Probability Year 10

Terminology

Probability measures the chance something happens. Formally, we say it measures how likely is the
outcome of an event. We write P(result) as a shorthand.

An event is some measurable test, measurement or observation. Sometimes trial is used instead.

The results of an event are often called its outcomes. The ones we are interested in are usually
termed successes, regardless of their actual type (so having blue eyes can be a “success”).

Mutually exclusive results are the exact opposites of each other: either one or the other must
happen. Mathematically, outcomes are mutually exclusive when their probabilities add to 1.

The use of “not” helps with mutually exclusive events: the opposite of “rainy” is best thought of as
“not rainy” rather than vague terms such as “fine”.

Another term seen sometimes is “complement”. The complement is marked with a dash, ' , and so
we can write that P(A) + P(A') = 1

Numerical value

Probability is a number between 0 and 1.

P(X) = 0 means that the result X is impossible
P(X) = 1 means that the result X is certain.

Many students prefer to think of probability as being from 0% to 100%, which is just the same thing
expressed differently.

Students need be prepared to leave descriptions such as “likely” behind. They are sometimes useful
to describe what a probability value means, but they are not suitable when asked for a probability.
In general “likely” means a probability of greater than 0.5, becoming very likely around 0.8.

No probability can be bigger than one. If students get an answer larger than 1 then they should
immediately recognise that as impossible.

Usually answers may be given in any appropriate form: decimal, fraction or percentage.
Percentages are the least useful because they have to be converted before any calculations are done.
In general fractions are the most useful.
 Calculating Simple Probability

Probabilities can be calculated experimentally. A number of trials are conducted (or events are
measured) and

Probability =

Experimental probability is only an estimate, and will change slightly each time it is measured.

Theoretical probability can be calculated similarly, so long as each result (outcome) is equally
likely:

Probability =

When calculating theoretical probability it is important to divide up the outcomes in such a way as
to make them equally likely.
3
needs to be thought of as so that the white is 6
of the area.

The probability of white is 3

6
.

The calculation of probability is expressed as a fraction because that is, in most cases, the best way
to think about the concept.
A probability of 0.2 means very little, but makes it much more obvious that it is one
chance in five.

When calculating experimental probability we ideally test the entire population. Sometimes that is
not possible because it is too difficult, too expensive etc or the population is infinite. In those cases
we take a sample and work from that. The sample needs to be:
– random
– independent (the outcome of one trial does not affect the result of the others)
– sufficiently large (how large depends on how accurate you need it)
Students may notice that these conditions are the same as those we need for proper sampling for all
statistics.
 Expected Value

The expected value is the number of successes that are expected for a certain number of trials.
Expected value = Probability × number of trials

The expected value is basically the same as the mean (average) number of successes, except that a
mean can be a decimal value, whereas an expected value is rounded because it is not possible to
have a fractional success.
The probability of 6 on a normal dice is = 16. If we roll the dice 20 times we calculate
1
6
× 20 = 3.33 sixes on average,
1
but the expected number of sixes is 3, since you can’t get 3
of a roll.

Calculations with Probabilities

If two results are possible from a single trial, then the probability of either is the probability of each
taken alone added together.
e.g. if a test has three possible outcomes: A, B or C, then for any one trial we can say that
P(A or B) = P(A) + P(B)

If two results occur in separate trials, then the probability of one then the other is the probability of
each taken alone multiplied together.
e.g. if a test has three possible outcomes: A, B or C, then we can say that
P(A both times) = P(A) × P(A)

It is important when calculating probabilities to consider different arrangements which give the
same result as effectively different.
e.g. if we toss two coins, then there are three possible outcomes (two heads, two tails,
one of each) but we have to consider HT and TH as different. It can help to consider the
coins as different colours, so that red H and blue T is obviously different to red T with
blue H.
The probability of one head and one tail from a toss of two coins is 24 , not 13.

Probability calculations are made much easier if real life differences which do not matter to the
question are ignored.
e.g. when calculating the probability of twice drawing a card from a standard deck
without getting a heart it is not helpful to count clubs, diamonds or spades as separate
options – for this purpose what matters is that the card is either a heart or not a heart.
P(no hearts) = ¾ × ¾ = 0.75 × 0.75 = 0.5625
 Probability Trees
Trees are generally the best way to resolve complicated problems with multiple paths to success.

● Start from what is actually happening. You need to be very clear on this.
● What are the potential results of the event you consider first? Draw branches to those
outcomes.
● Carry on for each following event from the ends of the previous branch. What potential
results can follow? Draw branches to those outcomes.
● When finished with the tree, write on each branch its probability.
● Decide which paths down the branches give the results you are looking for.
● The probability of each result is the probabilities on that branching multiplied.
● The final probability is the sum of all the individual paths of the result of interest.

Note: an event is some measurable test or observation. The results of an event are its outcomes.

Each vertical row of a tree represents a separate event: something that happens with a measurable
outcome (result). The branches go to those different outcomes, each with its probability attached.
Outcomes of Outcomes of
first event second event Probability

0.4 Success P(S, S) = 0.4 × 0.4 = 0.16

Success
0.4 0.6 Failure P(S, F) = 0.4 × 0.6 = 0.24

0.6
0.4 Success P(F, S) = 0.6 × 0.4 = 0.24
Failure
0.6 Failure P(F, F) = 0.6 × 0.6 = 0.36

Thus the probability of getting one, and only one, success is = 0.24 + 0.24 = 0.48

The key is getting the structure right first, before worrying about the rest. Each event adds another
family of branching, while more outcomes at each event means more lines at each branching.

Three events, each with two outcomes Two events, each with three outcomes
Generally when drawing trees it is wise to work through events in the order they occur, as it is less
confusing. If events occur simultaneously then it does not matter which order they are drawn, but
separate events must always have separate branching, even when they occur at the same time.

e.g. if two coins are tossed at the same time, the tree is

½ Heads P(2H) = 0.5 × 0.5 = 0.25

Heads
½ ½ Tails P(H, T) = 0.5 × 0.5 = 0.25

½ Heads P(T, H) = 0.5 × 0.5 = 0.25

½
Tails
½ Tails P(2T) = 0.5 × 0.5 = 0.25

Note that although the tosses are at the same time, it is important to distinguish the
two different ways a head and a tail can be reached. It can help to imagine the coins
are different colours, so that red H and blue T is obviously different to red T with
blue H.

It is not necessary to draw every branch in many cases. Once further branching has no meaning,
either because success or failure is automatic, it can be stopped.
e.g. if we want to draw a tree to see if three trials occur with at least one success, then
once any success is reached there is no need to continue the tree further. The tree
becomes:

Success
0.4

0.4 Success
0.6
Failure 0.4 Success
0.6 Failure
0.6 Failure

P(at least one success) = (0.4) + (0.6 × 0.4) + (0.6 × 0.6 × 0.4) = 0.784

Often it is quicker to work out answers by looking at the complementary (opposite) result.
In the case above, P(at least one success in three trials) = 1 – P(all three are failures)
= 1 – (0.6 × 0.6 × 0.6) = 0.784
 Sampling Without Replacement
Most questions at Year 10 do not need this, but it becomes increasingly important later.

If a selection leaves the pool of options changed, then each following selection is at slightly
different odds.

The key is to calculate the probability of each step separately.

Probability = as before, but now the number of results changes with each event.

After each step the number of possible results drops by one. The successful result also drops by one
if a success occurred previously, because that option is no longer possible.
There are 10 cards, of which five are red. Two are drawn at the same time.
P(1st is red) = the Probability the first card drawn is red = 5 out of 10.
P(2nd is red) = 4 out of 9, since one of the five red cards is gone, out of the 9 cards left
P(2 reds) = P(1st is red) × P(2nd is red) = × = 0.2

Things are slightly more difficult if the probability required mixes some successes and some
failures. Each step must be thought through on its own.

There are 10 cards, of which five are red. Two are drawn at the same time.
Seeking the probability that they are different colours.
4
/9 red
5
/10 red
5 5 5
/9 black 9 /10 × /9
5 5 5
/9 red 9 /10 × /9
5
/10 black
4
/9 black

5 5 5 5 5
P(different colours) = /10 × /9 + /10 × /9 = /18 = 27.8 %