How to perform in the new G2 SEC Mathematics examination becomes easier when the learner can reason in both directions. Some problems are transparent when solved forward from the given information. Others become clearer when the learner starts from the target, asks what must be true immediately before it, and works backwards until the known information is reached.
This forty-seventh Learner’s Guide focuses on backsolving and reverse reasoning. The central rule is: when the forward route is hidden, inspect the target and undo the structure. Reverse reasoning is not guessing the answer. It is using inverse operations, constraints and target conditions to reveal a valid forward solution.
For 2027, SEAB lists G2 Mathematics as K210. Use the SEAB 2027 G2 syllabus directory for current official details. The examples below are original eduKateSengkang learning material.
Forward solving and backward reasoning
Forward solving begins with what is known and applies operations until the target appears. Backward reasoning begins with the target and asks what previous quantity or condition would produce it.
Strong learners can switch between both without confusing the direction of the final proof or calculation.
Inverse operations are the foundation
- addition ↔ subtraction;
- multiplication ↔ division;
- squaring ↔ square roots, with domain care;
- percentage increase ↔ division by the growth multiplier;
- percentage decrease ↔ division by the remaining multiplier;
- factorisation ↔ expansion.
Backsolving often works because the final operation can be undone.
Backsolving versus checking
Backsolving can be a solving method or a checking method.
If you already found x = 7, substituting it into the original equation is a reverse check. If you do not know how to start, working backward from the required final value may reveal the equation itself.
Backsolving a simple equation
Suppose 3x + 5 = 26. Forward solving subtracts 5 and divides by 3. Backward reasoning asks: what number becomes 26 after 5 is added? 21. What becomes 21 after multiplication by 3? 7.
The route is mathematically the same structure viewed in reverse.
Backsolving a percentage
A price is $72 after a 20% discount. The final price is 80% of the original. To move backward, divide by 0.8: original = $90.
This is stronger than the vague rule “divide for original”. The learner knows which multiplier is being undone.
Backsolving a rate
A journey takes 2 hours at constant speed and covers 90 km. If speed is the target, divide distance by time. If time is the target and speed is known, divide distance by speed.
Reverse reasoning helps the learner see the same relationship from different targets.
Backsolving from a target total
A club needs $500 after receiving a 25% subsidy on the original cost. If the subsidy means the club pays 75%, the original cost must be 500 ÷ 0.75.
The target total reveals what fraction of the original remains.
Work backward from the final quantity
In multi-step word problems, write the last requested quantity and ask what input is needed to calculate it.
If final profit needs revenue and cost, then those become subtargets. If revenue needs price and quantity, continue backward until all needed quantities connect to given information.
The dependency chain
Long questions can be represented as a chain: given → intermediate A → intermediate B → target.
Backsolving reads that chain from right to left to discover what must be found first.
Backsolving in geometry
If an area is required, ask what dimensions are needed. If a missing dimension depends on Pythagoras or similarity, that becomes the earlier subtask.
The target tells the learner which geometric facts are relevant and which are distractions.
Backsolving in trigonometry
If a height is the target and the angle and horizontal distance are known, the target suggests a trigonometric relationship involving those quantities.
The learner should still label the triangle and choose the ratio correctly. Backward reasoning reveals what must connect; it does not replace geometric understanding.
Backsolving in graphs
If the question asks for the x-value where y reaches a target, begin at the target y-value and read across to the graph before reading down to x.
The reading direction follows the target.
Backsolving in tables
If a table contains a desired output, locate the row or input producing it.
If the exact target is absent, interpolation or algebra may be needed depending on the task and relationship.
Backsolving from a constraint
A budget problem may ask for the maximum whole number of items affordable.
Work backward from the budget: subtract fixed fees, divide by unit cost, then apply the whole-number constraint.
Backsolving from a minimum
A capacity problem may ask how many units are needed to meet at least a target.
Divide target demand by unit capacity, then interpret the result using the minimum whole-number requirement.
Candidate testing
Sometimes a problem offers a small set of possible answers. Testing candidates can be efficient when the conditions are clear.
A candidate is valid only if it satisfies all original constraints, not just one equation.
Do not confuse candidate testing with random guessing
Random trial has no structure. Candidate testing uses the problem conditions to reject or confirm values efficiently.
The learner should know why each candidate is being tested.
Testing an integer solution
If x must be a positive integer and the equation or inequality narrows x to a small range, testing the few remaining integers can be reasonable.
State the domain so the search is mathematically justified.
Testing a multiple-choice answer
In a Mathematics MCQ, substitute the options into the original condition if that is faster than solving symbolically.
This can be especially useful when the options are simple and the direct algebra is long. The method is valid because the question asks which option satisfies the condition.
Work backward from units
Units can reveal the required operation. If the target is km/h and the givens are kilometres and hours, division is structurally indicated.
If the target is area, the final representation must produce squared units.
Work backward from answer form
If the answer must be a percentage, probability, whole number or length, the final step must produce that form.
This can help the learner detect when an intermediate value has been mistaken for the target.
Backsolving from a graph feature
If the target is gradient, identify two points or the required change in y and x. If the target is intercept, locate where the graph crosses the relevant axis.
The feature tells the learner what evidence is needed.
Backsolving in simultaneous relationships
If a problem contains two unknowns and two independent relationships, the target may suggest which variable to eliminate or substitute first.
Reverse reasoning can make the algebra route more efficient without changing the mathematics.
Backsolving and representation switching
Use Vol 0031. Working backward becomes easier when the target is represented clearly as an equation, diagram, graph or table.
The target-first annotation
Before solving a long problem, write one short note beside the target: “need total time”, “need original value”, “need angle”, “need feasible maximum”.
This keeps the direction visible through intermediate calculations.
The what-before-this question
Ask repeatedly: what quantity must I know immediately before I can calculate this target?
Continue until the answer is a given value or a quantity that can be found directly from the givens.
The backwards plan
Write the plan in reverse, then solve it forward.
This is often safer than performing the arithmetic in reverse because the final written solution remains logically ordered from givens to conclusion.
Backwards planning versus backwards calculation
Backward planning identifies the route. Forward calculation executes the route.
The learner can reason from the target to discover subtargets, then present working in a clear forward chain.
Why this helps long problems
Long problems become difficult when too many intermediate quantities compete for attention.
Backwards planning filters the information: only quantities that feed the target remain central.
The irrelevant-information test
If a given value does not appear anywhere in the backward dependency chain, ask whether it is truly needed.
Do not discard it automatically; verify that no hidden condition uses it. But target-first planning can expose decorative or secondary data.
Backsolving and constraints
Use Vol 0035. Candidate answers found by reverse reasoning must still satisfy the original domain and practical conditions.
Backsolving and checking
Use Vol 0019. A reverse route is powerful because it can fail differently from the forward method.
Backsolving and error containment
Use Vol 0027. A dependency chain makes it easier to see which later quantities must be recomputed after one intermediate value changes.
Worked example one: reverse percentage
A membership fee is $102 after a 15% discount. The final value represents 85% of the original. Backsolving divides by 0.85 to obtain $120.
The reverse route is transparent because the last operation was multiplication by 0.85. Undo that operation.
Worked example two: fixed fee plus unit cost
A service costs $39 for a fixed booking fee plus five identical units. The fixed fee is $9. What is the unit cost?
Work backward from $39: remove the fixed $9 to get $30, then divide by five. Unit cost is $6.
Worked example three: multi-stage discount
An item is reduced by 10% and then by a further 20%, ending at $72. What was the original price?
The final multiplier is 0.9 × 0.8 = 0.72. Reverse by dividing 72 by 0.72 to obtain $100.
Do not reverse with +10% then +20%; inverse percentage changes are not obtained by adding the same percentage back.
Worked example four: average requirement
A student has three test scores of 62, 71 and 75 and wants a four-test average of at least 70. What minimum fourth score is required?
Work backward from the target total: 4 × 70 = 280. Current total is 208. Required score is at least 72.
The target average becomes a target sum, making the unknown easy to isolate.
Worked example five: target mean with unequal groups
Two groups together must have mean 15. Group A has eight values with mean 12. Group B has four values with unknown mean b.
Target total is 12 × 15 = 180. Group A contributes 96, so Group B must contribute 84. Its mean is 84 ÷ 4 = 21.
Worked example six: geometry target
A rectangle has area 96 cm² and length 12 cm. The width is needed for a later perimeter calculation.
Backward planning from perimeter says width is required. Area gives width = 96 ÷ 12 = 8 cm. Then perimeter = 2(12 + 8) = 40 cm.
The target tells you which intermediate quantity must be found first.
Worked example seven: Pythagorean backsolving
A right triangle has hypotenuse 13 cm and one shorter side 5 cm. Find the other side.
The target side is hidden inside a² + 5² = 13². Rearranging gives a² = 144, so a = 12 cm in this length context.
The negative algebraic root is excluded because a physical side length is positive.
Worked example eight: reverse scale
A map distance is 3.5 cm and actual distance is 700 m. Find the scale relationship.
Convert 700 m to 70,000 cm. Then 3.5 cm corresponds to 70,000 cm, so 1 cm corresponds to 20,000 cm. Scale is 1:20,000.
The reverse problem asks for the mapping rule instead of applying a given scale.
Worked example nine: probability target
A bag contains red and blue counters. Probability of red is 3/5 and there are 12 red counters. How many counters are there altogether?
If 12 is three-fifths of the total, total = 12 ÷ (3/5) = 20.
The target is the whole; the known quantity is a fraction of it.
Worked example ten: capacity target
A hall must seat at least 185 people. Each row contains 12 seats. What is the minimum number of full rows required?
185 ÷ 12 = 15.416…. Reverse reasoning finds the threshold; feasibility rounds upward to 16 rows.
The last step belongs to constraint interpretation, not ordinary nearest-number rounding.
Backsolving with equations versus mental undoing
Simple reverse chains can be handled mentally. Longer ones are safer as equations.
The learner should not treat mental backsolving as a replacement for written mathematical communication when essential working is needed.
When to define a variable
If several quantities or repeated operations are involved, define the unknown and formalise the relationship.
Backwards intuition can discover the equation; algebra can then execute it reliably.
When backsolving is efficient
- final value after percentage changes;
- target average;
- fixed fee plus variable cost;
- minimum or maximum under constraints;
- small candidate set;
- multi-step geometry where the target reveals a required intermediate;
- equations built from reversible operations.
When backsolving is less useful
Some problems are easier forward because the givens already form a direct route.
Do not force reverse reasoning when it creates more complexity than it removes.
The one-step-back rule
When stuck, move back only one dependency at a time. Ask what quantity is needed immediately before the target.
Avoid mentally reversing an entire long problem at once.
The backward-tree method
For complex tasks, draw a small tree with the target at the top and required inputs below it.
Continue until each branch ends in a given value or an easily computed quantity.
The forward-execution rule
After constructing the tree backward, calculate from the leaves upward.
This keeps numerical work in a stable forward order even though planning began at the target.
Backsolving and option testing
If answer choices are available, substitute candidates into the original relationship.
Start with values that are easiest to test or use order information to reduce the search.
The monotonic-search idea
If a quantity increases steadily with the variable, one failed candidate can tell you which direction to move.
Use this only when the relationship is actually monotonic over the allowed range.
Backsolving and inequalities
A target such as “at least $500” becomes an inequality rather than an equation.
Reverse operations must preserve inequality logic, including direction changes when multiplying or dividing by negative quantities.
Backsolving and domains
Candidate values must remain inside the domain.
An algebraic reverse step can produce impossible values; test them against the original problem before accepting them.
Backsolving and rounding
Do not round intermediate reverse calculations unless the question instructs it or the context requires a discrete interpretation.
A rounded intermediate can move the final candidate across a constraint boundary.
Backsolving and approximation
If the final target is approximate, keep sufficient precision while reversing the chain.
The final answer can then be rounded according to the task.
The reverse-check loop
- Solve forward.
- Reverse from the final answer.
- Recover the starting information.
- If it does not match, locate the first mismatch.
This is powerful because the checking route runs in the opposite direction.
The target-first drill
Take ten word problems and do not solve them. Write only the target and the quantity immediately needed before it.
This trains dependency recognition without arithmetic.
The inverse-operation drill
Give chains such as “start → ×1.2 → −5 → final”. Supply the final value and work backward.
Then reconstruct the same chain forward to verify.
The backsolving-versus-forward drill
Solve the same problem both ways and compare efficiency.
The learner should understand that method choice depends on structure, not on loyalty to one style.
The candidate-testing drill
Use problems with a small integer domain or multiple-choice options. Require the learner to state why candidate testing is efficient before using it.
This prevents random trial from being mistaken for strategy.
The dependency-tree drill
Use one long Paper 2-style scenario. Build the target tree before calculating.
After solving, compare the tree with the actual solution route and remove unnecessary branches.
The backsolving error ledger
- inverse operation chosen incorrectly;
- percentage multiplier reversed wrongly;
- constraint forgotten after candidate found;
- negative root kept in invalid context;
- intermediate rounded too early;
- target misidentified;
- candidate tested against only one condition;
- backward plan correct but forward execution inconsistent.
These categories show whether the issue is inverse reasoning, algebra, or interpretation.
The four-week reverse-reasoning build
Week 1 — inverse chains
Use equations, percentages and fixed-plus-variable situations.
Week 2 — target planning
Use geometry, averages and multi-stage word problems.
Week 3 — constraints and candidates
Use integer domains, budgets, capacity and multiple-choice testing.
Week 4 — timed mixed transfer
Place reverse-friendly and forward-friendly problems together. The learner must choose the efficient direction.
Use method selection
Use Vol 0015. Backsolving is one possible method, not a universal default.
Use boundary and counterexample reasoning
Use Vol 0043 when a reverse-derived candidate needs testing at boundaries or against a broad claim.
Use the Mathematics index
If reverse reasoning exposes a concept gap, return to the Complete Mathematics Index.
The PSLE bridge
The earlier rule Represent Before You Calculate still applies. The target-first plan is another representation of the problem structure.
Use Examination Craft
For time decisions and checking, continue through the Examination Craft hub. Backsolving should simplify the route, not become an extra detour.
Final rule
When forward solving becomes opaque, look at the target and ask what must be true immediately before it.
Undo reversible operations, build the dependency chain, test candidate values against all original conditions, then execute the solution clearly. Strong Mathematics can move in both directions while keeping the logic consistent.
Backsolving clinic one: target score
A learner needs an average of 68 across five tests. The first four scores total 261. The target total is 5 × 68 = 340, so the fifth score must be at least 79.
The backward step converts an average requirement into a total requirement.
Backsolving clinic two: target profit
A stall needs $240 profit. Fixed costs are $60 and each item contributes $6 after variable cost. The number of items must produce $300 contribution before fixed cost, so 300 ÷ 6 = 50 items.
The target tells the learner to restore the cost before dividing by per-item contribution.
Backsolving clinic three: target volume
A container must hold 900 cm³ and has base area 75 cm². Height = 900 ÷ 75 = 12 cm.
The target volume reveals which missing dimension is needed.
Backsolving clinic four: target area after scaling
A similar shape must have area four times the original. Since area scales with the square of the linear factor, the required linear factor is √4 = 2.
The target relationship is area; reverse reasoning takes the square root to recover the length factor.
Backsolving clinic five: target probability
A probability of success must be 0.6 and there are 18 favourable outcomes in an equally likely model. Total outcomes = 18 ÷ 0.6 = 30.
The known favourable count is a fraction of the unknown whole.
Backsolving clinic six: target gradient
A line must have gradient 3 and passes through a point. To construct another point, choose a convenient change in x and work backward from gradient = change in y / change in x.
If change in x = 2, change in y must be 6.
Backsolving clinic seven: target cost under budget
A budget is $150, fixed fee $18 and each unit costs $11. After removing the fixed fee, $132 remains. Maximum whole number of units is 12.
The constraint then confirms 12 × 11 + 18 = $150 exactly.
Backsolving clinic eight: target percentage
A population should increase to 156 from 120. Increase is 36; percentage increase is 36/120 × 100% = 30%.
Working backward from the final total to the change clarifies the correct base.
Backsolving clinic nine: target time
A trip must cover 84 km at an average 42 km/h. Required travel time is 2 hours.
If there is a 20-minute stop inside the total allowed time, the movement time and overall elapsed time become different quantities. Target wording matters.
Backsolving clinic ten: target integer
An inequality gives n ≥ 7.2 and n counts boxes. The minimum whole-number solution is 8.
Backward algebra finds the threshold; domain interpretation gives the final count.
Reverse reasoning with multiple branches
Some targets depend on two separate intermediate quantities. For example, final cost may require both usage charge and tax.
Build two backward branches, solve each from known data, then combine them forward.
Reverse reasoning with hidden givens
A question may give information in a graph, table or diagram rather than prose.
The target-first plan should point the learner to the representation containing the needed input.
Reverse reasoning and algebraic manipulation
The learner should understand inverse structure rather than memorise “move across, change sign”.
Backsolving is stronger when each reverse operation can be explained as undoing the forward operation.
Reverse reasoning and function machines
A function-machine representation is useful for chains such as ×3, then +5.
To recover the input from the output, reverse the order and apply inverse operations: −5, then ÷3.
Reverse order matters
Undo operations in the reverse order from which they were applied.
If a value is multiplied by 3 and then increased by 5, subtract 5 before dividing by 3.
The reverse-order error
A common mistake is to invert each operation but keep the original order.
This generally fails because composition of operations is order-sensitive.
Backsolving and brackets
Expressions with brackets make operation order explicit.
If y = 4(x + 2), reverse from y by dividing by 4 first, then subtracting 2.
Backsolving and powers
If y = x² in a domain allowing both signs, reversing gives x = ±√y.
The domain or context decides whether both values remain valid.
Backsolving and absolute constraints
A reverse-derived value may satisfy the equation but violate a contextual condition.
Always finish by testing domain and feasibility.
The reverse-planning worksheet
- Target quantity.
- Immediate inputs needed.
- Where each input comes from.
- Operation linking input to target.
- Constraint on final value.
- Independent check.
This structure is especially useful during early training on long word problems.
The backsolving decision question
Ask: would the target reveal the route more quickly than the givens do?
If yes, plan backward. If no, solve forward.
The two-direction check
A strong solution can often be verified both ways.
Forward work produces the answer; reverse work reconstructs the starting information. Agreement increases confidence.
The reverse-reasoning error ledger
- wrong final target identified;
- inverse operation incorrect;
- operations undone in wrong order;
- domain restriction forgotten;
- constraint applied before algebra complete;
- candidate satisfies derived equation but not original condition;
- backward plan never converted into clear forward working.
These errors show where reverse reasoning needs repair.
The 15-minute backsolving session
Use five short inverse-operation chains, three word problems and one multi-step target tree.
For each, state whether backward planning or forward solving is more efficient.
The target-switch drill
Use the same data but ask for a different quantity each time.
For example, distance, speed and time can all become targets from the same relationship. The learner practises reorganising the equation around the target.
The answer-choice drill
Provide four candidate answers and one original condition. Test candidates efficiently, but require the learner to state why candidate testing is valid.
This keeps substitution strategic rather than random.
The constraint-backsolve drill
Start from a budget, capacity or minimum requirement and work backward to the largest or smallest allowed input.
Then test the neighbouring integer to verify the boundary.
The geometry target drill
Give a composite shape and ask for one final measure. Before calculating, list every intermediate dimension actually needed.
The target filters irrelevant geometry.
The advanced standard
An advanced learner can use the target as information. They do not stare only at the givens.
They can plan backward, execute forward, test candidates, undo operations in the correct order and finish by checking the original conditions.
Final perspective
Backsolving is a way of seeing structure from the other end.
The target is not merely where the solution finishes; it is a clue to the route. Ask what must be true immediately before the target, continue until the chain reaches the givens, then calculate clearly forward. When both directions agree, the solution is easier to trust.
The final reverse-reasoning calibration
Backsolving should be evaluated by whether it makes the structure clearer. If the learner reaches the target quickly but cannot explain why each inverse step is valid, the method has become a shortcut without understanding.
During review, ask the learner to write both directions: the forward operation chain and the reverse operation chain. The order should reverse as the operations invert. If the two chains do not reconstruct each other, locate the first mismatch.
In longer problems, the best evidence of mastery is flexible choice. Some questions should be planned backward and calculated forward. Others should be solved directly from the givens. A mature learner can explain why one direction is more efficient here without turning that choice into a universal rule.
Finally, every reverse-derived candidate returns to the original conditions. Substitute it, check the domain, inspect the units and apply any practical constraint. Backsolving finds a route; verification decides whether the route has actually reached the right destination.
Backsolving under examination pressure
Under time, reverse reasoning should simplify rather than expand the solution. If a target-first plan produces a clear dependency chain in thirty seconds, use it. If it creates more notation than the forward route, return to the givens.
A useful timed habit is to write the final target and one immediate prerequisite only. For example, “need original price → need remaining percentage” or “need perimeter → need missing side”. Once the first prerequisite is identified, continue only if the route remains clearer than forward solving.
Backsolving is also a strong late-paper recovery tool because it can reveal which part of a long question still matters. If several intermediate values have already been calculated, work backward from the unfinished target and reuse only the values that belong to its dependency chain.
In checking, use the reverse direction independently. A correct forward route and a correct reverse reconstruction should meet at the same quantities. If they do not, the mismatch provides evidence about where to inspect.
The final standard is flexible direction. Strong learners are not committed to forward or backward solving. They choose the direction that exposes structure fastest, present the final working clearly, and verify that every reverse-derived candidate still satisfies the original question.
One final backsolving rule
When a reverse plan reaches a quantity that is not given directly, stop and ask whether that quantity can itself be obtained from another relationship. This prevents the learner from treating an intermediate target as though it were known.
Continue the dependency chain only until every branch ends at genuine givens or previously established results. Then execute forward. This separation between planning and calculation keeps reverse reasoning from becoming guesswork and makes long solutions easier to audit.
Backsolving is strongest when the learner can explain the reverse step and then verify it by running the relationship forward. That two-direction test turns a convenient shortcut into a reliable mathematical method.
If the reverse route produces several candidates, keep all mathematically possible values until the original domain and practical conditions decide which survive.
The final reverse-reasoning check is simple: if you work backward from the target and forward from the givens, the two routes should meet at the same intermediate quantities. When they do, the structure is internally consistent.
