Hurricane Erin became the Atlantic's first Category 5 of 2025 last week, broke the basin's twenty-four-hour rapid-intensification record, and killed nine people in Cape Verde before the WMO declined to retire its name. The decision to keep Erin in active rotation was unusual enough to draw a small wave of confused commentary. Most of the commentary repeated a folk theory that almost everyone in American naming culture believes: that hurricane names get destroyed by the storms that wear them. The folk theory is mostly wrong, and the place where it goes wrong is interesting.
What the SSA record actually shows
The folk theory, in its strong form, says that any name attached to a major hurricane will collapse in the years following the storm. The standard examples are Katrina and Sandy. Katrina lost 86 percent of its peak popularity in the years after 2005, going from a top-300 girls' name to outside the top 1000 by 2015. Sandy lost 78 percent in the years after 2012. These two cases get cited in every storm-name article for a reason: they are the two cleanest examples in the historical record. The problem is that they are also the two examples that prove a much more limited rule than the folk theory wants to believe.
If you look at the names of every Atlantic hurricane that has been retired by the WMO since 1953 — there are about ninety-three of them — and you compare each name's pre-storm SSA trajectory to its post-storm trajectory, what you find is that most retired hurricane names show no detectable acceleration in their decline. Andrew, after the 1992 Florida storm that killed 65 people, continued its existing decline at the same rate. Wilma, after the 2005 Yucatan storm, showed no SSA effect at all. Hugo, Ivan, Charlie, Floyd, Isaac, Maria, Ian — none of them produced visible SSA accelerations. The names that did produce accelerations are a small subset.
What's special about Katrina and Sandy
The names that did show real accelerations share three properties. First, the storm made U.S. mainland landfall in a major population center. Second, the casualty count among English-speaking Americans was high enough to dominate the U.S. news cycle for weeks rather than days. Third, the name was already in a vulnerable position — declining or just past peak — when the storm hit. Katrina checks all three. Sandy checks all three. Andrew, despite being a deadly Florida hurricane, fails the third condition because Andrew was a stable top-100 name with no demographic vulnerability.
The mechanism is, I think, fairly clear. American parents do not consciously associate names with storms in most cases. The unconscious association requires the storm to have produced enough English-language American grief — enough funerals, enough damaged neighborhoods, enough media coverage — that the name acquires what behavioral economists call an availability heuristic. The name becomes more available, in memory, in its negative association than in its neutral one. Most retired hurricane names do not reach that threshold. Two have.
Why Erin will not follow Katrina
Erin's storm was significant but did not check any of the three boxes. It made no U.S. mainland landfall. The casualties were in Cape Verde, which is heartbreaking but does not enter U.S. memory in the same way as a New Orleans hurricane. And Erin, the name, is not in a vulnerable position — it sits in the girls' top 350, declining slowly along with most other 1990s-popular Anglo names, with no acute trajectory weakness that could be exploited by a media event.
The WMO's decision not to retire the name is, in this light, structurally consistent with how most retired-name decisions actually shake out. The WMO's published criteria for retirement focus on damage and casualties; they do not directly consider the future linguistic life of the name. But the criteria they use happen to correlate well with the conditions that produce SSA accelerations, because the same factors — U.S. landfall, English-speaking casualties, length of news cycle — drive both the WMO's retirement decision and the unconscious naming response.
The names that survived their storms
The cleanest counter-examples to the folk theory are the names that are still in the SSA top 1000 despite having been worn by retired hurricanes. Andrew is the obvious one — top 60 in the boys' chart even today, despite being attached to a 1992 storm whose name was retired. Erin, similarly, has continued to function as a normal mid-decline 1990s name even after the 1995 hurricane Erin that brushed Florida and after numerous smaller storms in subsequent decades. Allison, despite being attached to a deadly tropical storm in 2001 in Houston, did not collapse — Allison was already declining and continued at roughly the same rate. Ian, the 2022 storm that hit Florida and killed over 150 people, has had no detectable SSA effect on Ian as a name as of the most recent data; the name has continued its long pre-storm decline at the same rate.
The Ian case is particularly useful because it's recent and because the storm did, in fact, dominate U.S. news for weeks. By the folk theory, Ian should have collapsed. It did not. The SSA effect, if there is one, is too small to distinguish from existing variation. Two years out from the storm, Ian sits in roughly the position the pre-storm trend would have predicted.
The deeper question of grief and naming
What I find interesting in the WMO data, beyond the specific Erin case, is what it suggests about how grief actually moves through naming culture. Names do not absorb negative associations from external events as easily as the folk theory wants to believe. Most of the time, the name continues. The cultural memory of a storm is shorter than the cultural memory of a name. Andrew survived 1992. Allison survived 2001. Ian is surviving 2022. The name continues to mean what it meant before the storm — friend's son, character in a book, person on the team — and the storm becomes one association among many rather than the dominant one.
Katrina is exceptional because the storm did not become one association among many. It became the association, for at least a generation of Americans, and that's a high bar. Sandy is exceptional because the storm reactivated a name that had already been losing ground and gave the existing decline a final, public shove. Both of these are unusual circumstances, not the typical result.
What this means for parents
If you are a parent considering a name that happens to be on the WMO's active hurricane rotation, I'd argue you can ignore the storm risk for the vast majority of names. The base rate of catastrophic naming damage from a hurricane is very low. The names that historically have been damaged were already vulnerable, and the storms that damaged them were unusual in ways that are visible in the news cycle while they are happening — you can tell, while a hurricane is occurring, whether it is going to be a Katrina or whether it is going to be one of the other ninety-one retired storms that left almost no naming wake.
Erin is not a Katrina. The storm was destructive, the casualties were tragic, but the cultural conditions that would produce a long-term naming effect are not in place. By 2027 SSA data, Erin will be exactly where its pre-2025 trajectory predicted it would be — possibly within a few places either side, the kind of variation that's invisible without a careful look. The WMO's decision to keep the name in rotation will probably look, in retrospect, exactly correct.
Data source: U.S. Social Security Administration. Analysis by NamesPop.
Found this helpful?
Share it with someone who’s picking a name.