Practical algebra guide
How I Solve for X Without Losing Track of the Algebra
By Evan LiUpdated
When I check an equation, the final value of x is not the part I trust first. I want to know which operation changed the equation, where a restriction entered the problem, and whether the answer still works in the original line. That habit is what I use here: simplify one move at a time, keep the equation balanced, and verify the candidates before I call them solutions.
I start by deciding what kind of equation I am looking at
My first mistake used to be reaching for a memorized formula before I had classified the equation. A line such as 2x + 5 = 15 only needs inverse operations. x² - 5x + 6 = 0 needs quadratic thinking. |2x - 5| = 9 creates two cases. A denominator containing x introduces forbidden values, while a square root can create extraneous roots after squaring. The method gets much easier once I know which of those situations I am in.
I also rewrite unusual symbols before doing any algebra. I treat × as multiplication, ÷ as division, x² as x^2, and I remove visual clutter that does not change the mathematics. That makes it much easier to see whether x appears only to the first power, inside a denominator, under a square root, or on both sides of the equals sign.
For linear equations, I use one reversible operation at a time
With 2x + 5 = 15, I subtract 5 from both sides before I divide by 2. I prefer writing each change on its own line because sign mistakes become visible immediately. The same idea works when x appears on both sides: with 5x - 3 = 2x + 12, I first collect the x terms, then move the constant, then divide by the remaining coefficient.
If the x terms disappear completely, I stop and inspect what is left. A false statement such as 5 = 9 means there is no solution. A true statement such as 2 = 2 means the original equation is an identity and has infinitely many solutions. I do not force those cases into a fake numerical answer.
For quadratics, I try factoring before I reach for the formula
For x² - 5x + 6 = 0, factoring is the cleanest path because the numbers 2 and 3 multiply to 6 and add to 5. That gives (x - 2)(x - 3) = 0, so the roots are 2 and 3. When factoring is not obvious, I move everything to one side, identify a, b, and c, and use the quadratic formula.
I keep the exact form when it carries useful information. For x² - 2 = 0, √2 tells me more than 1.41421356. I still use a decimal when I need a numerical comparison, but I treat it as an approximation of the exact root rather than a different answer.
Absolute value, rational, and radical equations each have one extra trap
For |2x - 5| = 9, I split the problem into 2x - 5 = 9 and 2x - 5 = -9. For a rational equation, I write the denominator restrictions before cross multiplication; if x - 2 is in a denominator, x = 2 is forbidden even if later algebra cancels that factor. For a radical equation, I expect squaring to create candidates that may fail the original equation.
Those are exactly the places where a quick-looking solution can be wrong. The extra check is not optional decoration: it is part of the method. A rational candidate that makes a denominator zero is not a solution, and a radical candidate that only works after squaring is extraneous.
I finish by checking the original equation, not the last simplified line
This is the habit that catches the most errors for me. I substitute every numerical answer back into the original input and compare the left and right sides. If both evaluate to the same value and the domain restrictions are satisfied, the result has survived a much stronger check than simply matching the last line of my algebra.
The calculator on this site follows the same idea. It separates the solving steps from verification so I can see both how a candidate was produced and whether it actually satisfies the input. When an equation is outside the supported forms, I would rather see an explicit unsupported message than a confident-looking guess.