Substitution Method Calculator

Two equations become one

The other two methods on this site work on the equations as wholes — adding them, scaling them, reducing them. Substitution does something different: it takes one unknown, writes it as an expression in the others, and then that expression travels. Wherever the unknown appeared, the expression goes instead, and one equation is left with one unknown in it.

x + y = 10 and x − y = 2 shows the whole shape in four lines. The first rearranges to x = 10 − y. That expression goes into the second, which becomes an equation in y alone, solves to y = 4, and putting 4 back gives x = 6.

It is also the method that handles a case the others find awkward: an equation already written as −2x + y = −1. There is nothing to rearrange — the substitution step is the first step — and this is exactly the shape a word problem produces when one quantity is defined in terms of another. x = 2, y = 3.

Isolate the one that costs nothing

Any unknown in any equation can be isolated, and the choice decides how much of the rest of the working is fractional. The rule is short: look for a coefficient of 1 or −1. Rearranging around it needs no division and introduces no fractions.

• E1 has a coefficient of 1 on x, so it rearranges to x = … with no division and no fractions.
• E1 has a coefficient of −1 on y, so it rearranges to y = … with no division and no fractions.
• No unknown has a coefficient of ±1 anywhere, so some division is unavoidable. x in E1 has the smallest coefficient (2), which keeps the fraction as simple as it can be.

When no coefficient is ±1, substitution is working uphill and that is a reason to consider elimination instead — it can often keep whole numbers where substitution cannot. 2x + 3y = 7 and 4x + 5y = 13 is that case: isolating anything produces a fraction, the working carries it through three lines, and the answer, x = 2, y = 1, is perfectly tidy. The fractions were in the method, not in the problem.

One more thing to watch when isolating around a negative coefficient. 4x − y = 5 solved for y gives 4x − 5, not the same expression with its signs the other way round. Moving a term across the equals sign changes its sign; moving the whole equation does not.

The brackets are not optional, and neither is the sign in front of them

This is the step the method is known for losing marks on. The expression replaces a single term in the other equation, so it has to be bracketed — and if that term had a coefficient or a minus in front of it, the coefficient multiplies everything inside and the minus flips everything inside.

In x − y = 2, substituting 10 − y for x gives (10 − y) − y = 2, and expanding that bracket is where the arithmetic actually happens. Write the substitution without brackets and the second term of the expression keeps a sign it should have lost.

The calculator prints the bracketed line and the expanded line as two separate steps for exactly this reason. They are two different operations — putting the expression in, and multiplying out — and collapsing them into one line is how the error hides. When you are working by hand, writing the bracketed version down before touching it costs three seconds and removes the failure mode.

Back-substitution: use the rearranged equation

Once you have one value, the other comes from putting it back — and there is a right place to put it. Use the rearranged equation, the one already in the form “unknown = expression”. It is solved for exactly the unknown you still need, so the last step is one substitution and no rearranging at all.

Putting the value into an original equation also works and costs an extra rearrangement. What does not work is putting it into the expanded equation you solved — that one has only one unknown left in it, and it will simply confirm the value you already have.

There is a difference from elimination worth noting here, because the two methods give opposite advice. Elimination back-substitutes into an ORIGINAL equation, deliberately, because an error in the multiplying step would survive a check against a transformed one. Substitution back-substitutes into the REARRANGED equation, because rearranging is reversible and introduces nothing to check — and then the answer is verified against both originals at the end regardless. Both methods end at the same place: the final check uses the equations you were given.

When substitution is the right choice

The two methods are not interchangeable in cost, and the coefficients say which is cheaper before you start:

• Substitution wins when some unknown has a coefficient of ±1, and wins outright when an equation is already in the form y = something. Both are common in word problems, where one quantity is usually defined in terms of another.
• Elimination wins when coefficients are equal or opposite, and generally when no coefficient is ±1 — because it can multiply where substitution must divide, and multiplying keeps whole numbers.
• Matrices win from three unknowns upward, and are the only route that says what the solution set looks like when there is not exactly one.

Substitution does extend to three unknowns, and the shape is the same: isolate, substitute everywhere it appears, and you have a system one size smaller. It is more writing than elimination at that size and considerably more than row reduction, which is why it is rarely the taught choice past two.

When the second unknown cancels as well

Sometimes the substitution removes both unknowns at once. As with elimination, what remains decides which of two opposite outcomes you have:

• 0 = 0. x + y = 5 and 2x + 2y = 10 does this. The second equation is the first doubled, so substituting turns it into a statement that is true for every value — infinitely many solutions, namely x = 5 − t and y = t.
• 0 = a non-zero number. x + y = 5 and 2x + 2y = 11 substitutes down to something false, and no values make it true. There is no solution.

Neither is a failed calculation, and the distinction is the whole answer. Reporting “no solution” because the unknowns vanished, without reading what was left, gets it wrong half the time.

Related calculators

The same system, other routes:

More in Math, or browse all calculators.

Sources and methodology

The method needs no external authority — it rests on the fact that equals may be replaced by equals, which is argued in the first section rather than cited. What is cited is the standard reference statement of the linear algebra the three methods share.

• NIST Digital Library of Mathematical Functions — §3.2, Linear Algebra
• How we calculate — methodology

Method. Every figure on this page comes from src/lib/substitution-method.ts over the exact-rational linear core in src/lib/algebra/linear.ts — shared with the elimination and system pages, so all three agree on the answer while showing different work. The engine picks the unknown to isolate by looking for a coefficient of ±1 first, which is the same rule the page gives, and renders the isolated expression the way it is written by hand rather than in coefficient order. That engine is verified on every change against 62 hand-written assertions, including that all three methods agree across a swept range of systems and that every answer is checked against the original equations. The count and the per-case breakdown are published on the formula verification page.

Read the guide

This page argues that substitution suits a coefficient of ±1 and struggles without one. Elimination, Substitution, or Matrices — Choosing a Route Through a System makes that a rule you can apply in five seconds, works one system through all three routes so the trade is visible, and sets out where each method characteristically fails. If the fractions on this page are getting out of hand, that guide explains why and what to switch to.