2.3.6 · D2Coordinate Geometry

Visual walkthrough — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

1,939 words9 min readBack to topic

We assume you know only two things: how to plot a point on a grid, and how to subtract. Everything else is built here.


Step 1 — What is a point on a grid?

WHAT: A grid has two number-lines crossing at right angles. The flat one is the ==-axis== (how far right), the upright one is the ==-axis== (how far up). A point is an address written : go right, then up.

WHY: Before we can say what a line is, we must be able to name where any single dot sits. The whole subject is: which dots obey a rule?

PICTURE: Below, the pale-blue dot lives at — three steps along the flat axis, two steps up the upright one.

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

Step 2 — Two dots, and the "rise" and "run" between them

WHAT: Take two points, call them and . The little subscript numbers and are just name tags — "point one" and "point two". The horizontal gap is the run ; the vertical gap is the rise .

WHY: A line is about steepness, and steepness is "how much up for how much across." We cannot talk about that until we can measure "up" and "across" between two dots.

PICTURE: The blue horizontal arrow is the run; the pink vertical arrow is the rise. Together they make a right-angled corner — a little right triangle with the line as its slanted top.

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

Step 3 — Slope: bottling steepness into one number

WHAT: Divide rise by run. That single number is the slope, written :

Each symbol: is the answer (steepness), the top is how far up, the bottom is how far across.

WHY divide? Because we want "up per one step across." Dividing rise by run gives exactly the up-amount for a run of . A slope means: every 1 step right, 3 steps up. One number now captures the whole tilt.

WHY is it the same everywhere on the line? Pick any two points on a straight line and the triangle you get is always the same shape (similar triangles) — so riserun never changes. This constancy is the secret ingredient of the next step.

PICTURE: Two different triangles on the same line — a small one and a big one. Their riserun ratios are equal, shown as equal slope .

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

Step 4 — The birth of point–slope form

WHAT: Fix ONE known point and the slope . Now let be a mystery point — any dot on the plane. Ask: when does the mystery dot lie on our line?

WHY this question? A line is an infinite crowd of points; we can't list them. Instead we hunt for the condition a dot must obey to belong. Here is the condition: the slope measured from to must equal our known — because slope is the same everywhere (Step 3).

Now multiply both sides by to clear the fraction:

WHY multiply? The fraction form secretly bans the point itself (it would give ). Multiplying through heals that: now plugging gives , true — so the known point is legally on its own line.

PICTURE: The fixed point (yellow) anchored, a mystery point (blue) sliding along; the right-triangle riserun always keeping the same slope .

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

Step 5 — Slope–intercept: choose the cleverest known point

WHAT: Instead of any known point, use the one where the line crosses the -axis: the point . The number is the ==-intercept== — the height of the line when .

WHY this choice? Because putting makes a term vanish, and the equation collapses to something you can read at a glance. We are not learning a new law — just feeding a smart point into Step 4.

Substitute into point–slope:

Add to both sides:

PICTURE: The line piercing the -axis at height (yellow dot on the upright axis), then climbing for each unit right.

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

Step 6 — Two-point form: when nobody handed you the slope

WHAT: Now you are given two points and but no slope.

WHY no new formula? Slope is computable from two points (Step 3). So: compute first, then pour it back into point–slope (Step 4). Two points pin down a line completely — there is exactly one straight line through them.

Divide both sides by to display the twin ratios:

PICTURE: Two given dots (yellow), the line drawn through them, and a mystery dot (blue) whose little triangle matches the big triangle of the two given dots — equal ratios.

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

Step 7 — Standard form: catching the lines slope–intercept misses

WHAT: Sweep everything to one side of the equals sign to get

The condition just forbids the useless "" that describes no line.

WHY do we need it? Start from (renamed intercept to so it doesn't clash). Move all terms left:

That handles every tilted line. But a vertical line () has no slope at all — you cannot write it as . In standard form it is simply . This is the whole reason standard form exists: it is the only form roomy enough to hold vertical lines too.

To read the slope back out, solve for (needs ):

PICTURE: Three lines on one board — a tilted one (blue) with its , a horizontal one (pink, impossible? no: ), and a vertical one (yellow, ) that slope–intercept simply cannot express.

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

Step 8 — The degenerate cases, shown honestly

WHAT & WHY: A derivation is only trustworthy if it survives the broken inputs.

  • Horizontal line — every point at the same height . Rise for any run, so . Equation . In standard form: .
  • Vertical line — every point at the same across-value . Run , so is undefined (you cannot divide by zero). No possible. Equation , i.e. .
  • Single point given, no slopenot enough information; infinitely many lines pass through one dot. You need a second fact (a slope or a second point).

PICTURE: The horizontal wall of flatness () and the vertical wall ( undefined) side by side, with the "run ⇒ can't divide" warning on the vertical.

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)

The one-picture summary

Everything above is one tree. Slope is the root. Point–slope is the trunk. The other three forms are branches — each just point–slope with a special choice or a rearrangement.

Figure — Equations of a line — slope-intercept, point-slope, two-point, standard (ax+by+c=0)
Recall Feynman retelling — the whole walk in plain words

Picture a ramp. First I learn to name any spot on the floor-and-wall grid: "3 across, 2 up." Then I put two dots on the ramp and measure how far across and how far up it goes between them — that's run and rise. Dividing up by across gives one honest number, the slope: how much I climb for each single step forward. Because a ramp is straight, that number is the same no matter which two dots I pick.

Now the trick: I fix one dot I know and let a mystery dot wander. The mystery dot is on my ramp exactly when the slope from my dot to it equals my known slope — write that down, clear the fraction, and out pops : the parent of every line-equation. Feed it the crossing-the-wall dot and it tidies into ("start at height , climb "). Give it two dots and I just compute the slope first, then reuse the parent — that's two-point form. Sweep it all to one side and I get , the only roomy enough box to also hold walls — vertical lines that have no "climb per step" and so refuse . Same seed, four hats.


Connections