?

a) Linda works at a bank.

b) Linda is a bank teller and is active in the feminist movement.

The probability that Linda both works at a bank and is active in the feminist movement can be written as $ P(\text{bank}) \cdot P(\text{feminist}) $. This is obviously less compared to the probability that she works at a bank, $ P(\text{bank})$. Option a) is thus most likely. 

We can illustrate this situation with a venn diagram. Let circle $ B $ be the probability that Linda works at a bank and let circle $ F $ be the probability that Linda is active in the feminist movement. Option a), the probability that Linda works at a bank, is then the whole circle $ B $.


Option b), the probability that Linda both works at a bank and is active in the feminist movement, is only the intersection of the circles $ B $ and $ F $.


Since b) is a subset of a), it is obviously more likely that a) is true than that b) is true. If b) is true, then by definition a) must be true as well. But a) may be true without b) being true.


Exercise 16

A number $p$ is randomly selected from the set $ \{11, 13, 15, 17, 19 \}$ and a number $q$ is randomly selected from the set $ \{1999, 2000, 2001, 2002,…, 2018 \}$. What is the probability that the product $p^q$ ends with a $1$?

Let us list the last digits for $ 11^x, 13^x, 15^x 17^x $ and $ 19x $ where $ x = 1, 2, 3, 4… $:

$ \text{Last digit} \ 11^x: \ \ \ \ \ 1, 1, 1, 1, 1… $

$ \text{Last digit} \ 13^x: \ \ \ \ \ 3, 9, 7, 1, 3… $

$ \text{Last digit} \ 15^x: \ \ \ \ \ 5, 5, 5, 5, 5… $

$ \text{Last digit} \ 17^x: \ \ \ \ \ 7, 9, 3, 1, 7… $

$ \text{Last digit} \ 19^x: \ \ \ \ \ 9, 1, 9, 1, 9… $

$ 11^x $ always ends in $ 1 $. $ 15^x $ never has $1$ as its last digit. $ 13^x $ and $ 17^x $ end in the digit $ 1 $ if and only if $ x $ is divisible by $ 4 $, and $ 19x $ ends with a $ 1 $ if and only if $ x $ is divisible by $ 2 $. Of the total $ 20 $ numbers that $ q $ is selected from, five of them are divisible by $ 4 $ and ten are divisible by $ 2 $. The probability that $ p^q $ ends with a $ 1 $ is thus:

$ \Big(\dfrac{1}{5} \cdot 1 \Big) + \Big(\dfrac{1}{5} \cdot \dfrac{5}{20} \Big) + $$\Big(\dfrac{1}{5} \cdot 0 \Big) + \Big(\dfrac{1}{5} \cdot \dfrac{10}{20} \Big) \ =$$ \ \pmb{\dfrac{2}{5}}$


Exercise 17

A secretary writes letters to eight different people and addresses eight envelopes with these people's addresses. He then places the letters randomly in the envelopes. What is the probability that exactly six of the letters end up in the correct envelope?

We will solve this problem by calculating the number of ways that exactly six of the letters can reach the right person and then dividing it by the total number. possible ways in which the letters can be sent to the various persons.

If exactly six of the letters reach the right person, it means that exactly two of the letters do not. We can select these two people in $\frac{8 \ \cdot \ 7}{2!} = 28 $ ways. The total number of ways the letters can be sent is $ 8! $. The answer to the problem is therefore:

$$\frac{28}{8!} = $$

$$\frac{7 \ \cdot \ 4}{8 \ \cdot \ 7 \ \cdot \ 6 \ \cdot \ 5 \ \cdot \ 4 \ \cdot \ 3 \ \cdot \ 2 \ \cdot \ 1} = $$

$$\frac{1}{8 \ \cdot \ 6 \ \cdot \ 5 \ \cdot \ 3 \ \cdot \ 2 \ \cdot \ 1} = $$

$$ \pmb{\frac{1}{1440}}$$


Exercise 18

Richard is out hitchhiking along a road. The probability that he will see a car sometime in the next $ 20 $ minutes is $\frac{609}{625}$. What is the probability that he will see a car sometime in the next five minutes? Assume that the probability of seeing a car is the same for the entire $ 20 $ minutes.

Take a closer look at the fraction $\frac{609}{625}$. Why do we have these two specific numbers? $ 625 $ is $ 5 \cdot 5 \cdot 5 \cdot 5 = 5^4$. The difference $625 - 609$ also happens to be $16$, which is the same as $2 \cdot 2 \cdot 2 \cdot 2 = 2^4$. Try to use this information to solve the problem on your own before you continue to read the rest of the solution.

Let the probability of seeing a car in the next $ 5 $ minutes be $ p $. The odds of not seeing a car in the next $ 5 $ minutes are then $ (1-p). $ This means that the probability of not seeing a car in the next $ 20 $ minutes is $ (1-p)^4 $. We also know that the probability that Richard will not see a car in the next $ 20 $ minutes is $\frac{16}{625}$. This gives us:

$$ (1 - p)^4 = \frac{16}{625}$$

$$ (1 - p) = \frac{2}{5}$$

$$\frac{3}{5 } = p $$ 

The probability that Richard will see a car in the next $ 5 $ minutes is thus $ \pmb{\frac{3}{5}}$.