The short version
Each of the four sections — Reading, Listening, Speaking, Writing — gets its own score on a 1.0–6.0 scale, in half-point steps. Your overall score is simply the average of the four section scores, rounded to the nearest half band. There's no separate "total points" system like the old 0–120 scale — the new scale is the section average, full stop.
Step 1 — Reading and Listening: objective, percentage-based
These two sections are auto-scored against fixed answer keys — no human or AI judgment involved. Your raw score (number correct ÷ total items) converts to a 1–6 band through a percentage-to-band table. Get roughly the top tier of questions right, and you land in the 5.5–6.0 range; a middling percentage lands you in the 3.5–4.5 range, and so on down the scale. This is exactly how a multiple-choice test has always been scored — the new part is just reporting it on 1–6 instead of a raw point total.
Step 2 — Speaking and Writing: rubric-graded
These sections involve actual production (spoken or written responses), so they're graded against an official rubric rather than a fixed answer key. Each task is scored on specific criteria: Speaking is judged on delivery (pace, clarity, pronunciation), language use (grammar, vocabulary), and topic development (did you actually answer the question with real content); Writing is judged on development, organization, and language use. Each criterion is scored, and the task's overall band comes from combining them. If a section has more than one task (like Speaking's two tasks, or Writing's three), the section score is the average of its task bands.
Step 3 — combining into an overall score
Once all four sections have a band (1.0–6.0 each), your overall score is their average, rounded to the nearest 0.5. So a candidate scoring Reading 5.0, Listening 4.5, Speaking 4.0, and Writing 4.5 gets an overall of 4.5 (the average of 5.0+4.5+4.0+4.5=18, divided by 4 = 4.5). There's no weighting — all four sections count equally toward the overall.
Why this matters for how you prepare
Because the overall is a simple average, your weakest section drags the total down the most relative to how much room it has to improve — a section already at 5.5 has little room to add to your average, while a section at 3.5 has a lot. If your goal is to move your overall band efficiently, time spent on your weakest section usually returns more than the same time spent polishing your strongest one.
What "adaptive" changes (and doesn't)
Reading and Listening use a multistage adaptive format — the specific items and difficulty can shift based on your performance as you go. This affects which questions you see, not how the resulting percentage converts to a band; the scoring math in Step 1 above is unchanged by adaptivity. See our full adaptive format explainer for details.
What each band actually represents
For the full breakdown of what 1.0 through 6.0 mean in terms of real ability (and their CEFR equivalents), see our band scores explained guide. For the old 0–120 equivalent, see our score conversion chart.
Find your current section-by-section breakdown
A free diagnostic gives you an estimated band per section, not just an overall number — so you can see exactly where the averaging is working for or against you.