Here is the habit. You simplify an expression. You get an answer. Don't walk away yet. Before you circle the result, pick a small value — x = 1, or x = 2 if x = 1 is inconvenient — evaluate the original at that value, evaluate your simplified version at the same value, and check that the two numbers match. If they do, your simplification is almost certainly correct. If they don't, you have a mistake to find, and you have found it in thirty seconds instead of after the exam.

Why this works

Two expressions are equivalent if they give the same value for every value of the variable. Simplification is nothing but rewriting an expression as an equivalent one in a shorter form — see evaluate vs simplify for the distinction between the verbs. So if you correctly simplified 3x + 5x + 2 into 8x + 2, the two expressions must produce the same number for every x you plug in. In particular, they must agree at x = 1.

The contrapositive is what makes the check useful. If the two expressions disagree at even a single input, they are not equivalent, and one of them has an error. The check doesn't prove correctness — two expressions could agree at x = 1 by coincidence and disagree elsewhere — but it catches almost every real-world algebra mistake, because real mistakes (a dropped minus sign, a missed constant, a coefficient off by one) change the value at most inputs, not just at the ones you happen to skip. Think of it as a cheap unit test: it won't prove your function is bug-free, but if it fails, you know there is a bug.

Why x = 1 or x = 2 specifically

Two criteria make a value good for this check: the arithmetic is small, and the value exercises the expression, meaning each term takes a non-trivial value so a mistake anywhere will show up.

x = 1 is the default. Every power of x becomes 1 — x^2 = 1, x^3 = 1, x^{10} = 1 — so an expression like x^2 + 3x + 2 collapses to 1 + 3 + 2 = 6. The arithmetic is trivial, and every coefficient still contributes: the expression reduces to the sum of the coefficients plus the constant term.

x = 2 is the backup when x = 1 is awkward. If the original has (x - 1) in a denominator, plugging x = 1 divides by zero. If (x - 1) is a factor the simplification was supposed to cancel, x = 1 makes both sides trivially zero and the check is vacuous. In those cases, jump to x = 2: still small numbers, still exercises every term.

Walk-through: does 3(x+2) - (x - 1) = 2x + 7 hold?

Suppose you simplified 3(x+2) - (x-1) and got 2x + 7. Before committing, run the check at x = 1.

Original at x = 1: 3(1 + 2) - (1 - 1) = 3 \cdot 3 - 0 = 9 - 0 = 9.

Simplified at x = 1: 2 \cdot 1 + 7 = 2 + 7 = 9.

Both give 9. Match. Your simplification is almost certainly right. Circle and move on. Total cost: about fifteen seconds.

When the check fails — how to use the failure

The check is even more valuable when it fails, because the nature of the failure often points at the mistake.

Suppose you had simplified 3(x + 2) - (x - 1) and, in a hurry, written 3x + 7 — you forgot to subtract the x in the second bracket. Check at x = 1: original gives 9 (as above), simplified gives 3 \cdot 1 + 7 = 10. Mismatch. The simplified answer is exactly 1 too big.

Now look at the difference and retrace. A difference of exactly 1 at x = 1 often means one coefficient or constant is off by one. Scanning the work, you spot it: the -(x - 1) should have contributed -x + 1, and you kept the +1 but forgot to subtract the x.

If the mismatch is large, suspect a sign error. A flipped sign on a middle term produces a difference of twice that term's value at x = 1. The check doesn't tell you exactly where the error is, but the size of the mismatch is a strong hint.

Use a second value if the first feels suspicious

If the check matches at x = 1 but something still looks off, re-run at x = 2. If both values agree, the probability of an undetected error drops sharply — two coincidences are much less likely than one. In exam conditions, a match at x = 1 plus a match at x = 2 is enough to be confident.

Why x = 0 is the weakest check

x = 0 looks attractive — even simpler arithmetic than x = 1 — but it is the worst choice. Every term containing x^k for k \geq 1 vanishes at x = 0, so the only thing you test is the constant term. Both 3x^2 + 2x + 5 and 99x^3 - 17x + 5 evaluate to 5 at x = 0, even though they have nothing else in common.

Use x = 0 only as a quick constant-term check, never as your sole check. A match at x = 0 and a match at x = 1 together test the constant and the sum-of-coefficients — a much stronger combination.

Applies to more than simplification

The underlying principle — two expressions claimed to be equal must actually be equal at every input — shows up all over school algebra.

"When in doubt, plug in a number" is one of the most reliable debugging moves in all of algebra.

Avoid a value that is a root of either form

One subtlety. If either expression has (x - 3) as a factor — meaning it equals zero at x = 3 — then plugging x = 3 makes that side zero. If the other side is also zero at x = 3 (which it will be, if the two are correctly equivalent), both sides are zero and you have learned nothing. The check is vacuously satisfied.

So scan the expressions before picking your test value. If (x - 1) appears as a factor anywhere, avoid x = 1 and switch to x = 2. Pick a value that is not a zero of either form.

The sanity-check process as a vertical flowA top-to-bottom flow diagram. The first box says you have a simplified answer. An arrow leads down to a box that says pick x equals 1 or x equals 2. From there two arrows fan out to a pair of parallel boxes, one labelled evaluate the original at that value, the other labelled evaluate the simplified form at that value. Arrows from both boxes converge into a decision diamond that asks do the two numbers match. If the answer is yes, an arrow points to a box that says done, simplification is almost certainly correct. If the answer is no, an arrow points to a box that says go back and find the error, the size of the mismatch is a hint. simplified answer in hand pick x = 1 (or x = 2) evaluate ORIGINAL get a number evaluate SIMPLIFIED get a number do the two numbers match? yes ✓ no ✗ done — almost certainly right find the error — size hints where
The check as a vertical process. One expression in, two evaluations out, one comparison at the end. The failure branch is as valuable as the success branch — the size of the mismatch usually points at the mistake.

Close

Thirty seconds of checking saves minutes of re-doing. Better, it saves the silent marks you lose when a grader sees a wrong simplification and moves on — you never get the chance to rework it. Make the check a reflex: simplify, plug in x = 1, evaluate both sides, confirm, then commit. The way a senior developer writes a unit test immediately after writing the function — not because they distrust themselves, but because the cost of checking is tiny and the cost of not checking is everything.

One expression. One value. One comparison. Run it every time you simplify, and watch your error rate drop.