If you have looked up the Victorian driving test pass rate, you have probably seen a number somewhere between 75% and 90%. Both circulate widely, usually without a stated denominator.
We wanted the raw numbers, so we asked for them under Freedom of Information. The Department of Transport and Planning released them in full on 15 July 2026: 492,504 car drive tests, January 2024 to April 2026, broken down by month, transmission, result, gender, age band, and how many times the candidate had sat the test before.
The answer that falls out is 72.2%.
The more interesting part is where the rest of the number went.
What actually happens to half a million drive tests
| Outcome | Tests | Share |
|---|---|---|
| Passed | 355,364 | 72.2% |
| Completed the test but did not meet the standard | 28,138 | 5.7% |
| Test stopped before completion | 109,002 | 22.1% |
Look at the bottom two rows. Of everyone who does not walk out with a pass, only about one in five actually finishes the test and falls short on points. The other four in five never reach the end. The assessment is halted, recorded as incomplete, and a re-test is required.
That is 109,002 tests over 28 months. Roughly 3,900 a month. About 130 a day.
How the same data produces 92.7%
Divide the passes by only those tests that ran to completion, and you get 92.7%. Every stopped test vanishes from the denominator, and Victoria suddenly has one of the friendliest driving tests in the world.
Neither number is a lie. They answer different questions:
- 92.7% answers "if the testing officer let me finish, did I pass?"
- 72.2% answers "I booked a test. Did I come home with a licence?"
If you are a learner deciding whether to book, the second question is the one you are actually asking.
The same split shows up on first attempts specifically: first-time candidates pass 93.2% of completed tests and 74.2% of all tests. We cannot tell you which convention produced any particular published figure, because the published ones generally do not say. What we can say is that a headline pass rate in the low nineties for Victoria is only reachable by leaving stopped tests out.
Why so many tests get stopped
Victoria's drive test runs in two stages. Stage 1 covers quiet residential streets and includes a low-speed manoeuvre such as a reverse parallel park or a three-point turn. Stage 2 moves into real traffic: multi-lane roads, roundabouts, busier intersections. Stage 1 is commonly described as taking around ten minutes and Stage 2 around twenty, though VicRoads does not publish a fixed split.
You do not coast through Stage 1. A critical error there ends the test on the spot, and you never see Stage 2. When that happens the test is not completed and you have to book again. In the FOI dataset these tests sit in their own result category, separate from tests that ran to the end and fell short.
This lines up with the one public statement we could find on the subject. RACV, citing VicRoads figures, reports that about three-quarters of failed Victorian driving tests are due to "immediate termination errors". In our data, stopped tests make up 79.5% of all unsuccessful outcomes, and that share is stable across 2024 (78.1%), 2025 (80.3%) and the first four months of 2026 (81.3%).
In other words, the typical unsuccessful candidate in Victoria is not someone who drove for forty minutes and lost on points. It is someone who made one disqualifying mistake and was driven back to the centre.
The bit that should change how you prepare
There is a strategy buried in this.
If most failure came from accumulating minor faults, the sensible response would be broad polish: get a bit better at everything. But that is not what the data shows. Failure in Victoria is overwhelmingly binary and early. One critical error, usually in the first ten minutes, on quiet streets that most learners assume are the easy part.
So the highest-value preparation is not another lap of general practice. It is making the specific disqualifying actions impossible for you: the head check you skip when you are nervous, the give-way on a right turn, the speed you drift over in a 50 zone you assumed was 60, the manoeuvre you have only done a handful of times.
Stage 1 is where the tests die. Practise it like it counts, because statistically it counts more than everything after it.
The honest limits of this data
We should be clear about what this dataset does and does not tell you.
It does not break down by test centre. That was the core of our original FOI request, and VicRoads refused it under section 19 of the FOI Act, saying test centre level data cannot be extracted from its systems without external contractors. So we cannot tell you which Melbourne centre is easiest, and anyone quoting official per-centre pass rates for Victoria should be asked where they got them.
The result codes are not documented. The release came without a data dictionary. Our reading of the third result category as "test stopped before completion" is an inference, not an official definition. It rests on three things: the match with the publicly reported figure that about three-quarters of failed Victorian tests are immediate terminations (reported by RACV, citing VicRoads); the fact that this category exists only for drive tests and never for the computer-based hazard perception or knowledge tests; and its tracking of candidate difficulty exactly as you would expect. We think it is solid. It is still an inference, and we will correct this article if VicRoads publishes a definition that contradicts it.
It covers car drive tests. Motorcycle, heavy vehicle and re-assessment categories exist in the data but are too small to say anything useful about.
Source
492,504 Victorian car drive test records, January 2024 to April 2026, released to us by the Department of Transport and Planning under FOI (reference CS019502).