This is a thinking bank for the parent topic. No heavy arithmetic here — every item targets a misconception or a boundary case. Cover the reveal, commit to an answer with a reason, then check.
Before the traps, ground the picture. A term is a number (the coefficient) glued onto a variable part. Think of the variable as a type of box, and the coefficient as how many boxes you have.
Look at the figure: three green boxes labelled x make 3x; five pink boxes labelled y make 5y. You can count more green boxes, but you can never turn a green box into a pink one — that is the whole reason 3x and 5y can't merge into one term.
Now watch what "combining like terms" looks like when you slide matching boxes together, and what the Distributive Property "pull-out" does:
The distributive step 3x+5x=(3+5)x is literally gathering all the x-boxes into one pile and counting them — the box type x is factored out front, and only the counts 3 and 5 add.
Both 3x and 5x share the box type x, so the Distributive Property lets us pull it out as (3+5)x; x and y share no common factor, so nothing can be pulled out.
Why does the coefficient change during addition while the variable part stays the same?
Adding like terms is counting how many boxes you have; the box (the variable part) is unchanged — you only update the count in front.
Why must the minus sign hit every term inside a bracket?
A leading minus means "multiply the whole bracket by −1", and by the distributive property −1 multiplies each term individually, flipping every sign.
Why do exponents stay fixed when adding but grow when multiplying?
Adding x3+x3 counts boxes of the same power → still power 3; multiplying x3⋅x3stacks the multiplications (x used 3+3 times) → power 6.
Why is "adding the opposite" a safer way to do subtraction of expressions?
Rewriting A−B as A+(−B) forces you to distribute the negative once, up front, so you never forget an inside sign later.
Why can a coefficient be a fraction or decimal but the counting picture still works?
You can have "two-and-a-half boxes" — a partial count is still a count, so 2.5x+1.5x merges its counts to 4x exactly like whole numbers do.
Is 5x−5x a "like-terms" problem, and what is the result?
Yes — same box type; combining gives (5−5)x=0x=0, a degenerate case where the term disappears entirely.
What is 0⋅x2+7?
The 0x2 term contributes nothing, so the expression is just the constant 7; a zero coefficient erases its term.
Are x and 1x different terms?
No — a variable with no visible coefficient has an understood coefficient of 1, so x and 1x are literally the same term.
Can an expression like 3x2+7x−10 be simplified further?
No — all three terms have different variable parts (x2, x1, x0), so there are no like terms to combine; it is already simplest. See Polynomials.
If you subtract an expression from itself, e.g. (4a+3)−(4a+3), what happens?
Every term cancels its twin: 4a−4a=0 and 3−3=0, leaving 0 — the additive-identity edge case.
Is x0 a variable term or a constant?
A constant — x0=1 for any nonzero x, so a term like 5x0 is really just the number 5 and combines with other constants. (Order matters here; see Order of Operations.)
Does the order you write like terms in the final answer matter, e.g. 8y+9x vs 9x+8y?
No — addition is commutative so both are equal; convention just usually writes higher-degree or alphabetically-first terms first for tidiness.
Can a coefficient like 21x+21x ever become a whole-number term?
Yes — fractional counts can sum to a whole: 21+21=1, giving exactly 1x=x.
Recall One-line survival summary
Only combine terms with the same variable and the same power; when you do, only the coefficients change (whole, fraction, or decimal); a leading minus flips every sign inside its bracket, worked deepest-bracket first.