What Are the Most Common Mistakes in IB Maths Exams?

Verdict: the four repeat offenders are premature rounding, calculator mode errors, missing method marks from unclear working, and misreading command terms such as 'hence' versus 'hence or otherwise'. According to the IB, the current Mathematics AA and AI guides were first examined in 2021, and examiner reports since then flag these same issues every session.

Because IB mark schemes award marks separately — M1 for method, A1 for accuracy, R1 for reasoning — so a right approach with a rounding slip, a missing constant, or an unjustified final line still drops the A1 or R1. I've seen this cost students a full grade boundary every November session.

The single biggest calculator mistake is leaving the GDC in the wrong angle mode — radians instead of degrees, or vice versa — which silently wrecks every trig, vector and complex-number answer on the page. A close second is not clearing statistics or finance-app memory between questions, so an old data set contaminates a new calculation.

Yes — IB mark schemes are built around method marks, so a correct final answer with no working can score as little as one mark out of four or five, especially on 'show that' questions where the answer is already given and only your reasoning earns credit. Always write every step.

Verdict: most 'silly' mistakes are really rushed-reading mistakes — misreading a negative sign, skipping a domain restriction, or copying a number wrong from one line to the next. Slow down on the transcription step between your working and your answer box; that's where marks vanish most often in mock scripts I've marked.

Very strict — unless a question states otherwise, the IB requires final numerical answers to be exact or given to three significant figures, and examiners apply an accuracy penalty (AP), typically deducting one mark once per paper, for early rounding that changes the final result. Round only your final answer, never intermediate values.

With ten minutes left, check three things in this order: does your answer make sense in context (a probability over 1, a negative length), does it match the number of significant figures asked for, and does it actually answer the question asked, not a sub-part along the way. That order catches the highest-mark errors first.

Yes — Maths AI students most often mis-apply a formula from the booklet without checking it fits the real-world context, such as using linear regression on clearly non-linear data. Maths AA students more often lose marks on algebraic manipulation and formal proof steps, particularly in complex numbers and calculus.

Paper 3 exists only for Mathematics AA HL and is a 20-mark, structured investigation. The most common mistake is treating each part as separate rather than building one conjecture across the whole paper — students who don't test a pattern with a second example lose the reasoning marks even when earlier working is correct.

They matter more than most students think: 'show that' requires you to work toward a given answer with full algebraic justification, 'hence' means you must use the previous result (a different valid method scores zero), and 'hence or otherwise' allows any valid method. Mixing these up is one of the most common ways to lose method marks.

Not really — the mistakes just move. Maths AI questions are shorter but context-heavy, so students lose marks from misreading a real-world scenario or misapplying a formula-booklet result rather than from algebra, while Maths AA students lose marks on manipulation and proof. Choose based on which mistake you'd rather train out, not which subject is 'easier'.

Yes — at SL, most lost marks come from basic algebra slips and calculator misuse on shorter questions, while at HL the mistakes shift toward depth: incomplete proof reasoning, weak justification in Paper 3 (AA HL only), and rushing multi-step calculus or complex-number questions under time pressure. HL rewards patience over speed.

Keep an error log — after every mock, write down the exact mistake, not just 'lost marks', tag it as method, accuracy, or reasoning, and revisit that specific skill before the next paper. I've watched students climb a full grade boundary in one term just from tracking three recurring error types honestly.

Past papers alone rarely fix mistakes, because students mark themselves generously and don't examine why a mark was lost. Past papers work best paired with the actual mark scheme and topic-specific worksheets that isolate the skill you keep dropping marks on, rather than repeating a full paper you already half-remember.