Violent crime in a city is not spread evenly across it but concentrates in a small number of very small places. Hot spots policing sends officers to those places specifically, and about thirty years of accumulated evidence shows it does reduce crime at the places treated. The question no one had answered is whether cutting crime in those small places also lowers the crime figures for the whole city, which should not be assumed, since the treated places are only a fraction of the city. This study draws on seven years of violent crime data from three large Texas cities. Dallas ran a city-wide violent crime reduction plan built on hot spots policing for two years, while Fort Worth and Arlington, in the same metropolitan area and running no comparable programme, served as comparison cities. The authors used interrupted time series analysis, which compares the trend in the data before and after the point an intervention begins. Dallas recorded a significant fall in violent crime immediately at the point of intervention; the other two cities did not.

Research Objectives

- Answer the question the existing literature had left open: whether localised crime reductions from hot spots policing lower crime rates across an entire city.

- Compare violent crime trends in the treated city against similarly situated cities in the same metropolitan area.

- Separate the effect of the intervention from pre-existing crime trends and from the impact of the COVID-19 pandemic.

- Test whether the city-wide reduction came from the treated places, by comparing treated grids against grids that never received treatment.

Methodology

- Murder, aggravated assault and robbery incidents in Dallas, Fort Worth and Arlington from May 2016 to April 2023, covering 84 months. February 2021 was excluded because all three cities recorded abnormally low counts during a severe winter storm, leaving 83 months: 59 before the intervention and 24 after.

- Certain aggravated assault cases were removed from the Dallas and Arlington data because the intervention was not designed to address family violence; Fort Worth's data could not be differentiated on this basis, making it the only city whose figures include family-related incidents.

- Dallas began its violent crime reduction plan in May 2021 under a newly appointed police chief, overlaying a grid network across the whole city with cells of 330 by 330 feet, roughly the length of an American football field on each side, producing 101,103 grids.

- Analysis before the intervention found that fewer than 6,000 of those 101,103 grids accounted for all violent crime in the city, and that at any given time just 50 to 60 grids accounted for 8 to 10 per cent of the city's violent crime, with that set of grids rotating every 60 days.

- Each treatment cycle therefore covered 50 to 60 grids, running 90 days at first and later shortened to 60 days once analysis showed most of the benefit occurred within the first 60 days, reaching 313 unique grids between May 2021 and December 2023. Fidelity, measured as an officer being present in the correct grid within the correct hour, ranged from 62 to 89 per cent and averaged roughly 80 per cent.

- Interrupted time series analysis (ITSA) was used with Newey-West corrections to address autocorrelation in the series.

Key Findings

- Dallas's average monthly violent crime fell from 630.81 incidents before the intervention to 540.83 after, a reduction of 14.26 per cent, while Fort Worth fell 4.98 per cent and Arlington rose 18.06 per cent.

- The ITSA results confirmed this. At the intervention point Dallas returned a coefficient of -53.269, statistically significant, indicating an immediate drop, while Arlington and Fort Worth showed significant increases of 40.556 and 64.536 respectively.

- Dallas's post-intervention trend also fell significantly more steeply than its pre-intervention trend (-4.916), whereas Fort Worth showed no change in trend.

- Over a four-year comparison window, treated grids fell 56.2 per cent, from a monthly average of 23.2 incidents to 14.8, while untreated grids rose 21.5 per cent, from 376.8 to 480.2. The city-wide figure still fell 20.8 per cent, because the reduction at the hot spots more than offset the increase elsewhere.

- The effect appeared strongest immediately after implementation, despite fidelity being at its lowest in that first period at 62 per cent. The authors read this as the visible presence of officers in hot spots producing a deterrent effect almost immediately.

Recommendations

- Police agencies facing staffing shortages can use hot spots policing as a cost-effective approach, concentrating limited resources on the small number of places generating the most violent crime.

- Fidelity must be kept high, since the authors state the results depend on the strategy being implemented carefully and with high fidelity.

- Target areas should be rotated according to current data, since the grids that constitute hot spots change every 60 days.

- Future research should replicate the study in other jurisdictions, examine the long-term sustainability of the reductions, and test whether complementary measures such as community-based interventions increase the effect. The authors report replications under way in Tacoma, Salt Lake City and San Antonio, with preliminary findings showing comparable reductions.

Key Takeaways

- Treating just 313 small grids in a city of more than a hundred thousand produced a violent crime reduction measurable at city level, because those places account for a disproportionate share of the crime.

- Crime in untreated areas rose 21.5 per cent over the same period, but the 56.2 per cent fall at the hot spots was large enough to absorb that increase and still leave a net reduction, so the city-wide figure went down.

- The authors note their study covers one metropolitan area in Texas and cannot fully rule out other factors affecting crime in Dallas. For Thailand the structure of this intervention resembles risk-based patrol deployment. This connection is our own; the study holds no Thailand data.

References

Tregle, B., Smith, M. R., Fahmy, C., & Tillyer, R. (2025). More than meets the eye: Examining the impact of hot spots policing on the reduction of city-wide crime. Crime Science, 14, Article 3. https://doi.org/10.1186/s40163-025-00247-9

Full text (Open Access): https://crimesciencejournal.biomedcentral.com/articles/10.1186/s40163-025-00247-9