Common Mistakes in Advanced Functions
304 common pitfalls across 76 concepts, including 76 recurring stuck points
These are the pitfalls students hit across all 76 advanced functions concepts, from Function, Domain and Range onward. Each entry names the concept it comes from, so you can tell a slip from a genuine gap and go straight to the explanation.
The procedure for function is the easy part; the trap is calling any equation in and a function. Asking "Does every allowed input give exactly one output, never two?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling any equation in and a function - check first that each gives only one (vertical line test).
Thinking one output can come from only one input - functions allow many inputs to share an output; they forbid one input giving many outputs.
Reading as times - it means the output of rule at input .
The procedure for domain is the easy part; the trap is listing where the function equals zero instead of where it is undefined. Asking "Which input values would make the rule undefined, and have I excluded exactly those?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Listing where the function equals zero instead of where it is undefined - the domain excludes undefined inputs, not roots.
Forgetting to check both a denominator and a radical in the same problem - scan for every operation that can fail.
Stating the domain as -values - the domain is always the set of inputs (), never outputs.
The procedure for range is the easy part; the trap is copying the domain as the range. Asking "Which output values does the function actually reach as runs over its domain?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Copying the domain as the range - inputs and outputs are different sets; compute outputs separately.
Forgetting outputs a rule can never produce - check for squares, absolute values, and exponentials that block negatives.
Reporting one peak value instead of the full set - the range is an interval or set, not a single number.
The procedure for inverse function is the easy part; the trap is reading as a reciprocal. Asking "If I know the output, does this rule hand back the exact input that produced it?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as a reciprocal - it means the reversing function, not .
Finding an inverse without checking one-to-one - if fails the horizontal line test, no inverse exists.
Swapping and but forgetting to solve for the new - the inverse must be expressed as output in terms of input.
The procedure for function composition is the easy part; the trap is applying before in . Asking "Is the output of one function being used as the input of another, in a fixed order?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Applying before in - the inner function always goes first.
Multiplying outputs instead of substituting - composition plugs in for the variable in , it does not multiply.
Assuming equals - order changes the result for most functions.
The procedure for exponential function is the easy part; the trap is putting the variable in the base instead of the exponent. Asking "Does the output multiply by the same factor for each equal step in ?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Putting the variable in the base instead of the exponent - exponential means the variable is the exponent, as in , not .
Treating constant-percent growth as constant-amount growth - a fixed percent is exponential, a fixed amount is linear.
Allowing the base to be or negative - exponential requires and .
The procedure for logarithm is the easy part; the trap is treating a log as division. Asking "Am I asking 'what exponent on the base gives this number?'" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating a log as division - is the exponent on , not .
Taking the log of zero or a negative number - the argument of a log must be positive.
Confusing which way it inverts - a log frees the exponent (), a root frees the base.
The procedure for euler's number is the easy part; the trap is treating as a variable to solve for. Asking "Is the growth happening continuously, with the natural base where the rate equals the amount?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating as a variable to solve for - it is a fixed constant, about .
Using base 10 or 2 for continuously compounded growth - continuous growth uses base .
Over-rounding to in precise work - keep enough digits () for accuracy.
The procedure for trigonometric functions is the easy part; the trap is mixing up which sides belong to sin versus cos. Asking "Am I linking an angle to a ratio of sides (or a point on the unit circle)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Mixing up which sides belong to sin versus cos - SOH-CAH-TOA: sin uses opposite, cos uses adjacent, both over hypotenuse.
Leaving the calculator in the wrong angle mode - match degrees or radians to the problem before computing.
Using right-triangle ratios for an obtuse or non-right triangle - those need the law of sines or cosines instead.
The procedure for periodic functions is the easy part; the trap is calling a drifting or growing pattern periodic. Asking "Does the function return to the exact same value after a fixed, repeating interval?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling a drifting or growing pattern periodic - true periodicity needs with no net change per cycle.
Reporting a multiple of the period instead of the smallest one - the period is the smallest positive repeat interval.
Confusing period with frequency - period is the time per cycle; frequency is cycles per unit, .
The procedure for polynomial functions is the easy part; the trap is allowing negative or fractional exponents. Asking "Is every term a constant times to a whole-number power, with only addition and subtraction joining them?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Allowing negative or fractional exponents - polynomial exponents must be non-negative integers.
Misreading the degree from a non-leading term - the degree is the highest power, not the first written.
Expecting asymptotes or breaks - polynomials are smooth and defined for all real inputs.
The procedure for rational functions is the easy part; the trap is forgetting to exclude inputs where the denominator is zero. Asking "Is the function a polynomial divided by another polynomial containing the variable?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forgetting to exclude inputs where the denominator is zero - those are never in the domain.
Cancelling a common factor and ignoring the resulting hole - the cancelled input stays excluded as a hole.
Assuming a vertical asymptote wherever the denominator is zero - if the factor also cancels, it is a hole, not an asymptote.
The procedure for asymptote is the easy part; the trap is believing a curve can never cross a horizontal asymptote. Asking "Is there a line the curve approaches arbitrarily closely as the input or output grows without bound?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Believing a curve can never cross a horizontal asymptote - it can cross horizontals; only verticals are uncrossable.
Confusing a zero with an asymptote - a zero is where the output is 0, an asymptote is a line approached at infinity or a forbidden input.
Reading the horizontal asymptote without comparing degrees - compare numerator and denominator degrees to find .
The procedure for piecewise function is the easy part; the trap is evaluating with the wrong piece. Asking "Does the formula used depend on which interval the input falls into?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Evaluating with the wrong piece - first locate which interval the input falls in, then use only that formula.
Ignoring the endpoint inequality - versus decides which piece owns a boundary value.
Letting intervals overlap - each input must belong to exactly one piece, or it is not a function.
The procedure for function transformation is the easy part; the trap is shifting the wrong direction inside the function. Asking "Is a known parent graph being moved, scaled, or flipped by added or multiplied constants?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Shifting the wrong direction inside the function - moves right, moves left (opposite of the sign).
Confusing inside (horizontal, on ) with outside (vertical, on ) changes - and act on outputs, and on inputs.
Applying horizontal stretches as the literal factor - a factor inside compresses by , not by .
The procedure for continuous function is the easy part; the trap is assuming defined-everywhere means continuous. Asking "Can the graph be drawn through this point without lifting the pencil?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Assuming defined-everywhere means continuous - a function can have a value at every point yet still jump.
Overlooking holes from cancelled factors - a removable hole still breaks continuity at that point.
Ignoring boundary matching in piecewise functions - the pieces must agree in value at the seams to be continuous.
The procedure for function as mapping is the easy part; the trap is rejecting a function because two inputs share an output. Asking "Does any input point to two different outputs?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Rejecting a function because two inputs share an output — repeated outputs are allowed.
Accepting a relation with one input and two outputs — that violates the function rule.
Thinking every graph is a function — use the vertical line test.
The procedure for input-output view is the easy part; the trap is letting one input map to two outputs in the box. Asking "Am I tracking what comes out for a given input, treating the rule as a sealed box?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Letting one input map to two outputs in the box - a function box gives exactly one output per input.
Reading as multiplication - it names the box's output, not times .
Thinking you must know the inner formula to use the box - you can reason from inputs and outputs alone.
The procedure for multiple representations is the easy part; the trap is treating equivalent forms as different functions. Asking "Do these different-looking forms encode the exact same input-output pairs?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating equivalent forms as different functions - check that the input-output pairs match before concluding they differ.
Forcing one representation everywhere - pick the view that answers the question most directly.
Mis-translating a verbal rule into a formula - 'three more than twice ' is , not or .
The procedure for one-to-one mapping is the easy part; the trap is confusing one-to-one with being a function. Asking "Does every output value come from at most one input?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Confusing one-to-one with being a function - a function needs one output per input; one-to-one needs one input per output.
Using the vertical line test for one-to-one - one-to-one is checked by the horizontal line test.
Calling a function one-to-one when an output repeats - any repeated output value breaks it.
The procedure for many-to-one mapping is the easy part; the trap is calling a many-to-one function 'not a function'. Asking "Do two or more distinct inputs produce the same output?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling a many-to-one function 'not a function' - it is valid; only one input giving two outputs is forbidden.
Expecting an inverse for a many-to-one function - it has none until the domain is restricted.
Confusing the direction - many inputs to one output is allowed; one input to many outputs is not.
The procedure for constant rate is the easy part; the trap is checking only one pair of points and declaring it constant. Asking "Does every equal step in the input add the exact same amount to the output?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Checking only one pair of points and declaring it constant - test several consecutive steps for the same change.
Confusing equal-ratio with equal-difference - constant rate means equal differences, not a constant unless it passes through zero.
Reading the starting value as the rate - the rate is , the per-unit change, not the value at .
The procedure for changing rate is the easy part; the trap is reporting one average rate as 'the' rate of the function. Asking "Do equal steps in the input produce different changes in the output?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reporting one average rate as 'the' rate of the function - state the interval, because the average rate generally differs from one interval to another.
Confusing average rate with instantaneous rate - average uses two points; instantaneous is the limit at one point.
Assuming a changing rate means the function always increases - the rate can change while the function rises, falls, or both.
The procedure for proportional function is the easy part; the trap is calling any straight line proportional. Asking "Does input give output , and is the ratio the same for every point?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling any straight line proportional - only lines through the origin (no ) are proportional.
Taking at a single point as the constant when the line has a starting value - for , is 3.5 at but 2.5 at ; check that the ratio is the same at every point.
Forgetting to check the origin - if it cannot be proportional even if it looks linear.
The procedure for step function intuition is the easy part; the trap is interpolating between steps. Asking "Is the output constant within each interval and changing only by sudden jumps at boundaries?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Interpolating between steps - within a step the value is constant, so a half-step input keeps the same output.
Mixing up floor and ceiling at thresholds - decide whether a partial unit rounds up (charged) or down (dropped).
Getting the boundary endpoint wrong - check whether the jump value belongs to the interval on the left or the right (open vs. closed).
The procedure for piecewise behavior is the easy part; the trap is assuming continuity for free. Asking "Does the function switch to a different rule depending on which region of the domain the input is in?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Assuming continuity for free - a piecewise definition can jump; check the limits at each boundary.
Using the wrong piece for a given input - match the input to its region before applying a rule.
Forgetting which piece owns the boundary point itself - the conditions ( vs ) decide which formula gives .
The procedure for growth vs decay is the easy part; the trap is reading as growth because 0.8 is positive. Asking "Is the quantity multiplied by the same factor each period, and is that factor above or below 1?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as growth because 0.8 is positive - any base below 1 (but above 0) is decay.
Confusing the base with the rate - convert percents: 3% decay is , not .
Treating a fixed-amount-per-period change as exponential - that's linear; exponential needs a fixed multiplier.
The procedure for saturation is the easy part; the trap is extrapolating early growth straight up. Asking "Does the quantity grow fast early, then slow and flatten toward a ceiling it never passes?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Extrapolating early growth straight up - saturation means the rate falls near the limit, so the curve bends over.
Confusing the ceiling with the final value being reached exactly - the curve approaches asymptotically, never equaling it.
Mistaking the flattening for decay - in saturation the quantity is still rising, just ever more slowly, not shrinking.
The procedure for feedback is the easy part; the trap is swapping positive and negative feedback. Asking "Does the system's output get fed back in as the input for the next step?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Swapping positive and negative feedback - positive amplifies/destabilizes, negative corrects/stabilizes.
Treating a feedback loop like a one-shot function - the rule is applied repeatedly, each output re-entering as input.
Assuming positive feedback always grows without limit - real loops often hit saturation or get checked by negative feedback.
The procedure for stability is the easy part; the trap is confusing where the equilibrium is with whether it's stable. Asking "After a small nudge, does the system move back toward the equilibrium rather than away from it?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Confusing where the equilibrium is with whether it's stable - first solve , then test the slope.
Using instead of for the test - stability depends on the derivative's magnitude, not the function value.
Forgetting the absolute value - is stable even if is negative (which adds oscillation).
The procedure for sensitivity is the easy part; the trap is calling a function 'sensitive' globally. Asking "Does a small change in the input produce a large change in the output here?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling a function 'sensitive' globally - sensitivity is local and varies with where you measure it.
Confusing a large output value with high sensitivity - what matters is the change in output per small input change, not the output's size.
Ignoring the input scale - sensitivity is a ratio , so the size of matters when reporting it.
The procedure for local vs global behavior is the easy part; the trap is generalizing a local fact to the whole function. Asking "Is the question about the function right around one point, or about its behavior across the whole domain?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Generalizing a local fact to the whole function - a tangent or approximation only describes one neighborhood.
Calling a local max the global max - a neighborhood high point may be beaten elsewhere on the domain.
Reading end behavior off a small window - global trends as aren't visible by zooming in near one point.
The procedure for functional modeling is the easy part; the trap is forcing a line onto exponential data. Asking "Are you choosing and building a function to represent a real-world relationship?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forcing a line onto exponential data - check for equal ratios (exponential) vs equal differences (linear) before choosing.
Skipping the pattern diagnosis - identify the signature in the data before writing any equation.
Trusting the model far outside the data - extrapolation assumes the pattern holds, which it often doesn't.
The procedure for dependency graphs is the easy part; the trap is drawing the arrow toward the variable that's known first. Asking "Are you mapping which variables directly influence which others with directed arrows?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Drawing the arrow toward the variable that's known first - means depends on ; the arrow points to the dependent.
Allowing a cycle when one isn't intended - a pure 'computes from' dependency graph should be acyclic (a DAG).
Confusing direct with indirect dependence - an arrow means a DIRECT influence; indirect ones come from following a path.
The procedure for scaling functions is the easy part; the trap is applying the horizontal rule like the vertical one. Asking "Is the function the same shape multiplied by a constant on the output or input (not shifted)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Applying the horizontal rule like the vertical one - compresses (factor ), it does not stretch.
Confusing amplitude (vertical) with period (horizontal) - changes amplitude; changes period.
Forgetting a negative multiplier also reflects - scales while the sign of flips the graph.
The procedure for shifting functions is the easy part; the trap is reading as a left shift. Asking "Is the graph the same shape just moved by an added constant (not stretched or flipped)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as a left shift - inside shifts reverse: moves the graph RIGHT 3.
Mixing up inside and outside constants - inside affects horizontal, outside affects vertical.
Thinking a shift changes the shape - translation only moves position; size and orientation are untouched.
The procedure for reflecting functions is the easy part; the trap is swapping which negative flips which axis. Asking "Is the graph a mirror image of the parent caused by a negative sign (not a slide or a stretch)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Swapping which negative flips which axis - outside is the -axis flip; inside is the -axis flip.
Calling asymmetric - it's even, equal to its own -axis reflection.
Confusing a reflection with a rotation - a flip mirrors across a line; it does not rotate the graph.
The procedure for composition chains is the easy part; the trap is evaluating left to right. Asking "Is one function's output being fed as the input to the next in a fixed order?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Evaluating left to right - composition is inside-out; the rightmost (innermost) function runs first.
Assuming - composition is generally not commutative; order changes the result.
Plugging the input into the wrong stage - feed into the innermost function, then pass each result outward.
The procedure for function families is the easy part; the trap is treating and as variables. Asking "Do these functions all share one general form, differing only in the values of their constants?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating and as variables - parameters are fixed for one member; only and vary.
Mixing members of different families - and aren't the same family; the general FORM must match.
Forgetting the parent - every transformed function is a member of a family with a baseline parent shape.
The procedure for invariants under transformation is the easy part; the trap is assuming all features are invariant. Asking "Does this property remain exactly the same after the transformation is applied?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Assuming all features are invariant - shifts change intercepts, scales change amplitude; only some properties survive.
Confusing invariance with symmetry - symmetry is a function being its own image; invariance is a property being preserved.
Naming the wrong feature - check the specific property against the specific transformation before claiming it's invariant.
The procedure for unit circle is the easy part; the trap is swapping and in the coordinates. Asking "Are you reading sine and cosine of an angle as coordinates on a circle of radius 1?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Swapping and in the coordinates - , , in that order.
Forgetting the sign by quadrant - cosine is negative in quadrants II and III, sine negative in III and IV.
Assuming the radius scaling - only on the UNIT circle do coordinates equal directly; a radius- circle needs a factor of .
The procedure for radian measure is the easy part; the trap is plugging a degree value into a calculator set to radians. Asking "Is the angle measured by arc-lengths-of-radius around the circle rather than by a slice of 360?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Plugging a degree value into a calculator set to radians - match the calculator mode to the unit you actually have.
Forgetting the unitless only converts degrees to radians, not the reverse - multiply by to go back to degrees.
Treating 1 radian as a 'nice' round angle - it is about 57.3°, not 60°, so estimates drift.
The procedure for trigonometric function graphs is the easy part; the trap is using directly as the period. Asking "Is the curve a repeating wave whose height, repeat length, and center I can name from , , , ?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using directly as the period - the period is , so divide.
Forgetting to factor out before reading the phase shift - the shift is , not .
Letting a negative change the amplitude - amplitude is ; the negative only flips the wave.
The procedure for inverse trigonometric functions is the easy part; the trap is treating as . Asking "Am I starting from a ratio and asking for the angle, with the answer pinned to one restricted range?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating as - the superscript means inverse function, and reciprocal is .
Reporting an angle outside the principal range - returns only , only .
Assuming for every - it only holds when is already in the restricted domain.
The procedure for pythagorean trigonometric identities is the easy part; the trap is reading as . Asking "Do I have squared trig functions I can collapse to 1 or swap using the unit-circle relation?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as - it means , sine first then square.
Misremembering the derived forms - divide by to get , not .
Setting the sum equal to the angle - is always 1, independent of .
The procedure for sum and difference identities is the easy part; the trap is splitting as . Asking "Is the argument of the trig function a sum or difference of two angles I want to break apart?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Splitting as - it expands to , a product mix.
Getting the cosine sign wrong - in cosine the inside becomes a between the products ( flips).
Using matching signs for cosine - sine and tangent keep the sign, cosine reverses it.
The procedure for double-angle identities is the easy part; the trap is writing . Asking "Is the angle exactly twice another, so I can express it from single-angle trig values?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Writing - it is , the sum identity with equal angles.
Picking the wrong form - use when you only know sine, when you only know cosine.
Doubling the function for cosine too - ; it equals .
The procedure for logarithm properties is the easy part; the trap is turning into . Asking "Is the log's argument a product, quotient, or power I can split into add, subtract, or a front multiplier?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Turning into - the sum rule applies to products , never sums.
Leaving the exponent in place - the power rule pulls out front as .
Reversing quotient and product - division gives subtraction, multiplication gives addition.
The procedure for natural logarithm is the easy part; the trap is treating as base 10. Asking "Is the base (continuous growth), so the inverse I want is rather than a base-10 log?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating as base 10 - is base ; the base-10 log is written .
Forgetting and - the log of the base is 1, the log of 1 is 0.
Not using and as inverses - and cancel directly.
The procedure for change of base formula is the easy part; the trap is flipping the ratio. Asking "Is the base something other than or 10 that I must turn into a computable ratio of logs?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Flipping the ratio - it is (target over base), not base over target.
Switching intermediate bases mid-problem - use the same base in numerator and denominator.
Thinking the new base must be 10 or e - any base works; those two are just the ones on the calculator.
The procedure for solving exponential equations is the easy part; the trap is dividing instead of taking a log. Asking "Is the unknown stuck up in the exponent, so I need a logarithm to bring it down?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Dividing instead of taking a log - an exponent comes down only via a logarithm, never division.
Forgetting the power rule after taking the log - rewrite as to expose .
Confusing it with a log equation - here the variable is the exponent, so you take a log, not exponentiate.
The procedure for solving logarithmic equations is the easy part; the trap is skipping the domain check. Asking "Is the unknown the argument of a log, so I rewrite as to release it?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Skipping the domain check - reject any solution that makes a log's argument zero or negative.
Converting before combining - merge multiple logs into one with the properties, then exponentiate.
Mixing up direction - the variable is inside the log, so exponentiate; do not take another log.
The procedure for equation of a circle is the easy part; the trap is reading the center sign backwards. Asking "Are and present with equal positive coefficients and a constant on the other side?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading the center sign backwards - gives center , so means center at .
Forgetting to square-root the right side - the equation gives , so radius is .
Treating it as an ellipse - a circle requires equal coefficients on and .
The procedure for ellipse is the easy part; the trap is using for the foci. Asking "Are both squared terms positive, added, with different denominators equaling 1?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using for the foci - an ellipse uses ; the plus version is for hyperbolas.
Assuming is always under - the major axis lies under the LARGER denominator, which may be .
Confusing it with a circle - unequal denominators mean an ellipse, not a circle.
The procedure for hyperbola is the easy part; the trap is using for the foci. Asking "Is one squared term subtracted from the other (opposite signs) with the result equaling 1?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using for the foci - a hyperbola adds: .
Reading the opening direction wrong - the branch opens along the variable of the POSITIVE term.
Mistaking the larger denominator for - in a hyperbola sits under the positive term regardless of size.
The procedure for parabola (focus-directrix definition) is the easy part; the trap is equating with the coefficient . Asking "Is the curve the set of points equally far from a single point and a single line, with only one variable squared?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Equating with the coefficient - convert via ; is the vertex-to-focus distance.
Putting focus and directrix on the same side of the vertex - they sit opposite, each a distance away.
Expecting two squared terms - a parabola has exactly one variable squared.
The procedure for conic sections overview is the easy part; the trap is reading as a root count. Asking "Am I deciding the TYPE of conic from a general equation rather than analyzing one specific known conic?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as a root count - its SIGN names the conic type, not the number of solutions.
Ignoring the refinement - a negative discriminant is an ellipse, or a circle if also and .
Skipping classification - identify the conic first, then pull the matching standard form.
The procedure for polar coordinates is the easy part; the trap is reading as . Asking "Is the location given as a distance from the origin plus an angle, rather than horizontal and vertical amounts?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as - the first number is a distance, the second an angle.
Mishandling the angle's quadrant in - arctan alone can land in the wrong quadrant; check the signs of and .
Forgetting a point has many polar names - adding to (or negating and adding ) gives the same point.
The procedure for polar graphs is the easy part; the trap is miscounting rose petals. Asking "Is the equation expressed in terms of , traced by sweeping the angle around?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Miscounting rose petals - odd gives petals, even gives .
Plotting as on a square grid - use the polar grid of concentric circles and radial lines.
Ignoring negative - when the point is plotted in the opposite direction, completing many curves.
The procedure for parametric equations is the easy part; the trap is believing eliminating keeps everything. Asking "Are and each written as a function of a separate parameter that drives both together?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Believing eliminating keeps everything - the resulting -equation drops direction and speed.
Forcing the curve to be a function - parametric paths can loop or cross, failing the vertical-line test.
Treating as - is an independent clock that drives both and , not the horizontal axis.
The procedure for parametric graphs is the easy part; the trap is connecting points by order. Asking "Am I sketching or analyzing the actual traced path of , including its direction or tangent?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Connecting points by order - join them in increasing- order, since the path can reverse.
Taking as - it is , the -rate over the -rate.
Ignoring direction arrows after eliminating - the Cartesian shape hides which way the curve is traced.
The procedure for compound interest is the easy part; the trap is using the annual rate directly when compounding more than once a year. Asking "Does each period's interest get added to the balance so the next period earns on a larger amount?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using the annual rate directly when compounding more than once a year - divide by to get the periodic rate and use exponent .
Forgetting to convert the percent to a decimal - enters the formula as , not .
Treating compound interest like simple interest by multiplying - you must raise to the power, not multiply principal by rate by time.
The procedure for annuities is the easy part; the trap is using the annual rate as . Asking "Is the same amount paid repeatedly at fixed intervals, rather than once?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using the annual rate as - the periodic rate is the annual rate divided by payments per year (e.g. monthly gives ).
Confusing with years - is the TOTAL number of payments (months years for monthly), not the number of years.
Mixing up future-value and present-value forms - use to find what deposits grow to, and to find what a future stream is worth now.
The procedure for present and future value is the easy part; the trap is discounting when you meant to grow (or vice versa). Asking "Am I moving one sum of money forward (growing) or backward (discounting) along the timeline?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Discounting when you meant to grow (or vice versa) - dividing by moves money BACK to today, multiplying moves it forward.
Comparing future and present amounts directly without converting - bring both to the same point in time before judging which is larger.
Using the wrong sign or direction in NPV - the initial cost at is divided by and is usually negative.
The procedure for lines in 3d is the easy part; the trap is confusing the point with the direction vector. Asking "Does the line live in space and need a direction vector rather than a single slope?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Confusing the point with the direction vector - is where the line is, is which way it goes; they play different roles.
Writing symmetric form when a direction component is zero - if you cannot divide by it, so keep that coordinate as a separate equation like .
Assuming two lines with different direction vectors must cross - in 3D they can be skew (never meet and never parallel).
The procedure for planes in 3d is the easy part; the trap is using a direction vector instead of the normal vector. Asking "Is the object a flat surface fixed by a perpendicular (normal) vector and one equation ?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using a direction vector instead of the normal vector - a plane is built from the vector PERPENDICULAR to it, not one lying in it.
Reading the coefficients wrong - in the normal is (the coefficients), and alone does not affect orientation, only position.
Forgetting the absolute value or the in the distance formula - distance is always nonnegative and divided by the normal's length.
The procedure for function notation is the easy part; the trap is reading as times . Asking "Are the parentheses holding an input to a named function (not a multiplication)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as times - the parentheses hold the input; multiplication never happens here.
Substituting only part of a compound input - in replace every in the rule with the whole expression .
Confusing the input with the output - in , the 3 goes in (input) and the 10 comes out (output), not the reverse.
The procedure for symmetric functions is the easy part; the trap is judging symmetry from a rough sketch instead of algebra. Asking "When you replace with , do you get back (even), (odd), or neither?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Judging symmetry from a rough sketch instead of algebra - substitute and compare; a picture can mislead.
Assuming a function must be even or odd - many are neither, so the test can come back 'no symmetry.'
Forgetting to distribute the negative through powers - but ; sign behavior depends on the exponent.
The procedure for restricted domain is the easy part; the trap is restricting to an interval where the property still fails. Asking "Am I deliberately limiting the inputs so the function gains a property (like passing the horizontal line test)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Restricting to an interval where the property still fails - check that the chosen interval truly makes outputs unique (passes the horizontal line test).
Confusing the natural domain with a restricted one - the natural domain is forced by the rule; a restriction is a deliberate extra choice.
Forgetting the restriction carries into the inverse - the restricted domain becomes the inverse function's range, which is why only.
The procedure for horizontal line test is the easy part; the trap is using vertical lines instead of horizontal. Asking "Does every horizontal line cross the graph at most once?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using vertical lines instead of horizontal - vertical lines test 'is it a function,' horizontal lines test 'is it one-to-one.'
Declaring a function has no inverse forever - it fails on its full domain but can pass after a domain restriction.
Misreading 'at most once' as 'exactly once' - a horizontal line may miss the graph entirely; what matters is that it never hits twice.
The procedure for amplitude is the easy part; the trap is using the full crest-to-trough height as amplitude. Asking "Am I measuring the vertical distance from the midline to a peak (half the total swing)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using the full crest-to-trough height as amplitude - amplitude is half that: .
Taking as signed when it can be negative - amplitude is , always nonnegative.
Confusing amplitude with period - amplitude is vertical (height), period is horizontal (cycle length).
The procedure for frequency is the easy part; the trap is reporting frequency as the period. Asking "Am I counting how many complete cycles happen per unit (not the length of one cycle or its height)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reporting frequency as the period - they are reciprocals, so , never equal unless .
Confusing frequency with amplitude - frequency is how often it repeats, amplitude is how tall it is.
Mixing frequency with angular frequency - the coefficient inside the sine is , not itself.
The procedure for parent functions is the easy part; the trap is treating a shifted/stretched graph as a new family. Asking "Is this the simplest untransformed template that all others in the family are built from?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating a shifted/stretched graph as a new family - peel off the transformations to find the underlying parent.
Forgetting the parent's own domain and range - starts at , so its transformations inherit that restriction.
Mixing up parent shapes - is a U, is a sharp V, is an S; learn the distinct templates.
The procedure for exponential growth is the easy part; the trap is modeling percent growth as linear (adding a fixed amount). Asking "Is the quantity multiplied by the same factor each period (rather than having a fixed amount added)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Modeling percent growth as linear (adding a fixed amount) - growth by a rate is multiplication by , so use a power, not a slope.
Adding the rate instead of using - growth multiplies by each period, not adds .
Confusing the growth factor with the rate - in , but the multiplier each step is .
The procedure for even and odd functions is the easy part; the trap is assuming 'not even' means 'odd'. Asking "Does equal (even), equal (odd), or neither?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Assuming 'not even' means 'odd' - run both tests; a function can be neither.
Sign-handling powers wrong - (an even power cancels the negative sign) but (an odd power keeps it).
Eyeballing symmetry from a sketch - confirm algebraically by substituting , not by visual guess.
The procedure for radians is the easy part; the trap is leaving the calculator in degree mode for radian work (or vice versa). Asking "Is the angle measured by arc-length-over-radius (so a full circle is , not )?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Leaving the calculator in degree mode for radian work (or vice versa) - match the mode to the unit before computing a trig value.
Converting wrong - multiply degrees by to get radians, by to go back.
Treating as having units of length - the radius cancels, so radians are dimensionless (a pure ratio).