Crime is seasonal in data that spans decades: per FBI published records, homicide, assault, and nearly all property offenses rise through the summer months and fall in winter, with monthly averages for some offenses moving 10 to 20 percent or more between seasonal peaks and troughs. The pattern is stable enough that criminologists treat it as a baseline, and any month-to-month crime claim that ignores it starts from the wrong comparison.
This site publishes information, not legal advice.
What does the summer crime increase actually look like?
The strongest documented pattern is violent. Per FBI monthly offense tables, homicide and aggravated assault run consistently higher from May through August than from November through February, with July and August typically the peak months. The seasonal amplitude for homicide has been measured at roughly 25 to 35 percent between winter lows and summer highs in national series, a swing large enough to swamp most single-month year-over-year changes. Property offenses follow the same direction with smaller amplitude: burglary and larceny rise modestly in summer, and motor vehicle theft shows a summer bump with year-to-year variation.
Two long-standing exceptions appear in the published record. Burglary of unoccupied homes has historically shown a winter component in some analyses, tied to darkness and travel patterns, and shoplifting-related larceny peaks around the November-December retail season rather than summer. Seasonality is real, but it is offense-specific, not a single wave that lifts every category.
Why does crime rise in summer?
The explanations in the research literature fall into three groups, and the data supports all of them at once. First, opportunity: warm weather puts more people outside and in contact, and more unoccupied homes, which raises both interpersonal conflict and property opportunity. Second, routine activity: school vacations and changed work patterns alter where people and property are. Third, the heat-aggression hypothesis, supported by studies showing violence rising with temperature even within a single day and even in climate-controlled comparisons — though its independent effect is harder to isolate.
What the seasonal pattern does not establish is any single cause. The FBI's own publications present the pattern descriptively; causal weight comes from academic studies, and those studies disagree about how much of the summer increase is heat versus routine activity. A careful reader can accept the pattern without accepting any one mechanism.
Which seasonal claims do not survive the data?
Several durable folk claims fail against the records. Full moons have no demonstrated effect on offense levels in rigorous analyses. Holiday-specific claims mostly fail outside retail theft: per FBI monthly tables, most offenses do not spike at Halloween or in the summer's late weeks beyond the ordinary seasonal curve. The claim that crime rises immediately after major public events or elections is not supported in national series, though specific local events have documented short-lived local effects.
The most consequential bad comparison is the month-over-month one: comparing this July to last June rather than last July mistakes a season for a trend. Per FBI published data, a typical June-to-July increase in assault is expected behavior, while a July-to-July change is the actual signal.
Related stories: What NIBRS data shows about property crime trends · Why single-cause explanations of the crime drop rarely survive scrutiny.
How do analysts adjust for seasonality?
Standard practice compares the same calendar months across years, or uses twelve-month rolling totals that average the seasons together. Statistical agencies apply formal seasonal adjustment — estimating the typical monthly pattern from years of history and removing it — the same way they do for employment figures. The FBI's quarterly and annual releases mostly present annual or same-month comparisons; detailed seasonal adjustment appears more often in academic work and in analyses of city-level monthly data.
Adjustment choices matter at the edges. During and after the 2020 pandemic disruption, ordinary seasonal patterns broke — lockdowns flattened the usual summer rise in some categories in 2020, and the rebound ran through 2021 and 2022 — which is why per-month comparisons across that period carry extra caveats in the published literature.
Does seasonality differ by region and city?
It does, and the differences are informative. Northern cities with sharp winters tend to show the largest seasonal amplitude — the summer peak is higher relative to a deep winter trough — while Sun Belt cities show flatter curves, with violence present year-round. Domestic violence research complicates the summer story: some studies find intimate partner violence peaking in summer months while others find it tracking weekends, holidays, and hot individual days more than seasons. Categorized by offense, the regional spread is smaller for homicide seasonality than for property crime, which follows local climate more directly.
The practical implication is for readers of city-level data: a New York monthly curve and a Phoenix monthly curve are not comparable without seasonal context, and a mild winter in a northern city can produce a scary-looking January-to-January increase that is mostly weather.
What did the pandemic years do to the pattern?
The 2020 and 2021 data is the strongest recent demonstration of why the baseline matters. During lockdown months, the usual rhythms flattened: overall property crime fell while the ordinary summer rise was muted or absent, and violence shifted away from downtown commercial zones and toward residential neighborhoods, per analyses of city-level data published through 2021 and 2022. As restrictions lifted, the seasonal curve reasserted itself, and 2022 showed a return toward ordinary patterns even while annual levels remained elevated above pre-pandemic baselines.
What should readers do with a seasonal crime claim?
- Check whether the comparison is same-month or same-quarter across years, not adjacent months.
- Check the offense: summer effects are strongest for violence, weakest or reversed for some property categories.
- Check whether an unusual year — 2020 through 2022 in particular — is inside the comparison window.
- Distrust single-month percentages; the seasonal swing for homicide is larger than most real annual changes.
Seasonality is one of the most replicated findings in crime data, and per FBI published records it holds across decades, regions, and reporting systems. That stability is useful exactly because it is a baseline: once the seasons are accounted for, whatever remains is the trend a reader is actually looking for.
For more context, read How clearance rates are calculated, and why they often mislead.
For more context, read nibrs.
For more context, read victimization survey.