A large share of A-Level Physics marks are lost by students who understand the physics. They go to assumptions left unstated, to signs and directions handled loosely, and to mathematics that is correct but not fluent enough to be reliable under time pressure. These are all fixable, and none of them are fixed by re-reading the content.
Working out which is costing you is the first useful thing to do.
Why the step up from O-Level is steep
Mathematics stops being incidental. Calculus, vectors, exponentials, logarithms and graph analysis are the working language of the subject. A student who is competent but slow at the maths loses time and accuracy on questions that are not testing maths at all — and it reads as a physics weakness.
Models replace descriptions. O-Level physics can be handled by knowing what happens. A-Level asks you to model a situation: decide what to neglect, state the assumptions, apply the relationship, and say what the result means physically.
Topics stop being separate. A single question may involve mechanics, energy and a graph interpretation. Weakness anywhere in the chain fails the whole question.
Precision is marked. Directions, signs, units, significant figures. Loose handling that passed at O-Level is penalised here — which students often experience as pedantry and is actually the subject testing whether the model was applied correctly.
The four places marks actually go
Take a marked paper and sort every lost mark. The proportions matter more than the total.
1. Unstated assumptions and reasoning. The answer is right; the reasoning that earns the marks is not on the page. Or the question said “explain” and received a description. This is usually the largest pile and it responds to strict marking faster than to more teaching.
2. Vectors, signs and directions. Getting the magnitude right and the direction wrong, losing a sign in a substitution, treating a vector quantity as scalar. Persistent and specific, and it usually traces back to a shaky treatment of vectors early on.
3. Mathematical execution. Rearranging, handling exponentials and logs, reading gradients and intercepts properly. Not a physics gap, and more physics teaching will not touch it.
4. Not knowing the physics. A genuine content gap. Real, and smaller than most students assume.
A student whose losses sit in piles one to three has been buying the wrong thing if they have been paying for more content coverage.
What each pile needs
For unstated reasoning: write full answers to “explain” and “show that” questions and have them marked hard against the criteria. State assumptions explicitly even when they feel obvious — that is frequently where the mark is. This is uncomfortable and it is the highest-return activity available.
For vectors and signs: go back and rebuild vector handling properly, including sign conventions, and then apply it deliberately across mechanics, fields and circuits. It feels like going backwards while the course moves on, which is why it is rarely done and why it keeps costing marks.
For mathematical execution: targeted maths practice, not physics practice. Fluency is the goal — the operations need to be automatic so attention is free for the physics.
For genuine content gaps: teaching, which is what tuition is best at, and which is the smallest category for most students.
The topics that reliably cause trouble
Without asserting any particular weighting, the areas where difficulty concentrates:
Mechanics, because everything later assumes it and because it is where vector handling is established or not.
Fields — gravitational, electric, magnetic. Structurally similar to each other, which helps students who see the pattern and confuses those who learn each separately.
Quantum and nuclear physics, where the concepts resist intuition and students fall back on memorised statements they cannot apply.
Practical work and data analysis — uncertainties, error propagation, graph linearisation. Consistently neglected until late and entirely learnable.
What “state your assumptions” actually earns
The instruction is given constantly and rarely explained, so students treat it as a formality and lose the marks attached to it.
An A-Level physics question is asking you to model a situation, and a model is defined by what it leaves out. When you neglect air resistance, treat a string as inextensible, assume a field is uniform over the region of interest, or take a resistance as constant with temperature, you are making the problem solvable — and the examiner wants to see that you know you did it.
Two distinct marks are usually available. One for identifying the assumption; one for recognising where it breaks down and what would change if it did. Students who write “assuming no air resistance” collect the first and miss the second, which is why partial credit on explanation questions so often stalls at half.
The same logic governs “show that” questions. The answer is given, so the marks are entirely in the route — every step visible, every substitution justified. A student who works backwards from the printed answer produces something that looks complete and earns very little.
Graphs are doing more work than students think
Graph questions are treated as a separate skill and are actually where several topics get tested at once.
A gradient is a physical quantity, not a number. Knowing what it represents in the particular situation — and what its units tell you — is usually the point of the question.
An intercept is often the most informative feature, because it isolates one term when another goes to zero.
Linearisation is a recurring demand. Rearranging a relationship so that plotting one thing against another gives a straight line, then identifying what the gradient and intercept correspond to. Students who can do the algebra and cannot see why they would want to lose these marks routinely.
Error bars and lines of best fit carry their own marks in practical contexts, and they connect directly to the uncertainty work that most students leave until late.
What to ask a prospective tutor
- “After a few sessions, can you tell me whether my problem is reasoning, vectors, maths or content?” A specific answer means diagnosis is happening.
- “Will you mark my written explanations strictly?” The largest pile goes unaddressed otherwise.
- “Would you spend sessions on maths fluency if that’s what’s costing me?” A tutor who only covers physics topics cannot fix a physics grade that maths is holding down.
Frequently asked questions
I understand the concepts but my marks are poor. Why? Most commonly because the reasoning is not on the page — assumptions unstated, “explain” answered with a description — or because signs and directions are handled loosely. Both respond to strict marking of written work.
Do I need to be strong at maths for A-Level Physics? You need fluency, not brilliance. Calculus, logarithms, vectors and graph analysis need to be automatic enough that they do not consume attention during a physics question.
Why is mechanics so important if it’s early in the course? Because vector handling and modelling are established there, and later topics assume both. A shaky foundation in mechanics surfaces as difficulty in fields and circuits.
Is H2 Physics much harder than O-Level? It is different rather than simply harder — modelling and mathematical fluency replace description and recall. Students who did well through memorisation feel the change most.
How quickly can marks improve? Reasoning and presentation problems can move within a term because the fix is technique. Vector and mathematical fluency gaps take longer, because something is being rebuilt while the course continues.
—