Which Auckland Junctions Have More Fatal or Serious-Injury Crashes Later? Comparing Historical Crash Rankings

Objective. Ranking junctions by fatal and serious-injury (DSI) crashes in the last five years is a simple way to pick safety sites, but the counts are noisy and unlucky sites tend to look better later. This study tests that rule, as a benchmark rather than any agency's procedure, against all-injury lists. Data. Public NZ Transport Agency crash records and OpenStreetMap junctions: 21,789 Auckland junctions within 500 m of a 2006 to 2019 injury crash, ranked on 2015 to 2019 and tested on 2022 to 2025. Outcomes are crashes, not people. Tests were pre-registered, including the 1,000-junction primary list size, set on 9 November 2026 by a power rule before any test-year junction outcome was opened (originally 310, and 85 before that). Findings. At 1,000 junctions, an all-injury list with empirical-Bayes smoothing and a predicted DSI share caught 285 later DSI crashes against 192: ratio 1.484 (95% interval 1.257 to 1.705). At 310 it is inconclusive: 1.089 (0.914 to 1.366). Exploratory checks suggest most of the gain comes from using all injury crashes (plain all-injury count: 1.359) and from the DSI rule filling its list with single-DSI sites; the extra modelling (not separated here) adds little. The 310 top-DSI junctions held 36.3% of junction DSI crashes in 2015 to 2019 but 10.4% in 2022 to 2025, mixing regression to the mean with any safety work. What is new. No earlier held-out test of these rankings on urban junctions was found (search not verified as complete). Practical use. For analysts shortlisting about 1,000 junctions from crash counts alone, results favour all injury crashes over DSI alone; at about 310, there is no evidence either way.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-16
DOI
https://doi.org/10.5281/zenodo.24280918
Primary Topic
Traffic and Road Safety
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Which Auckland Junctions Have More Fatal or Serious-Injury Crashes Later? Comparing Historical Crash Rankings

Jayden Ling
1 links
Zenodo (CERN European Organization for Nuclear Research)
Traffic and Road Safety
preprint

Which Auckland Junctions Have More Fatal or Serious-Injury Crashes Later? Comparing Historical Crash Rankings

Jayden Ling
preprint en

Abstract

Objective. Ranking junctions by fatal and serious-injury (DSI) crashes in the last five years is a simple way to pick safety sites, but the counts are noisy and unlucky sites tend to look better later. This study tests that rule, as a benchmark rather than any agency's procedure, against all-injury lists. Data. Public NZ Transport Agency crash records and OpenStreetMap junctions: 21,789 Auckland junctions within 500 m of a 2006 to 2019 injury crash, ranked on 2015 to 2019 and tested on 2022 to 2025. Outcomes are crashes, not people. Tests were pre-registered, including the 1,000-junction primary list size, set on 9 November 2026 by a power rule before any test-year junction outcome was opened (originally 310, and 85 before that). Findings. At 1,000 junctions, an all-injury list with empirical-Bayes smoothing and a predicted DSI share caught 285 later DSI crashes against 192: ratio 1.484 (95% interval 1.257 to 1.705). At 310 it is inconclusive: 1.089 (0.914 to 1.366). Exploratory checks suggest most of the gain comes from using all injury crashes (plain all-injury count: 1.359) and from the DSI rule filling its list with single-DSI sites; the extra modelling (not separated here) adds little. The 310 top-DSI junctions held 36.3% of junction DSI crashes in 2015 to 2019 but 10.4% in 2022 to 2025, mixing regression to the mean with any safety work. What is new. No earlier held-out test of these rankings on urban junctions was found (search not verified as complete). Practical use. For analysts shortlisting about 1,000 junctions from crash counts alone, results favour all injury crashes over DSI alone; at about 310, there is no evidence either way.

Zenodo (CERN European Organization for Nuclear Research)
Good health and well-being, Sustainable cities and communities
Traffic and Road Safety
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon