Substitution vs Elimination: Solving Systems of Equations
Two standard methods for solving a system of two linear equations are substitution and elimination (also called the addition method). Substitution replaces one variable with an expression from one equation into the other. Elimination adds or subtracts the equations to cancel one variable directly.
Both methods always give the same answer — the choice is about which is faster and cleaner for a given system. This guide compares them side by side with step-by-step examples and clear guidance on when each method wins.
| Property | Substitution | Elimination |
|---|---|---|
| Core idea | Solve for one variable, substitute into other equation | Add/subtract equations to cancel one variable |
| Best when | One equation is already solved for a variable | Coefficients are equal or easy to make equal |
| Number of steps (typical) | 3–4 (solve, substitute, solve, back-substitute) | 3–4 (multiply, add, solve, back-substitute) |
| Fractions introduced | Often, if variable has coefficient ≠ 1 | Less often — can multiply to avoid fractions |
| Works for non-linear? | Yes — used in quadratic-linear systems | Rarely — designed for linear systems |
| Handles parallel lines | Gets 0 = non-zero (no solution) | Gets 0 = non-zero (no solution) |
| Handles coincident lines | Gets 0 = 0 (infinitely many solutions) | Gets 0 = 0 (infinitely many solutions) |
| Graphing interpretation | Replaces one axis | Eliminates one axis |
| Preferred in | One variable has coefficient 1 or −1 | Coefficients are multiples of each other |
| Matrix connection | Gaussian back-substitution | Row reduction (Gaussian elimination) |
Substitution Method — Step by Step
The substitution method replaces one variable with an expression derived from one equation, turning a two-variable system into a one-variable equation.
Steps: (1) Solve one equation for one variable (pick whichever variable has coefficient 1 or −1 to avoid fractions). (2) Substitute that expression into the other equation. (3) Solve the resulting single-variable equation. (4) Back-substitute to find the first variable.
Example: Solve: 2x + y = 10 and x − 3y = −5. Step 1: From the second equation, x = 3y − 5 (coefficient of x is 1 — ideal). Step 2: Substitute into first: 2(3y − 5) + y = 10 → 6y − 10 + y = 10 → 7y = 20 → y = 20/7. Step 3: x = 3(20/7) − 5 = 60/7 − 35/7 = 25/7. Solution: (25/7, 20/7).
When fractions appear: If neither variable has coefficient ±1, substitution introduces fractions early. For example, solving 3x + 5y = 11 for x gives x = (11 − 5y)/3 — messy. In this case, elimination is usually cleaner.
Non-linear systems: Substitution is often the only practical method when one equation is quadratic. Example: y = x² and y = x + 2. Substitute: x² = x + 2 → x² − x − 2 = 0 → (x − 2)(x + 1) = 0. Elimination cannot remove a squared term.
Elimination Method — Step by Step
The elimination method multiplies one or both equations by constants so that one variable has equal and opposite coefficients, then adds the equations to cancel that variable.
Steps: (1) Write both equations in standard form (Ax + By = C). (2) Multiply one or both equations so the target variable has opposite coefficients. (3) Add the equations — the target variable cancels. (4) Solve the resulting single-variable equation. (5) Back-substitute to find the eliminated variable.
Example: Solve: 3x + 2y = 16 and 5x − 2y = 8. Step 2: y already has coefficients +2 and −2 — no multiplication needed. Step 3: Add: (3x + 5x) + (2y − 2y) = 16 + 8 → 8x = 24 → x = 3. Step 5: 3(3) + 2y = 16 → 2y = 7 → y = 3.5. Solution: (3, 3.5).
When to multiply: 2x + 3y = 7 and 4x − y = 5. To eliminate x: multiply first equation by −2 → −4x − 6y = −14. Add to second: (4x − 4x) + (−y − 6y) = 5 − 14 → −7y = −9 → y = 9/7. Then back-substitute for x.
Choosing which variable to eliminate: Look for the smallest common multiple of the coefficients. If coefficients are already equal or one is a multiple of the other, elimination requires at most one multiplication step.
When to Use Each Method
Choose substitution when: One equation is already solved for a variable (y = 2x − 3 form). One variable has coefficient 1 or −1, making isolation fraction-free. The system is non-linear (quadratic-linear, for example). There are three or more variables and one equation is already in simple form.
Choose elimination when: Both equations are in standard form (Ax + By = C). Coefficients share a common factor (e.g., 4x in one and 2x in another — multiply by −2). You want to avoid fractions — choose which variable will cancel based on LCM of coefficients. You're working with a matrix and applying row operations.
Either works equally well when: Both variables have coefficient 1 in both equations. The system is simple with small integer coefficients.
Rule of thumb: Scan both equations for a coefficient of 1 or −1. If you find one, use substitution. If none, use elimination. This single check usually identifies the faster path.
Special Cases — No Solution and Infinite Solutions
No solution (parallel lines): If applying either method yields a false statement like 0 = 5, the system is inconsistent — the lines are parallel and never intersect. Example: x + y = 3 and x + y = 7. Subtract: 0 = −4. False → no solution.
Infinitely many solutions (coincident lines): If the method yields a true statement like 0 = 0, the equations represent the same line — any point on it is a solution. Express the solution as a parametric set: let x = t, then y = 3 − t. Example: 2x + 4y = 8 and x + 2y = 4. Multiply second by 2: 2x + 4y = 8. Subtract from first: 0 = 0. Infinitely many solutions.
Checking consistency before solving: Write both equations in y = mx + b form. Same slope, different intercepts → parallel (no solution). Same slope, same intercept → coincident (infinite solutions). Different slopes → one solution.
Three-Variable Systems
Both methods extend to three variables, but the workflow becomes more involved. Elimination is generally preferred for 3×3 systems because it mirrors Gaussian row reduction — a systematic process that scales to any size.
Elimination approach for 3 variables: (1) Use two pairs of equations to eliminate the same variable, creating a 2×2 system. (2) Solve the 2×2 system. (3) Back-substitute to find the third variable.
Substitution approach for 3 variables: Solve one equation for one variable, substitute into both other equations to get a 2×2 system, then solve. Works if one equation is already simple (e.g., z = 3).
Matrix shortcut: For 3×3 or larger systems, write as an augmented matrix [A|b] and apply row reduction. This is essentially systematic elimination — the matrix form makes it faster and less error-prone for larger systems.
Verdict
Substitution is faster when one variable is isolated or has coefficient ±1. Elimination is faster when equations are in standard form with matching or easily matched coefficients. For non-linear systems, substitution is often the only option. Both methods always produce the same result.
- ✓Scan for a coefficient of 1 or −1 first. If found, substitution avoids fractions. If not found, elimination is usually cleaner.
- ✓Elimination mirrors matrix row reduction — it scales to larger systems more naturally than substitution.
- ✓For quadratic-linear systems (line intersecting parabola), substitution is the standard approach.
- ✓Both methods reveal inconsistent systems (no solution) and dependent systems (infinite solutions) through contradiction (0 = k ≠ 0) or tautology (0 = 0).
- ✓In practice, many students use substitution by default — but developing fluency with elimination prepares you for linear algebra and matrix methods.