7 topics. Each one opens with its key idea, then the detail you need to use it.
1. Parametric equations: tracing a curve and eliminating the parameter
Key idea: Solve the simpler equation for t (or use a Pythagorean identity), substitute, and then check what values of x and y the parameter actually allows.
A parametric curve gives x = f(t) and y = g(t). Each value of t produces one point, so the curve comes with a starting point and a direction of travel. To find the start, plug in the first allowed t. To find the direction, check how x and y change for slightly larger t. For x = cos t, y = sin t, the point at t = 0 is (1, 0). Just after that, sin t turns positive, so the point rises into the first quadrant. The motion is counterclockwise.
To eliminate the parameter, solve the simpler equation for t and substitute into the other. From x = t + 1 you get t = x − 1, so y = 2t − 3 becomes y = 2(x − 1) − 3 = 2x − 5. With trig parameters, use an identity instead. For x = 2 cos t, y = 5 sin t, write cos t = x/2 and sin t = y/5, then square and add to get x²/4 + y²/25 = 1, an ellipse. For x = sec t, y = tan t, the identity sec² t − tan² t = 1 gives the hyperbola x² − y² = 1.
Watch the domain. If x = eᵗ and y = e²ᵗ with t ≥ 0, then y = x², but x = eᵗ can never drop below e⁰ = 1. The curve is only the piece of the parabola with x ≥ 1. The Cartesian equation alone does not tell the whole story, so state the restriction.
2. Parametric derivatives: slope and concavity
Key idea: dy/dx is dy/dt over dx/dt; the second derivative is the t-derivative of dy/dx, divided by dx/dt again.
The slope of a parametric curve is dy/dx = (dy/dt) ÷ (dx/dt). This works because the chain rule says dy/dt = (dy/dx)(dx/dt), and you just solve for dy/dx. For x = t² and y = t³, you get dy/dx = 3t²/(2t) = 3t/2. At t = 2 the slope is 3. Notice that 3t/2 is not the answer to a question asking for the slope at t = 2. You must substitute.
Horizontal tangents happen where dy/dt = 0 and dx/dt ≠ 0. Vertical tangents happen where dx/dt = 0 and dy/dt ≠ 0. If both are zero at once, the formula gives 0/0 and you need more work, so do not call it either kind.
The second derivative is where students lose the most points. You cannot just differentiate dy/dx with respect to t. The correct rule is d²y/dx² = [d/dt (dy/dx)] ÷ (dx/dt). For the same curve, d/dt (3t/2) = 3/2, and dividing by dx/dt = 2t gives 3/(4t). At t = 2 that is 3/8, which is positive, so the curve is concave up there. If you had stopped at 3/2, you would have the derivative with respect to the wrong variable.
One more trap: differentiate each component carefully. The derivative of t² is 2t, not t. Using t would make the slope 3t, or 6 at t = 2.
3. Arc length and speed
Key idea: Speed is √(x′² + y′²); its integral over the time interval is always the distance traveled, and it equals the arc length only when no part of the path is traced twice.
Over a tiny time step, a particle moves Δx sideways and Δy up. By the Pythagorean theorem, the distance it covers is about √((Δx)² + (Δy)²). Divide by Δt and take the limit, and you get speed: √((dx/dt)² + (dy/dt)²). Adding speed over time gives the length of the path:
L = ∫ from a to b of √((dx/dt)² + (dy/dt)²) dt.
The integral of speed over [a, b] is always the total distance the particle travels. It also equals the arc length of the curve, but only when the particle traces no part of the path more than once. For x = t², y = t³ from t = 0 to t = 1, the integrand is √(4t² + 9t⁴) = t√(4 + 9t²). A u-substitution gives (13^(3/2) − 8)/27, about 1.440. On the calculator section, most arc length integrals have no clean antiderivative, so set up the integral correctly and let the calculator evaluate it.
Speed at an instant is just the integrand evaluated at that time. If x′(3) = 4 and y′(3) = −3, the speed is √(16 + 9) = 5. The sign of y′ does not matter, because squaring removes it. Speed can never be negative.
The common confusion is with displacement, the straight-line change in position. Displacement uses ∫x′ dt and ∫y′ dt separately. Distance uses the square-root integral. Distance is always at least as large as the size of the displacement.
4. Vector-valued functions: position, velocity, acceleration
Key idea: Differentiate each component separately; velocity gives direction and slope, its magnitude gives speed, and the particle speeds up when the derivative of speed is positive.
A vector-valued function writes position as r(t) = ⟨x(t), y(t)⟩. Calculus on a vector works one component at a time. Velocity is v(t) = r′(t) = ⟨x′(t), y′(t)⟩. Acceleration is a(t) = v′(t) = ⟨x″(t), y″(t)⟩. There is no new rule to learn, just two ordinary derivatives side by side.
The velocity vector points in the direction the particle is moving, so it is tangent to the path. Its slope, y′/x′, is the same dy/dx you found with parametric derivatives. Its length is the speed.
The signs of the components tell you about direction. If x′(t) > 0, the particle is moving right. If y′(t) < 0, it is moving down. Questions like "Is the particle moving toward the y-axis at t = 2?" are answered by comparing the sign of x(2) with the sign of x′(2). If they have opposite signs, x is heading toward zero, so the particle is moving toward the y-axis.
To decide whether the particle is speeding up, do not look at one component of acceleration. Write the speed as a function of t and check whether its derivative is positive. For v(t) = ⟨2t, t² − 1⟩, the speed is √(4t² + (t² − 1)²) = √(t⁴ + 2t² + 1) = t² + 1. Its derivative is 2t, which equals 2 at t = 1, so the particle is speeding up there. A quick check uses the dot product: speed increases when v · a > 0. Here a(t) = ⟨2, 2t⟩, so at t = 1 you get ⟨2, 0⟩ · ⟨2, 2⟩ = 4 > 0, which agrees.
5. Motion problems: integrating vectors to find position
Key idea: Position later = starting position + integral of velocity, done component by component; distance needs the speed integral instead.
Free-response motion problems usually give you velocity, or its components, plus a starting position. You must recover position later. The Fundamental Theorem of Calculus does this one component at a time:
x(b) = x(a) + ∫ from a to b of x′(t) dt, and y(b) = y(a) + ∫ from a to b of y′(t) dt.
Suppose v(t) = ⟨2t, t² − 1⟩ and the particle is at (1, 2) at t = 0. Then x(3) = 1 + ∫₀³ 2t dt = 1 + 9 = 10. And y(3) = 2 + ∫₀³ (t² − 1) dt = 2 + (9 − 3) = 8. The particle ends at (10, 8). Forgetting to add the starting value is a very common way to lose a point.
Now compare two different questions about the same motion. The displacement from t = 0 to t = 3 is ⟨9, 6⟩, with length √117, about 10.82. The distance traveled is ∫₀³ speed dt. Here the speed is √(4t² + (t² − 1)²) = √(t⁴ + 2t² + 1) = t² + 1, so the distance is ∫₀³ (t² + 1) dt = 9 + 3 = 12. The path curves, so it is longer than the straight line from start to finish.
On calculator questions, components like x′(t) = e^(sin t) cannot be integrated by hand. Write the full expression, including the initial value, before you type it in. Readers award the setup point for the written integral, not for the number alone.
6. Polar coordinates and polar slope
Key idea: Polar slope is dy/dθ over dx/dθ with x = r cos θ and y = r sin θ; dr/dθ measures change in distance from the origin, not slope.
A polar point (r, θ) sits at distance r from the origin along the ray at angle θ. The conversions are x = r cos θ and y = r sin θ, with r² = x² + y². A negative r means you go in the opposite direction of the ray, so (−2, π/2) is the same point as (0, −2) in Cartesian form.
A polar curve r = f(θ) is a parametric curve in disguise, with θ as the parameter. Substitute r to get x = f(θ) cos θ and y = f(θ) sin θ. Then the slope is dy/dx = (dy/dθ) ÷ (dx/dθ). Using the product rule on each piece gives
dy/dx = (r′ sin θ + r cos θ) ÷ (r′ cos θ − r sin θ), where r′ = dr/dθ.
Try r = 2 sin θ, a circle of radius 1 centered at (0, 1). Here x = 2 sin θ cos θ = sin 2θ and y = 2 sin² θ = 1 − cos 2θ. So dx/dθ = 2 cos 2θ and dy/dθ = 2 sin 2θ. At θ = π/6 these are 1 and √3, so the slope is √3.
A big mistake is to treat dr/dθ as the slope. It is not. dr/dθ tells you how fast the distance from the origin changes as θ grows. When r and dr/dθ have the same sign, the point is moving away from the origin. When they have opposite signs, it is moving toward the origin. Exam stems test this meaning directly.
7. Polar area: one curve and two curves
Key idea: Area is (1/2)∫r² dθ over the interval that traces the region exactly once; between curves, subtract the squares of outer and inner radii.
The area swept out by a polar curve between θ = α and θ = β is A = (1/2)∫ from α to β of r² dθ. Each thin wedge is a sector of a circle with area (1/2)r² dθ, and the integral adds them up.
The hard part is choosing the limits. For r = 1 + cos θ, a cardioid, the curve closes after one full turn, so you integrate from 0 to 2π. Expanding (1 + cos θ)² and integrating gives an area of 3π/2. For r = 2 sin θ, the circle is fully traced between 0 and π. If you integrate from 0 to 2π, the curve traces itself twice and you get 2π instead of the true area π. Before integrating, find where r = 0, because those angles usually mark where a loop or petal starts and ends.
For the region between two curves, find where they intersect by setting the r-values equal. This misses any intersection at the pole, because each curve can reach r = 0 at a different θ, so check the pole separately. Then subtract squares, not radii: A = (1/2)∫(R² − r²) dθ, where R is the outer curve. For the region inside r = 3 cos θ and outside r = 1 + cos θ, the curves meet where 3 cos θ = 1 + cos θ, so cos θ = 1/2 and θ = ±π/3. The area is (1/2)∫ from −π/3 to π/3 of (9 cos² θ − (1 + cos θ)²) dθ, which equals π. Writing (R − r)² instead of R² − r² is a classic mistake that gives the wrong answer.
On calculator questions, intersection angles are often messy decimals. Store them in your calculator rather than rounding early.