UK Driving Test Fail Rate 2026: 51.3% Fail, And Where the Marks Go

#Why 51.3% and not 50% or 60%

The UK driving test fail rate has hovered around 50% for over a decade. The exact number moves by half a percentage point either way each year (49.8% in 2018-19, 50.4% in 2021-22, 51.3% in 2024-25), but the broad order of magnitude has been stable. Most other UK certifications and exams have far higher pass rates, the driving test is deliberately calibrated to be a genuine threshold not a participation award.

The 50% figure is no accident. DVSA examiner training, route design, and the marking rubric are calibrated to produce a fail rate that means the test reliably distinguishes ready candidates from not-yet-ready ones. A fail rate of 20% would mean the test was too easy to identify under-prepared candidates safely on the road. A fail rate of 80% would mean the test was failing genuinely competent drivers. 50% sits in the band where the test does the work it is meant to do.

#Where the fail marks actually go

A fail is normally one serious or dangerous fault plus a cluster of supporting minors. The serious fault is the one the examiner names on the marking sheet at the end of the test. The supporting minors usually concentrate in the same skill category as the serious. Across 1.9 million annual tests, the fault marks are not spread evenly, five categories together account for around half of every fault recorded.

#Junction observation: why this dominates

Junction observation is the most-recorded fault category by a wide margin. The category covers any failure to look properly at a junction, missed mirror checks, incomplete head turns, looking but not seeing, looking at the wrong moment, not holding the look long enough to register what is approaching. Examiners record it on most tests because the opportunities to slip are high frequency, a typical 40 minute test passes through 15 to 25 junctions of various kinds.

The fix is mechanical, not abstract. A full head turn at every junction, hold the look until you have actually scanned both directions, time the look to give you usable information for the move you are about to make. A glance taken too early sees nothing useful, a glance taken too late is already inside the move. Most candidates know the rule, the candidates who do not fail on it have made the routine automatic through repetition. Drilling junctions is the highest-yield activity for any retake candidate whose marking sheet shows junction observation as a serious.

#Fail rate by centre: a 38 point spread

The UK average of 51.3% is the volume-weighted aggregate across roughly 570 centres. The individual centre fail rates range from around 28% (Lerwick on Shetland, 72% pass rate) to 67% (Belvedere in London, 33% pass rate). The 38 percentage point gap is the largest single variable in your statistical chance of failing.

#Why London centres dominate the fail rankings

The hardest centres in the UK are almost all in London. Belvedere, Wanstead, Erith, Hayes, Chingford, Croydon, Tolworth, all fail at 60% or higher. The pattern is not random and not about marking variance. The DVSA trains every examiner to the same rubric and audits the marking nationally. What differs between London and Lerwick is the road environment.

A 40 minute test at a London centre includes more multi-lane junctions, more bus lanes, more pedestrian crossings, more parked-vehicle obstructions, more complex roundabouts, more lane discipline opportunities, and more time-of-day traffic pressure than a 40 minute test in rural Scotland. The fault opportunities per minute are simply higher. Every London junction is a chance to miss a check. Every London lane change is a chance to forget a mirror. The marks accumulate faster, the test reaches the 16-minor cliff or a serious sooner. See the why London test centres hard guide for the full breakdown.

#The retake fail rate: does it get better?

Second-attempt candidates fail at 50.4% (versus 51.1% first-attempt). Marginal improvement, around one percentage point. The Reddit conventional wisdom that "everyone passes second time" is wrong. The lift comes from candidates addressing the specific faults the first marking sheet recorded, not from the second test itself. A learner who books a retake without taking targeted lessons in between gets roughly the same odds they had the first time.

By the third attempt, the fail rate has climbed above the first-attempt baseline. This is not because the test gets harder, it is because the cohort still testing on attempt three or later increasingly contains candidates with structural issues that simple drilling does not resolve. The implication for a multi-fail candidate is that the addressable lever is rarely "more lessons of the same kind". A real change to instructor, centre, or learning pattern is usually what breaks the cycle. See the driving test after failing guide for the retake planning detail.

#How fail rates vary by demographics

DVSA publishes demographic breakdowns annually. The headline patterns: male candidates fail at 50.4%, female candidates at 52.4%, a 2 percentage point gap that has narrowed from around 4 points a decade ago. Younger candidates (17-19) fail at around 48%, older candidates (40+) at around 60%. First-attempt candidates fail at 51.1%, those who learn over a year or more typically fail less often than those compressing into intensive courses.

None of these are large effects compared to centre choice. The 2 point gender gap is dwarfed by the 38 point centre spread. The 12 point age gap is real but mostly addressable with more lesson density rather than fewer. The full demographic analysis is on the male vs female pass rates guide and the driving test pass rate by age guide.

#The five faults most likely to fail you

#What the fail rate does not mean

Three misinterpretations of the 51.3% headline are worth correcting. The first is that "half of UK drivers cannot drive". The fail rate measures test attempts, not driver competence at large. Most candidates who fail eventually pass, the licence holding population is the people who passed at some point, not the people who passed first time.

The second is that the test is "unreasonable". The 51.3% rate is deliberate calibration. A 20% fail rate would mean the test was failing to identify under-prepared candidates. The current rate sits in the band where the test reliably distinguishes ready from not-yet-ready, which is what a public-safety driving certification is meant to do.

The third is that "everyone fails first time". The first-attempt pass rate is 48.9%, nearly half of all first-time candidates pass. Treating a fail as inevitable can produce under-prepared booking decisions (book early because "I will fail anyway"), which then becomes a self-fulfilling prophecy. A well-prepared first-time candidate has a real shot, and treating the first attempt as a serious assessment rather than a rehearsal raises the odds.

#How the fail rate connects to wider test stats

The why do people fail driving test guide covers each fail category in more depth. The how many minors fail driving test UK guide covers the specific 15-minor threshold. The easiest vs hardest test centres guide covers the centre-by-centre fail rate distribution. For learners using the fail rate data to make booking decisions, the should I travel for easier test guide covers the cost-benefit of optimising for centre choice.

#Sources and further reading

The figures, fees, and procedures referenced in this article are verifiable on the official gov.uk pages below. PassRates.uk is built on the Driver and Vehicle Standards Agency’s open data, published under the Open Government Licence.

• Driving test (gov.uk)
• Driving test fees (gov.uk)
• If you fail your driving test (gov.uk)
• Change or cancel your driving test (gov.uk)
• Driving licence categories (gov.uk)
• Open Government Licence v3.0 (nationalarchives.gov.uk)