Graphing Before You Calculate
Most precalculus classes teach you to plug numbers into algebra first and check the graph later. That habit causes problems when you hit limits. A graphical approach flips the order. You look at the shape of the function, identify what is actually happening near the point of interest, and then verify with algebra if the situation is ambiguous. The method works because limits are fundamentally about behavior in a neighborhood, not about the value at the point itself. The core procedure is straightforward once you get past the textbook diagrams that make everything look clean. Start by sketching or plotting the function around the x-value you care about. I recommend using Desmos or GeoGebra for the initial plot, then moving to a hand sketch once you understand the structure. Look for three things: where the curve approaches from the left, where it approaches from the right, and whether either side blows up or jumps. If the left and right paths meet at the same height, the limit exists. If they diverge, the limit does not exist. I worked through a problem last semester that exposed how easily students misread a removable discontinuity. The function was f(x) = (x² - 4)/(x - 2). At x = 2, the graph shows a clean hole at y = 4. A student told me the limit did not exist because the function was undefined there. That is the mistake. The hole is not an asymptote. The limit still exists and equals 4. I had them zoom into Desmos at x = 2 with the window set to [1.9, 2.1] on the x-axis and [3.5, 4.5] on the y-axis. Once they saw the curve passing through that gap without changing height, the concept clicked. The algebra confirms it: factor the numerator to (x + 2)(x - 2), cancel the problematic term, substitute x = 2 into x + 2, and you get 4. The graph tells you what to expect. The algebra gives you the exact number.
Vertical asymptotes behave differently and cause more trouble than holes ever will. Consider lim(x0) of 1/x² versus lim(x0) of 1/x. The first one approaches positive infinity from both sides because squaring eliminates the sign. The second approaches negative infinity from the left and positive infinity from the right. Students frequently conflate these two cases. The graphical fix is to observe the arrows on your sketch. If both sides point upward, the limit is infinity. If they point in opposite directions, the two-sided limit DNE. Writing infinity as an answer is technically shorthand for an unbounded limit. The limit does not exist in the strictest sense, but infinity is the conventional way to describe which kind of non-existence you are dealing with. Your instructor usually wants to see "DNE" or "" depending on how strict the course is. One counter-intuitive detail that rarely makes it into introductory texts involves piecewise functions at the boundary. Take a function defined as x² for x
1 and 2x - 1 for x 1. The limit as x approaches 1 from the left uses x², which gives 1. The limit from the right uses 2x - 1, which also gives 1. The function is continuous there. But now change the second piece to 2x + 3 for x 1. The right-hand limit becomes 5 while the left-hand limit stays at 1. The graphical approach reveals an immediate jump discontinuity. The curve breaks. No algebra is required to conclude the two-sided limit does not exist. The visual evidence is the primary argument. You only reach for algebra when the graph looks like it might be approaching a single value but you cannot read the exact number from the plot. Squeeze theorem applications are another area where graphics provide intuition before formal proof. For lim(x0) of x²sin(1/x), the sine term oscillates between -1 and 1 while x² shrinks toward zero. Plotting this shows the function trapped between the parabolas y = x² and y = -x², both of which meet at the origin. The squeezing visual makes the limit obvious even though the function oscillates infinitely near zero. The formal proof follows the sandwich inequality, but the graph is what tells you the answer is actually 0 and not some messy undefined expression.
Horizontal asymptotes map directly to limits at infinity. If you see the graph leveling off at y = 3 as x grows large in either direction, then lim(x) f(x) = 3 and lim(x-) f(x) = 3. This relationship is exact. The asymptote is the limit. No trickery involved. Rational functions follow the degree comparison rules, but the graphical reading works regardless of how complicated the rational expression appears. Long division, partial fractions, and polynomial reduction all produce the same visual endpoint. The main limitation of relying on a graphical approach is precision. You can tell whether a limit exists by looking at a graph. You cannot reliably extract an exact irrational value from a screen plot. When the answer involves or 2 or some other non-terminating decimal, the graph points you in the right direction but cannot replace the algebraic manipulation. I usually spend about 30 seconds scanning the graph to confirm my algebraic result, not 30 minutes trying to read coordinates off a grid. The workflow is graph first for intuition, algebra second for the precise value. A second bottleneck occurs with functions that have extremely sharp turns or micro-scale behavior near the point of interest. A function like sin(1/x) near x = 0 oscillates faster than any reasonable graphing window can display. The plot will show a black smear. The graphical approach fails here because you cannot visually distinguish the oscillation pattern. The algebraic reasoning about bounds and the squeeze theorem becomes necessary. I learned this the hard way when a student insisted the limit of sin(1/x) as x approaches 0 was zero because their graph looked like it settled at the axis. It had not settled. The oscillation amplitude stayed at 1 the entire time. The window was just too wide to show the truth.
Get the Full Details
For anyone working through this material, the practical recommendation is to treat the graph as a hypothesis generator. It tells you what might be true. Then verify with the appropriate analytical method. When the graph and the algebra disagree, one of them is wrong. Usually the graph is wrong because of zoom level or resolution. Check your window settings first. Then check your algebra. That order saves time.
