What a mechanical reasoning test actually measures
Mechanical reasoning is the ability to look at a machine you have never seen before and predict what it will do. It sits somewhere between spatial reasoning and applied physics: you are not being asked to recall a formula, and you are not being asked to rotate an abstract cube in your head either. You are being asked whether you can trace a force or a motion through a chain of parts and arrive at the right end state. That skill is remarkably stable across people and it transfers directly to real equipment, which is exactly why it has survived a century of employment testing while fashions in psychometrics came and went.
The eighteen questions on this page are drawn rather than described. Every stimulus is a diagram, so the test does not reward reading speed, and the wording of each question is kept deliberately plain. What varies is the mechanism: meshed gear trains, open and crossed belts, chain drives, first and second class levers, fixed and movable pulleys, block and tackle arrangements, wheel and axle systems, a loaded beam and an inclined plane. Each one has exactly one defensible answer that follows from ordinary physics.
The two rules that cover almost every question
Nearly every item on a mechanical reasoning test reduces to one of two ideas, and it is worth fixing both in your mind before you start. The first is about direction and speed of rotation. Two gears that mesh tooth to tooth always turn in opposite directions, so a train of gears alternates: the first and third turn together, the second and fourth turn against them. A belt running straight between two pulleys, or a chain between two sprockets, keeps both wheels turning the same way; a belt crossed into a figure eight reverses the second wheel. Speed then follows from size. A small gear driving a large one turns it slowly and powerfully; a large gear driving a small one spins it quickly and weakly. The ratio is simply the tooth count of the driver divided by the tooth count of the driven wheel, and an idler gear sitting between two others changes the direction but never the ratio.
The second idea is the law of the lever, and it covers everything with a pivot: seesaws, balance beams, crowbars, wheelbarrows, nutcrackers and the arm of a crane. Multiply each force by its distance from the pivot. If the two products are equal, the system balances; if one is larger, that side goes down. The same accounting explains why a longer handle makes a job easier: increasing the distance on the effort side lets a smaller force match the same load. Pulleys are the rope version of the same bargain. Count the rope segments that actually support the moving load — with two supporting segments you pull half the weight over twice the distance, with four you pull a quarter of it over four times the distance. A single fixed pulley supports nothing at all; it only changes the direction of your pull, which is convenient but never easier.
How the questions are built
Every diagram is generated from the geometry of the mechanism rather than copied from a question bank, and every option is checked against the physics before it is keyed. The wrong answers are chosen to be the mistakes people actually make: reversing the direction of a crossed belt, inverting a gear ratio, counting the rope segments of a pulley block wrong, or forgetting that a fixed pulley gives no mechanical advantage. That means you cannot reach the right answer by eliminating obviously silly options, and it also means a wrong answer tells you something specific about which rule slipped.
How to read your score
Your raw score out of eighteen is converted to the familiar hundred-point scale where the population average is 100 and fifteen points is one standard deviation. Because the test is short, treat the number as the centre of a band roughly ten points wide rather than as a precise value.
| Correct | Score | What it suggests |
|---|---|---|
| 17–18 | 135+ | Exceptional; comfortably above the bar for engineering entry tests |
| 15–16 | 122–131 | Strong; the range technical employers screen for |
| 12–14 | 105–118 | Above average; solid practical mechanical sense |
| 9–11 | 90–101 | Average for adults without technical training |
| 6–8 | 75–86 | Below average; the core rules are worth revisiting |
Getting a better result on the real thing
Mechanical reasoning responds to practice more than most reasoning tests do, because the rule set is small and finite. A few concrete habits help. Trace the mechanism part by part with a finger rather than trying to see the whole thing at once; in a four gear train, label each wheel with an arrow as you go. When a question involves numbers, ask yourself first whether the answer should be bigger or smaller than the input, and only then work out by how much — this single check catches most inverted ratios. On pulley questions, count only the rope segments that pull upward on the moving block, and ignore the segment in your hand if it runs downward. Finally, answer every item: nothing is deducted for a wrong answer, so an eliminated guess is always worth more than a blank.
How this differs from other tests on the site
A culture fair matrix test measures abstract pattern reasoning with no real-world content at all, and a spatial reasoning test asks you to manipulate shapes in your head. Mechanical reasoning overlaps with both but adds a layer of naive physics: an intuition for how force, motion and leverage move through a structure. People with hands-on experience — cyclists, mechanics, sailors, anyone who has rebuilt a gearbox or rigged a hoist — usually score higher here than their abstract test results would predict, and that gap is real rather than an artefact. If you are preparing for a trade apprenticeship, an armed forces entrance exam or a maintenance technician assessment, this is the closest format on the site to what you will actually sit.
Frequently asked questions
What is a mechanical reasoning test?
It is a test of how well you predict the behaviour of simple machines. Each question shows a drawing of gears, pulleys, levers or a balance beam and asks what happens next. The physics involved never goes beyond turning direction, gear ratios and the law of the lever.
Do I need to know physics formulas to pass?
No formula sheet is required. Two ideas cover every item: meshed parts turn in opposite directions while belted or chained parts turn the same way, and force times distance from the pivot must match on both sides. Everything else follows from those two.
How many questions are there and how long do I get?
Eighteen questions with an eighteen minute limit, so roughly one minute each. Most people finish in eleven to fifteen minutes, and the test scores itself automatically if the clock runs out.
What is a good score on a mechanical reasoning test?
Twelve of eighteen correct is a solid average result for adults with no technical training. Fifteen or more places you in the range employers look for in engineering, maintenance and skilled trade roles.
Why do employers use mechanical reasoning tests?
Because the score predicts how quickly someone learns to work with real equipment. Apprenticeships, armed forces entry, maintenance technician jobs and machine operator roles all use this format as an early screening step.
Can I improve my mechanical reasoning with practice?
Yes, more than on most reasoning tests. The underlying rules are few and learnable, so people who practise gear trains, pulley systems and balance problems for a few hours usually gain several points on a real assessment.