Your TOEFL score is not out of 120 any more

Four sections, one overall, all on 1–6

If you have a target written down somewhere — 100, 105, 90 for a particular programme — that number is now in a language the test no longer speaks. From January 2026 the TOEFL iBT reports four section scores and one overall score, each on a 1–6 scale. Not 0–30 per section. Not 0–120 overall. Six bands, with half-bands in between.

Run that arithmetic on yourself and something uncomfortable appears. Four sections at 5.0 gives you a 5. Three at 5.5 and one at 3.5 gives you 5.0 as well — the strong sections carry the weak one, but only just, and the effort to move three sections up half a band each is far greater than the effort to move one section up from 3.5.

A straight average rewards fixing your worst section, not polishing your best. If you came from the 0–120 world, where a 26 and a 22 both looked like “fine, roughly”, the band scale makes that gap legible in a way the old scale hid.

Your old target still means something — for about two years

ETS is publishing a 0–120 comparable score alongside the band, for a two-year transition period after January 2026. So an admissions page that still says “100 minimum” has a number to compare against, and you have a way to read your result in the scale you already understand.

Two years is not very long, and it is running now. If you are applying in this cycle, the comparable score is genuinely useful. If your test date is eighteen months out, learn the band scale and stop converting.

What nobody can tell you yet

Here is the honest limit, and you should be suspicious of any site that skips it. ETS has published the ranges — Listening is 0–35 raw points mapping to 1–6, and so on for each section. What ETS has not published is the cut scores: the raw totals where a 4 becomes a 5. Those are derived statistically and vary by test form, which is why there is no table to look up.

That means every raw-score-to-band converter you find online, including ours, is an estimate. We say so on our own reports, because a converter presented as official is worse than no converter — it invites you to grind toward a threshold that may not be where you think it is. Use band estimates to find your weakest section. Do not use them to decide you are two questions away from a 5.

The part you cannot work out from here

Everything above is arithmetic. The one thing it cannot tell you is the number that actually matters — which of your four sections is the low one.

Almost everybody guesses this wrong. People nominate the section that feels worst, and feeling worst is usually about discomfort rather than performance: Speaking feels worse than Reading because you can hear yourself doing it. The average does not care how it felt.

You need four section bands, produced under the clock, side by side. At the end of it you know which section to spend the next month on, instead of guessing.