08-05-2026, 09:04 AM
All done and dusted. It's a bit of a joke though, them claiming that "errors are mercifully rare"!!!
Whilst not all private parking PCNs are issued through ANPR, the majority are and this does not support Horizon’s claim that errors are "mercifully rare". The unregulated private parking industry issues over 14 million PCNs per year. The vast majority of these come from ANPR-controlled sites because most major operators use ANPR as their primary enforcement method.
Even if only half of all PCNs were ANPR-generated, applying known real-world ANPR error rates still produces extremely large numbers of incorrect charges. Independent data shows ANPR misread rates between 3% and 10%. At 7% error, 7% of 7 million ANPR-triggered PCNs is 490,000 errors. At 10% error, it is 700,000 errors. Even the lowest credible rate of 3% still produces 210,000 errors.
These numbers are not "rare" by any definition. Horizon’s claim is statistically false because any system producing hundreds of thousands of errors per year cannot be described as "mercifully rare", regardless of whether some PCNs are issued through other evidential methods.
The known data on ANPR failures comes from three independent and credible sources. First, consumer testing by Which? found that private‑sector ANPR systems misread around 3% of plates in real‑world conditions. Second, DVLA enforcement data shows an 8.9% misread rate based on manual review of suspect ANPR images, which is one of the few audited datasets available. Third, technical studies of ANPR performance in normal operating environments consistently report error rates between 7% and 15%, caused by factors such as weather, camera angle, plate condition, and OCR limitations. These three sources form the basis of all reliable estimates of ANPR failure rates.
Whilst not all private parking PCNs are issued through ANPR, the majority are and this does not support Horizon’s claim that errors are "mercifully rare". The unregulated private parking industry issues over 14 million PCNs per year. The vast majority of these come from ANPR-controlled sites because most major operators use ANPR as their primary enforcement method.
Even if only half of all PCNs were ANPR-generated, applying known real-world ANPR error rates still produces extremely large numbers of incorrect charges. Independent data shows ANPR misread rates between 3% and 10%. At 7% error, 7% of 7 million ANPR-triggered PCNs is 490,000 errors. At 10% error, it is 700,000 errors. Even the lowest credible rate of 3% still produces 210,000 errors.
These numbers are not "rare" by any definition. Horizon’s claim is statistically false because any system producing hundreds of thousands of errors per year cannot be described as "mercifully rare", regardless of whether some PCNs are issued through other evidential methods.
The known data on ANPR failures comes from three independent and credible sources. First, consumer testing by Which? found that private‑sector ANPR systems misread around 3% of plates in real‑world conditions. Second, DVLA enforcement data shows an 8.9% misread rate based on manual review of suspect ANPR images, which is one of the few audited datasets available. Third, technical studies of ANPR performance in normal operating environments consistently report error rates between 7% and 15%, caused by factors such as weather, camera angle, plate condition, and OCR limitations. These three sources form the basis of all reliable estimates of ANPR failure rates.
Never argue with stupid people. They will drag you down to their level and then beat you with experience. - Mark Twain
