Where implicit differentiation beats explicit

Critical points of implicit relations sit at the intersection of two AP Calculus skills that examiners test every year: implicit differentiation and the first- or second-derivative test for local extrema. The first time a candidate meets the pattern, it usually appears in a free-response question worth roughly half of the paper's marks, where the curve is given as an equation in x and y rather than as y = f(x). The candidate's job is to differentiate both sides, solve dy/dx = 0, find candidate x-values, then justify whether each candidate is a local maximum, local minimum, or neither, using either a sign chart of dy/dx or the second-derivative test applied to d²y/dx². None of the steps is exotic; what trips students up is the bookkeeping of y terms, the algebra when dy/dx collapses to a fraction with y² in the denominator, and the written reasoning that the rubric's communication point demands. This article walks through the pattern end-to-end, with the kind of language a marker is trained to credit, so the next time a free-response question opens with a curve such as x² + y² = 1 or x³ + y³ = 3xy, the critical-point method feels routine rather than improvised.

Why implicit critical points deserve a separate preparation track

Most AP Calculus students meet critical points in the explicit world first, where y = f(x) and dy/dx is a function of x alone. The exam, however, reserves a chunk of marks for problems where the relationship between x and y is locked inside a single equation that cannot be solved for y as a single explicit function, or where solving it would produce a piecewise mess that is mathematically dishonest to ignore. The classic exam-shaped cases include circles, ellipses, lemniscates, foliums, and the family of curves sometimes labelled "Cartesian ovals". The defining feature is the same: differentiating cleanly still gives you dy/dx, and solving dy/dx = 0 still gives you candidate critical points, but you cannot peel y off the right-hand side without losing information. That is exactly the situation AP Calculus examiners like to use, because the only way forward is the technique the syllabus names explicitly: implicit differentiation.

For most candidates, this is the part of the course that quietly decides whether the multiple-choice section finishes in time and the free-response section picks up the second or third point on a six-point rubric. The topic shows up in Unit 3 of the AP Calculus BC syllabus and the relevant portion of Unit 3 in AB, and it is one of the items that the official Course and Exam Description flags as a connection between differentiation and the analytic framework for sketching curves. The trick is that an implicit critical-point problem is not really a new idea; it is a chain-rule problem whose result must be set to zero, with the added constraint that the answer for y will be a function of x, not a number. Students who internalise that mental move usually pick up the marks. Students who try to convert the implicit equation into y = f(x) first usually run out of time and lose the communication point because their setup is already tangled.

A useful study heuristic is to treat the topic as a three-part ritual: differentiate both sides, factor or rearrange the resulting dy/dx expression so that dy/dx = 0 is a clean equation in x and y, and then build a sign chart or second-derivative argument using the original constraint to substitute y-values back in. The rest of this article goes through that ritual step by step, with two worked examples and the language patterns that match the AP rubric's expectations.

The chain-rule mechanics examiners want you to show

Implicit differentiation is the chain rule applied to y as an unknown function of x. The two mechanics the rubric is looking for are treating y as a function of x on every term that contains it, and applying d/dx to those terms using the chain rule, which produces a dy/dx factor each time. A common candidate mistake is to differentiate y² as 2y, dropping the dy/dx. The marker does not award the point for derivative setup when the dy/dx is missing on a y-term. The cleanest way to avoid that error is to write a small annotation next to each y-term, the way an experienced tutor would on a whiteboard: y² becomes 2y · dy/dx, y³ becomes 3y² · dy/dx, and so on. For trigonometric terms, sin y becomes cos y · dy/dx, and the same logic handles eʸ and ln y, where the derivative of eʸ is eʸ · dy/dx and the derivative of ln y is (1/y) · dy/dx.

Once the differentiation is done, collect the dy/dx terms on one side and the non-dy/dx terms on the other. The result is a linear equation in dy/dx, which factors out cleanly. From there, divide by the coefficient of dy/dx and you have the derivative expressed as a fraction whose numerator and denominator are functions of x and y. At this point the marker expects to see that the candidate is not afraid to leave dy/dx in this form, and that they are willing to substitute the constraint to simplify, especially when the problem asks for the slope at a particular point.

When the problem asks for critical points, the next mechanical move is to set dy/dx = 0. Because dy/dx is a fraction, that reduces to a numerator-zero condition, provided the denominator is not also zero. If both numerator and denominator vanish, the slope may or may not be defined, and the curve may have a cusp or vertical tangent, which is its own category of question. For ordinary critical points, the candidate writes down the numerator, sets it to zero, and uses the original implicit equation to eliminate y. The resulting x-values are the candidate x-coordinates of critical points. Plugging those x-values back through the original equation yields the y-values, and the candidate has the points themselves, which are the substrate for the next rubric line: deciding what kind of critical point each one is.

The chain-rule work is mechanical, and the more of it you can do without error, the more attention you can pay to the harder rubric line, which is justification. Examiners will not give full credit for an answer that simply lists points where dy/dx = 0. They want to see why each candidate is a local maximum, local minimum, or neither, and they want to see the reasoning tied to the curve defined by the original equation. The most efficient way to do that in a timed exam is to choose between two strategies, both of which are described next.

Choosing between a sign chart and the second-derivative test

For implicit problems, the sign-chart method is usually faster than the second-derivative test, because the second-derivative test requires another full implicit differentiation, this time of dy/dx, and then a substitution that is algebraically heavier. The sign chart, by contrast, uses the original numerator expression: pick x-values on either side of the candidate x, plug them into the numerator expression together with the y that the original equation forces, and check the sign of dy/dx. If the sign goes from positive to negative, the candidate is a local maximum; if it goes from negative to positive, it is a local minimum; if the sign does not change, the candidate is not a local extremum at all, and the candidate must say so. That last move, explicitly stating "not a local extremum" when the rubric requires classification, is where the communication point often lives. For most candidates, I'd personally pick the sign chart for a six-mark problem, because the algebra is one layer of substitution, not two.

Worked example 1: critical points of the curve x² + y² = r²

The circle is the simplest implicit curve, but the rubric rewards exactly the same moves as for a folium or a lemniscate, so practising it once with the full written language is worth the time. Take the unit case r = 1 for clarity, so the curve is x² + y² = 1. Differentiate both sides with respect to x: 2x + 2y · dy/dx = 0. Solve for dy/dx: dy/dx = −x/y. Set the numerator to zero to find candidates: −x = 0, so x = 0. Substituting x = 0 into the original equation gives y² = 1, so y = 1 or y = −1. The candidates are (0, 1) and (0, −1). At each point, check the sign of dy/dx in a small neighbourhood. For (0, 1), pick x slightly positive: then y is slightly less than 1 to keep x² + y² = 1, so y is positive. The numerator −x is negative, the denominator y is positive, so dy/dx is negative. Pick x slightly negative: y is still positive, the numerator −x is positive, so dy/dx is positive. The sign changes from positive to negative, so (0, 1) is a local maximum of the upper semicircle, which is consistent with the geometry. The same reasoning with signs reversed shows that (0, −1) is a local minimum of the lower semicircle.

What the rubric wants to see, line by line, looks like this: "Differentiating implicitly with respect to x gives 2x + 2y · dy/dx = 0, so dy/dx = −x/y. Setting the numerator equal to zero gives x = 0. Substituting into the original equation gives y = ±1. The candidate critical points are (0, 1) and (0, −1). At (0, 1), dy/dx is positive for x < 0 and negative for x > 0, so (0, 1) is a local maximum. At (0, −1), dy/dx is negative for x < 0 and positive for x > 0, so (0, −1) is a local minimum." That language is the kind of finished paragraph the AP marker can score quickly, and it lines up the three rubric categories: derivative setup, candidate identification, and justification. A common scoring trap is to write the candidate points in a list without the classification sentence; even a correct list without the classification loses the third point.

Worked example 2: critical points of the folium x³ + y³ = 3xy

The folium of Descartes is a frequent AP-style implicit curve, and it gives the candidate a chance to demonstrate the technique on an equation where the y-substitution is non-trivial. Start with x³ + y³ = 3xy. Differentiate implicitly: 3x² + 3y² · dy/dx = 3y + 3x · dy/dx. Collect the dy/dx terms: (3y² − 3x) · dy/dx = 3y − 3x². Solve: dy/dx = (y − x²) / (y² − x). Set the numerator to zero: y − x² = 0, so y = x². Substitute y = x² into the original equation: x³ + (x²)³ = 3x · x², which simplifies to x³ + x⁶ = 3x³, so x⁶ − 2x³ = 0, factor x³(x³ − 2) = 0, giving x = 0 or x = ∛2. For x = 0, y = 0, and the candidate point is (0, 0). For x = ∛2, y = (∛2)² = ∛4, and the candidate point is (∛2, ∛4). The candidate should check the denominator at these points: y² − x = (x²)² − x = x⁴ − x. At x = 0, the denominator is 0, so the derivative is undefined there, and the slope test cannot be applied. The geometry of the folium is that the curve passes through the origin with a self-intersection, so (0, 0) is not a local extremum; the candidate should say so in writing, because the rubric usually requires it. At (∛2, ∛4), the denominator is (∛4)² − ∛2 = ∛16 − ∛2, which is positive, so a sign chart is legitimate.

To classify (∛2, ∛4), pick an x slightly less than ∛2 and an x slightly greater, and read the sign of y − x² and y² − x. For x slightly less, y is forced to be close to ∛4 from the constraint, and y − x² is positive because y ≈ ∛4 and x² is smaller. For x slightly greater, y² − x is positive. The numerator y − x² is positive on the left, negative on the right (since x² overtakes y), so the sign changes from positive to negative, and the point is a local maximum. The candidate should write the classification as a sentence rather than a label, and include a brief sign-table description, because the rubric's communication point is usually awarded when the marker sees a written argument rather than a calculation dump.

The second-derivative test is also a valid route for the folium, and is worth practising once. Implicitly differentiate dy/dx = (y − x²) / (y² − x) a second time. The cleanest way is to write dy/dx as N/D with N = y − x² and D = y² − x, and apply the quotient rule. The result is a fraction with the chain-rule derivatives of N and D, which means dy/dx and d²y/dx² both appear on the right-hand side. Substituting dy/dx = (y − x²)/(y² − x) gives d²y/dx² as a function of x and y, and at the candidate point (∛2, ∛4) the sign of d²y/dx² confirms the local-maximum classification. For most candidates, the sign chart is faster; the second-derivative test is a backup when the sign chart is algebraically awkward.

What the rubric is really scoring, line by line

The AP Calculus free-response rubric for an implicit critical-point problem is usually a three-line rubric worth roughly six points, distributed across derivative setup, candidate identification, and justification. The first point is for the derivative: the candidate must show the implicit differentiation, with dy/dx appearing on every y-term, and the result must be a correct expression for dy/dx. The second point is for the candidate points: setting dy/dx = 0, solving for the candidate x-values, and substituting back through the original equation to find the y-values. The third point is the justification, which the marker reads as a finished argument: a sign chart with a sentence, a second-derivative calculation with a sign conclusion, or an equivalent written classification. The communication point, when the rubric has one, sits on top of all three, and is awarded for the readability of the work, the use of correct calculus notation, and the presence of concluding sentences.

The most common reason a candidate loses the justification point is that they have the right candidate point but the wrong sign-chart argument. A sign chart is only credible if the test x-values are on either side of the candidate and the y-values are forced by the original equation. Markers are trained to spot the candidate who picks a test point that does not actually lie on the curve, and they will not award the point. A second common reason is that the candidate writes "it is a maximum" without any supporting sentence, which the rubric's communication line tends to refuse. The third common reason is that the candidate forgets to check the denominator, so the sign chart is being applied at a point where dy/dx is undefined, which makes the whole argument invalid even if the candidate's classification is correct by accident. For most candidates reading this, the single best move is to write three sentences: one stating the test points, one stating the signs, and one stating the classification.

The table below summarises the relationship between the rubric line, the work that earns the point, and the most common error that loses it. It is a useful self-check before submitting a free-response answer.

Common pitfalls and how to avoid them

The most expensive mistake on implicit critical-point problems is dropping the dy/dx when differentiating a y-term. The reason it is expensive is that the rest of the algebra flows from that error, and the candidate ends up with a numerator and denominator that no longer represent the slope of the curve. The fix is mechanical: every time you differentiate a y-term, write the dy/dx on the right-hand side of that term, then collect. If you are not confident in the technique, do a quick sanity check on the circle x² + y² = 1, where the answer dy/dx = −x/y should fall out cleanly. If your answer does not have a y in the denominator, you have almost certainly dropped a dy/dx.

The second pitfall is treating the implicit equation as if it can be solved for y. In some questions, the equation can be solved, but the resulting function is piecewise, and the candidate is asked to consider the whole curve, not one branch. The cleanest exam move is to refuse to solve for y at all, and to keep the original equation as a constraint that is used to eliminate y at the end. Solving for y is a time sink that the rubric does not reward; the rubric rewards implicit differentiation and the use of the original equation as a constraint.

The third pitfall is to forget that critical points include points where dy/dx is undefined as well as points where dy/dx = 0. The AP Calculus definition of a critical point is a point in the domain of f where f'(x) is zero or undefined. For implicit curves, the analogous idea is a point on the curve where the slope is zero, infinite, or undefined. Candidates who only set the numerator to zero miss the cusp and vertical-tangent cases, and the marker is trained to look for that. A useful heuristic: when the denominator of dy/dx is zero at a candidate point, switch to a geometric argument or a parametric description, and classify the point accordingly.

The fourth pitfall is the sign-chart trap. Picking test x-values that are not near the candidate, or that do not actually lie on the curve, gives a sign chart that does not represent local behaviour. The fix is to pick x-values immediately on either side of the candidate, and to compute y by substitution into the original equation, not by guesswork. A sign chart that is technically correct but uses test points that are too far apart will not be marked down, but a sign chart that uses the wrong y will lose the point.

How to build a preparation routine that locks the technique in

For a topic this small, the most efficient preparation is targeted: do ten to fifteen free-response questions that all ask for critical points of an implicit curve, time yourself to roughly nine minutes per question, and mark your own work against the official scoring guidelines. The variety matters: a circle, an ellipse, a folium, a lemniscate, a curve with a parameter, and a curve that includes a trigonometric y-term. The repetition matters more: each time, the candidate writes the same three-sentence justification, the same implicit-differentiation step, and the same candidate identification, and the muscle memory of the rubric language becomes automatic. By the tenth question, the technique is no longer a thinking step; it is a habit.

A second preparation move is to keep a small notebook of the rubric language, written out in the candidate's own words. The rubric's exact phrasing is public, and the candidate is free to memorise it; the marker is not assessing originality, only correctness. Phrases like "dy/dx changes from positive to negative, so the point is a local maximum" or "the derivative is undefined at the origin, so the origin is not a local extremum" are reusable across questions, and writing them out once means the candidate does not have to invent them under timed conditions. The communication point is largely a reading-speed benefit to the marker, and these prefabricated sentences make the work scannable.

A third preparation move is to practise the second-derivative test on at least one implicit curve, even if the candidate plans to use the sign chart in the exam. The reason is that the second-derivative test exposes a different kind of error, namely a second implicit differentiation with a quotient rule applied to a fraction. The candidate who has done this once will be calmer in the exam, even if they choose the sign chart. The candidate who has not done it is more likely to switch strategies halfway through a question and lose the thread of the argument.

Finally, practise the transition from implicit differentiation to the graph. The free-response questions sometimes ask for a sketch, and the rubric will award a point for correctly identifying the maximum and minimum on the curve, plotted in roughly the right place, with a tangent line drawn at a separate point. A sketch is a quick way to pick up an additional point that is often under-claimed. The candidate should sketch the curve from the implicit equation, plot the critical points, and label them with the words "local maximum" and "local minimum" so the marker does not have to infer it.

Tying the technique back to the wider AP Calculus syllabus

Implicit critical points are not a standalone topic; they sit at a junction where differentiation, the chain rule, the chain rule applied twice, and the analytic framework for curve sketching all converge. The skill of implicit differentiation shows up in related rates, in the second-derivative test applied to implicit curves, and in the slope of a tangent line at a given point on a curve that is not a function. The skill of setting dy/dx = 0 and using the original equation as a constraint shows up in the Lagrange-multiplier-style problems that appear in BC, where the candidate is asked for extrema on a curve and the marker wants to see the same three-line argument. The skill of writing a finished justification sentence shows up in every free-response question on the paper, and is a habit that compounds across the rest of the exam.

For candidates targeting a 5 on the AP Calculus exam, the topic is a high-value item. It is small in syllabus weight, but it is reliable in appearance and reliable in scoring, which makes it a good place to bank marks. For candidates targeting a 3 or 4, the topic is a place to avoid leaving marks on the table, because the alternative is to lose points on a question whose underlying technique is mechanical and well-documented in the official Course and Exam Description. Either way, the time spent on ten free-response questions and ten sign charts is time well invested, and the language patterns described above will continue to be useful in the second half of the paper, where implicit and related-rates problems often share rubric vocabulary.

Conclusion and next steps

Critical points of implicit relations are a high-leverage AP Calculus topic: small in syllabus weight, but rich in rubric marks. The technique is a three-step ritual, implicit differentiation, candidate identification through dy/dx = 0 plus the original equation, and a written sign chart or second-derivative test, and the language of the rubric rewards exactly the same three sentences each time. Candidates who have practised ten to fifteen free-response questions of this type, who have memorised the rubric's phrasing, and who have built the habit of writing the conclusion in a sentence are well placed to pick up full marks when the topic appears on the exam. The work to do next is the worked problem set: a circle, a folium, an ellipse, a curve with a trigonometric y-term, and a curve with a parameter, marked against the official scoring guidelines and timed at nine minutes each. TestPrep Europe's targeted FRQ drills on implicit-relation critical points are a natural starting point for candidates building that repetition set.

Frequently asked questions