Original data study
PTE mock test statistics: what 716 attempts reveal
We analysed 716 real, AI-scored PTE Academic mock attempts taken on PTE Mocks between 6 June to 3 August 2026. The picture is sobering, and useful: the average PTE mock score is 50 out of 90 (median 52), only 21% reach 65, just 3% reach a 79, and speaking is the wall almost half of everyone hits first.
50
Mean overall score
out of 90 — top of the Competent band
3%
Reached 79 or above
despite almost everyone targeting 79
29
Average points short of a 79 target
the typical gap left to close
The PTE score distribution: most people are in the lower half
Share of the 716 attempts falling in each overall score band. Nearly two thirds scored below 59, and only 7% reached the Strong-or-Superior range (73+).
The shape matters more than any single number. The distribution is not a bell curve centred on a comfortable pass — it is stacked toward the bottom, with a long, thin tail reaching the high scores. The median attempt scores 52, and each step up gets steeper: 21% of attempts clear 65, but only 3% clear 79. The jump from Competent to Superior is where the field thins out fastest.
Is the PTE hard? What the numbers say
Judged by these results, yes — but not for the reason most people assume. Individual tasks are learnable; the difficulty is that a strong overall score demands a strong score in every skill on the same day. Reading a 79 in reading does nothing for your total if speaking is stuck at 43. That is why the median sits at 52 and only 3% reach 79: the test rewards the person with no weak skill, and almost everyone has one.
The practical takeaway is that raising your PTE score is usually a game of lifting your single weakest skill, not polishing your strongest. Find it with a scored mock, drill it specifically, then re-measure.
Speaking is the hardest skill by a wide margin
Mean score per skill, and the share of test-takers for whom that skill is their single weakest. Speaking is both the lowest-scoring skill and the most common weakest link, for 49% of everyone.
Mean score per skill (out of 90)
| Skill | Mean | Weakest skill for | Reached 65+ | Reached 79+ |
|---|---|---|---|---|
| Speaking | 43 | 48% | 7% | 0% |
| Listening | 51 | 18% | 25% | 4% |
| Writing | 52 | 27% | 23% | 3% |
| Reading | 60 | 7% | 46% | 13% |
Per-skill figures are computed over attempts where the skill was actually tested. Only 7% of attempts reached 65 in speaking, and none reached 79 — the clearest signal in the whole dataset.
Speaking · mean 43
The lowest-scoring skill and the most common weakest link. 52% of speaking attempts land below 43, and none reached 79.
Listening · mean 51
Middle of the pack. A quarter clear 65, but only 4% reach 79 — the passages move fast and note-taking makes the difference.
Writing · mean 52
Slightly ahead of listening. The people who jump fastest here adopt a single repeatable essay and summary structure.
Reading · mean 60
The strongest skill by far. Nearly half clear 65 and 13% reach 79 — for most people reading is already carrying their overall score.
The reading–speaking gap is the clearest pattern in the data
Of every relationship in the dataset, one is almost universal: people read far better than they speak. Reading averages about 16 points above speaking, 93% of test-takers score higher in reading than in speaking, and nearly half — 48% — have a speaking score fifteen or more points below their reading score.
This is the single most useful thing to know before you plan your prep. If your reading is already clearing 60 and your speaking is stuck in the 40s, your overall score is being held down by speaking alone — and the fastest route to a higher total is almost never more reading practice. It is fixing speaking: fluency, continuous phrasing, and recording conditions.
Why speaking scores lag
Two forces pull speaking scores down. The first is real: speaking is a produced skill, scored on fluency and pronunciation, which is genuinely harder than selecting a reading answer. The second is fixable: a share of low speaking scores come from recording conditions, not ability. A weak or distant microphone, background noise, and long pauses that split a sentence into fragments all lower the score, on our mocks and on the real exam, which grades Oral Fluency the same way.
That is why we added a microphone-clarity check before every test and an honest note on the report when a recording comes through unclear. If your speaking score looks far below your other skills, check your mic before you conclude anything about your English.
What is a good PTE score — and how far is it, really?
"Good" depends entirely on your goal. Here is how the common targets map onto our data, so you can see how far each one actually sits from the typical starting point.
| Score | Level | What it unlocks | Share who reach it |
|---|---|---|---|
| 50 each | Competent | Many university entry requirements; 0 Australian PR points | ≈ half of attempts |
| 65 each | Proficient | Higher university bars; 10 Australian PR points | 21% of attempts |
| 79 each | Superior | Top university bars; 20 Australian PR points | 3% of attempts |
Migration-point thresholds are per skill on the official 79/65/50 bands. University requirements vary — check your programme.
A realistic path from 50 to 65 to 79
The data points to a clear order of operations for the average test-taker sitting around 50:
- 1Find your true weakest skill. For 48% of people it is speaking. Take one full scored mock and read the per-skill breakdown before you decide what to practise — do not guess.
- 2Fix speaking first if it is your floor. It is the lowest skill for nearly half of everyone and the biggest drag on the total. Work on continuous phrasing, pacing, and a clean recording setup. This is usually the fastest way to lift the overall score off 50.
- 3Add a repeatable writing structure. Writing is the next most common weak skill (27%). A single reliable essay and summary framework moves the score quickly because it removes the guesswork.
- 4Protect your strong reading — don't over-invest in it. Reading is already the strongest skill for most people. A little maintenance is enough; extra reading practice rarely lifts the overall score.
- 5Re-measure every 10 to 14 days. Practising blindly hides plateaus. A fresh mock every couple of weeks tells you whether the weakest skill is actually moving.
Put it into practice: run a full scored mock, then drill your weakest area with focused speaking, writing, reading, or listening section tests.
Methodology and honest caveats
- The dataset is 716 completed PTE Academic mock attempts on PTE Mocks, 6 June to 3 August 2026, each scored on the official 10 to 90 scale by our AI engine.
- Figures are anonymised aggregates. No individual scores, emails, or answers are included.
- Mock-takers are self-selected practisers, not a random sample of all PTE candidates, so absolute levels skew lower than the general population. Read the relative pattern (which skills are hardest, how far targets sit from reach), not the absolute averages, as the exam-wide truth.
- A mock score is a directional estimate to guide practice, not an official Pearson result.
- Per-skill percentages are over attempts where the skill was tested. We update these figures periodically; cite the review date shown above.
Citing this? Please attribute to PTE Mocks and link back to this page. Our methodology is in our editorial standards.
See where you land
Take a full scored mock and get your own per-skill breakdown against this data. Find your weakest skill, then drill it.
Related: PTE score chart · how to hit Superior (79) · PTE for Australian PR
Frequently asked questions
Across 716 scored PTE Academic mock attempts on PTE Mocks between 6 June to 3 August 2026, the mean overall score was 50 out of 90. That sits at the top of the Competent English band. The distribution is heavily weighted toward the lower half: 39% of attempts scored 42 or below, and only 7% reached 73 or above.
In our data, only 3% of attempts reached an overall score of 79 or higher, even though almost everyone was targeting 79 (the old flat Superior English bar and a common university requirement). The average gap between a test-taker's score and a 79 target was 29 points, which shows how demanding a 79 in every skill really is.
Speaking. It was the single weakest skill for 49% of test-takers, nearly half, and had the lowest mean of any skill at 43. Only 7% of attempts reached 65 in speaking and none reached 79. Reading was consistently the strongest skill, with a mean of 60 and 46% of attempts clearing 65.
Two reasons show up in the data. First, speaking is a produced skill scored on fluency and pronunciation, which is harder to fake than a multiple-choice reading answer. Second, a meaningful share of low speaking scores trace to recording conditions, a weak microphone, background noise, or long pauses that fragment speech, rather than ability. That is why we now run a microphone-clarity check before every test.
This analysis covers 716 completed and AI-scored PTE Academic mock attempts taken on PTE Mocks between 6 June to 3 August 2026. Every attempt was scored on the official 10 to 90 Global Scale of English by our AI scoring engine. Figures are anonymised aggregates and contain no individual data.
No. A mock score is a directional estimate to guide practice, not an official Pearson result. Mock-takers are also self-selected practisers rather than a random sample of all candidates, so absolute levels skew lower than the general test-taking population. The value is in the relative pattern: which skills people struggle with most, and how far the typical target sits from reach.
It depends on the goal. Australian skilled migration awards 0 points at Competent (around 50 in each skill), 10 points at Proficient (65 in each) and 20 points at Superior (79 in each). Most universities accept 50 to 79 overall. Our data shows the 79 Superior bar is the hardest to hit, reached by only 3% of attempts.
Start by finding your true weakest skill with a full scored mock, then drill that skill specifically. For most people that is speaking, where Read Aloud and Describe Image practice fix fluency fastest, followed by writing, where a repeatable essay structure lifts scores quickly. Re-measure with a fresh mock every 10 to 14 days rather than practising blindly.
Yes, and it is one of the most common hidden causes of a surprisingly low speaking score. A weak or distant microphone and background noise garble the audio, and the scoring engine, on the real exam and in mocks, can only score what it captures. Long pauses inside a sentence also fragment your speech and lower Oral Fluency. Use a headset in a quiet room and speak in continuous phrases.
The PTE is demanding, and the numbers show it. In our data the median mock score is 52 out of 90, only 21% of attempts reach 65 (the Proficient level worth 10 Australian PR points), and just 3% reach 79. It is not that the tasks are tricky in isolation; it is that a strong score requires 65 or 79 in every skill at once, and most people have at least one skill — usually speaking — dragging the total down.
It depends on your goal, but here is a rough map against our data. A 50 overall sits at the top of the Competent band (most people land here) and clears the lower university requirements. A 65 in every skill is Proficient — reached by 21% of attempts — and worth 10 points for Australian skilled migration. A 79 in every skill is Superior, reached by only 3% of attempts, and worth 20 PR points. So a "good" score for a university is often 50–65 overall; a "good" score for maximum migration points is a 79 in each skill, which is genuinely elite.
You are not alone: 93% of test-takers in our data score higher in reading than in speaking, and reading averages about 16 points above speaking. Nearly half (48%) have speaking 15 or more points below reading. Two reasons: speaking is a produced skill graded on fluency and pronunciation, and a share of low speaking scores come from recording conditions — a weak mic, background noise, or pauses that fragment your speech — rather than your actual English.
Yes. You are welcome to cite these figures with attribution to PTE Mocks (ptemocks.com). Please link back to this page as the source. The dataset is anonymised aggregates from 716 mock attempts and we update the figures periodically, so cite the review date shown on the page.