How do mass shooting trends excluding gang violence vary by state and city from 2015–2024?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
Mass-shooting counts that exclude gang-related and other felony-motivated shootings show patterns that are localized, rare, and sensitive to how researchers define and count incidents: some long-run datasets show an overall rise over decades while others report recent year-to-year declines, and the signal varies widely by state and city because incidents are both uncommon and unevenly distributed [1] [2] [3].
1. Definitions drive the map
Any comparison across states and cities begins with a definitional problem: major databases differ in whether they count injuries as well as deaths, whether they include gang, drug or domestic-motivated events, and what victim threshold they use (three or four killed versus four or more shot), and these methodological choices materially change which incidents are tallied for a given place [4] [5] [6].
2. National trendlines: mixed short-term declines, long-term increases
Recent annual reporting shows a meaningful drop from 2023 to 2024 in several trackers — The Trace reported 24% fewer mass shootings in 2024 than in 2023 and characterized 2024 as part of a decline from the pandemic-era surge [2] — yet multi-decade research products such as the Rockefeller Institute’s factsheet continue to describe an overall upward trajectory over many decades even as their year-to-year counts can be low and variable [1] [7].
3. State and city variation: concentrated, rare, and episodic
Mass public shootings that exclude gang violence are concentrated in a small number of states and cities and are episodic events whose occurrence in any state or city is driven more by chance variability and isolated incidents than by steady, predictable rates; RAND and other analysts warn that the rarity of such events makes robust statistical comparisons across states and cities fragile and highly sensitive to outliers [3].
4. Policy differences correlate but causation is hard to prove
Some analyses and commentators point to state-level policy differences — for instance, states that tightened firearm rules have enacted site-specific restrictions and other measures — and correlate these with lower mass shooting activity in particular years, but rigorous causal claims are difficult because the outcome is rare, models risk overfitting, and different datasets produce different spatial patterns [2] [3] [8].
5. Databases disagree about place-specific burden
Because the Gun Violence Archive, Mother Jones, the Violence Project, the Regional Gun Violence Research Consortium, and media-compiled databases use different inclusion rules, a city or state can appear to have many mass shootings under one tracker and few under another — for example, 2024 national incident totals range from several dozen in curated “public mass shooting” compilations to hundreds in broader, injury‑based trackers — so any state-by-city ranking depends on which source is chosen [5] [1] [2].
6. Shooter profiles and local context matter for comparison
Research into perpetrator histories finds common features — prior criminal records, violence, and in some series military backgrounds or suicidality — that are relevant to prevention but not evenly distributed geographically; local mental‑health, law‑enforcement, and social services environments interact with those individual risk factors, so two cities with similar raw counts can have very different causal mixes and policy needs [9] [10].
7. How to interpret city/state differences responsibly
Because single incidents can skew small-area statistics and because definitions and reporting intensity vary by jurisdiction, responsible comparisons should use multiple datasets, examine multi-year windows (not single years), and treat per-state or per-city counts as indicative rather than definitive; studies caution against overinterpreting short-term declines or spikes without considering definitional and sampling uncertainty [6] [3].
8. What the available evidence allows and what it does not
The reporting allows confident statements that trends differ by place, that 2024 saw a notable decline from 2023 in several trackers, and that long‑run counts used by some research groups show an upward pattern, but the evidence does not support a single, definitive ranking of states or cities because of inconsistent definitions, rare-event statistics, and database discrepancies [2] [1] [6].