Understanding Trigonometric Functions as

Functions

April 1, 2025

A function maps every permissible input (every value in the domain) to a

single output. This is why the vertical line test works–if there was an input
with two outputs (which simply would not make sense anyway), a vertical line
could be drawn through more than one point when the inputs are graphed on
the horizontal axis and the outputs are graphed on the vertical axis.

Sine and cosine (and tangent) follow patterns. Sure every function is arguably
a pattern of sorts, but when we extend the definition of sine and cosine (and
tangent), we obtain very useful, periodic functions. This is not the first time
we have extended a definition. Think about exponents:

34 = 3 · 3 · 3 · 3

3n = 3| · 3 · 3{z
. . . 3 · 3}
n times

We develop rules around multiplying and dividing powers, which tell us that
when we divide powers with the same base, we subtract the exponents.

33
= 33−3 = 30
33
3
We also know that 333 = 1. So we define the meaning of an exponent of zero.
Anything to the power of zero is 1. This is an extension of the meaning of
exponents. We can make sense of this extension by saying we are multiplying 1
by our base as many times as the exponent tells us.

35 = 1 · 3 · 3 · 3 · 3 · 3

So for an exponent of zero, we just have 1.

From exponent rules, when we calculate powers of powers, we multiply. This
means that if we allowed ourselves to have fractional exponents, we could write:
1
(5 2 )2 = 51 = 5

1
and
1
(5 3 )3 = 51 = 5
Remember, we know that √
( 5)2 = 5
and √
( 5)3 = 5
3

1 √ 1
If we compare these equations, we see that 5 2 should be 5, and 5 3 should be
1
5 3 . We have just extended powers to include fractional exponents.
We keep going, asking ourselves, if

3x = 10

what is x? We cannot find any fraction that will work. We do two things.
If we have an irrational number exponent, i.e. it cannot even be expressed as
a fraction, the value of the power will√practically be the same as the decimal
approximation of the exponent: since 3 ≈ 1.732
√ 17
3
2 ≈ 21.7 = 2 10
173
and we can do fractional exponents, though 21.73 = 2 100 is a better approxima-
1732
tion, and 21.732 = 2 1000 is an even better approximation. So we have extended
powers to include irrational exponents, but we still have not solved for x. For
that, we had to develop . . . logarithms! That is quite the extension. I will
say one more thing: sometimes extensions require exclusions. For example, we
cannot write fractional exponents of negative numbers if the fraction reduced
to lowest terms has an even number denominator. This is because we cannot
do the square root (or 4th root or 6th root. . . etc.) of a negative number. For
1 √
example, (−1) 2 = −1 which is “imaginary”. We also cannot use irrational
numbers as exponents for negative numbers, because the fraction approximation
may have an even number denominator. Since we exclude all irrational numbers
(all non-terminating non-repeating decimals), we do not define functions like

f (x) = ax

when a is negative, or zero for that matter.

The point is that we have extended definitions before, and you have rolled
with it. The key is that the original definitions retain their meaning for the
original cases. So if we are extending the definitions of the trigonometric func-
tions (ratios), the original “SOH CAH TOA” should still be true for right angle
triangles.
If we are analysing angles within right angle triangles, the angles must be be-
tween 0◦ and 90◦ . So if we were to write sin 280◦ or cos(−10◦ ), we cannot apply
“SOH CAH TOA”, since right angle triangles cannot have angles over 90◦ or
negative angles. Now, just like in the exponents comparison, we generally extend

2
definitions because it is useful . For now, I am going to ask you to trust that
extending the definition of the trigonometric functions (ratios) beyond “SOH
CAH TOA” is super useful, because it helps us study periodic functions.

We draw a circle centred at the origin, i.e.

x2 + y 2 = r 2 (1)

We draw an angle in standard position, i.e. drawing a line segment from the
origin to the circle. The length will be the radius of the circle. We mark the
point on the circle, and then we mark the x-coordinate on the x-axis. These two
points along with the origin (so three points in total) form a right triangle. See
the attached sheet titled “L5 - Trig Applications Part 1”. So if we calculate the
trigonometric functions (again, ratios) based on “SOH CAH TOA”, we label
the three sides of the triangle: the line segment labelled y is the side opposite
θ, and the line segment labelled x is the side adjacent to θ. The hypotenuse is
clearly the radius.

This means that when you have an angle in standard position (i.e. drawn
counterclockwise from the positive x-axis about the origin), you can draw a
circle through your point (x, y), which will have a radius based on Equation 1.
Based on “SOH CAH TOA”, we can write:
y
sin θ = (2)
r
x
cos θ = (3)
r
y
tan θ = (4)
x
Note that we cannot always do this, since right triangles are not always oriented
in standard position! When they are, we can use this method, but the more
important point is this: if we need the value of a trigonometric function, the
easiest way to determine it from geometry is to draw a Cartesian plane, draw a
circle centred at the origin with radius 1 (i.e. x2 + y 2 = 1, also called the unit
circle), draw the angle in standard position, and measure the x and y coordi-
nates of the point the angle hits on the circle. Again, for angles between 0◦ and
90◦ , this corresponds with “SOH CAH TOA”.

However, we can draw any angle in standard position and obtain a point of in-
tersection with the unit circle. As previously mentioned, if the angle is greater
than 90◦ or negative, the angle could never be in a right angle triangle, and
“SOH CAH TOA” could not apply BUT we could still use

sin θ = y (5)

cos θ = x (6)

3
 Figure 1: Reference Triangles

y
tan θ =
x
for the point of intersection with the unit circle. Before you ask why , I want
to explain one more concept. When we draw an angle in standard position that
is either greater than 90◦ or negative, we can still draw a useful right angle
triangle, in which an acute angle is measured about the origin, the hypotenuse
is the radius of the circle (so 1 for the unit circle), and one of the other sides
is aligned with either the x-axis (blue triangle) or the y-axis (red triangle) (see
Figure 1).

So we can figure out b and h using the reference triangles, blue or red. Since
the hypotenuse is the radius, which is equal to 1,
adj b
cos α = = =b
hyp 1
opp h
sin α = = =h
hyp 1
adj h
cos β = = =h
hyp 1
opp b
sin β = = =b
hyp 1
We then write x = −b since x < 0, and we write y = −h since y < 0. Therefore,

sin θ = y = −h = − sin α = − cos β

cos θ = x = −b = − cos α = − sin β

y −h − sin α − cos β 1
tan θ = = = = tan α = =
x −b − cos α − sin β tan β
This is a demonstration of how to use reference (right) triangles (hypotenuse
is the circle radius, one of the other sides is aligned with the x- or y- axis) to

4
figure out trigonometric ratios for angles outside 0◦ to 90◦ . If you want things
to be very easy, use reference (right) triangles aligned with the x-axis, and then
it is a matter of applying negative signs based on the “CAST Rule”. Remember
that the hypotenuse and radius are always positive, and for right triangles, i.e.
where θ is between 0◦ and 90◦ , the opposite and adjacent sides must be positive
too, and so trigonometric ratios are always positive for angles between 0◦ and
90◦ . So when we use reference (right) triangles aligned with the x-axis, if we
call the actual angle θ and the angle in the reference triangle α (just like in
Figure 1), we simply write:

sin θ = (signS ) sin α (7)

cos θ = (signC ) cos α (8)

tan θ = (signT ) tan α (9)
How do we determine the signs? We use the “CAST Rule”, which says that in
the first quadrant, All ratios are positive, and in the other quadrants, only one
of the three ratios is positive (and the other two are negative).

So, why do we want trigonometric functions to apply beyond 0◦ to 90◦ ? We

want this because the unit circle clearly repeats every 360◦ . If you go around
one full extra time, it does not affect the point where you are on the circle.
This in turn means that sine, cosine, and tangent repeat every 360◦ . They are
periodic functions, and so we can use combinations of them to represent any
periodic signal, such as an electromagnetic wave of a telephone signal or sound
wave in a music file.

Note that the period of tan θ is actually 180◦ . You can prove this: Draw
any angle A in the first or second quadrants, and then draw A + 180◦ . You
can prove using the “CAST Rule” and a reference triangle with respect to the
x-axis, and you will end up finding that:

tan(θ + 180◦ ) = tan θ (10)