Coconuts Kill 150 People a Year — But That Number Is Meaningless

The Statistic That Launched a Thousand Bar Arguments

Somewhere in the mid-2000s, a number escaped from academia and went viral: coconuts kill roughly 150 people per year, making them more deadly than sharks. It appeared in travel guides, pub quizzes, and the kind of listicles that promise to “blow your mind.” The implication was clear: we’ve been afraid of the wrong thing all along.

The figure is often attributed to a 1984 paper by Dr. Peter Barss published in The Journal of Trauma, which documented head and spine injuries from falling coconuts at a hospital in Papua New Guinea. But Barss never claimed 150 global deaths. That number appears to originate from a 2002 travel insurance press release by Club Direct, not from any peer-reviewed research. The Barss paper was careful, methodologically sound work about a real hazard in a specific context. What happened next (a fabricated global death toll laundered through citation chains until it sounded authoritative) is a lesson in how statistics travel badly.

Why Everyone Dismisses It

The “coconuts vs. sharks” comparison works as trivia precisely because both numbers seem absurd. Sharks kill ten people a year. Coconuts kill 150. Neither sounds like a serious threat when you divide either figure by eight billion. The result is that people internalize the statistic as a curiosity rather than actionable information.

It becomes a punchline: watch out for coconuts, not sharks! The joke does real epistemic damage, because most people walk away believing both risks are negligible, conflating “low on a global average” with “low for me, in my situation.”

The 150-per-year figure is fine. What’s missing is the denominator.

The Exposure Problem

A “global annual risk” divides by eight billion people, but the overwhelming majority of them never stand under a coconut palm. They live in temperate regions, in cities, in landlocked countries. The coconut palm (Cocos nucifera) grows in a band between 26°N and 26°S latitude, in tropical and subtropical coastal environments spanning parts of Southeast Asia, South Asia, the Pacific Islands, the Caribbean, and coastal East Africa.

Even within those regions, most people have limited daily exposure to palms in fall zones. Meaningful exposure concentrates among a few groups:

• Agricultural workers harvesting coconuts
• Residents of villages where palms line footpaths and shade homes
• Tourists walking along tropical beaches
• Workers at resorts where palms are planted for aesthetics without regard for who walks beneath them

The Barss study was conducted at Alotau Hospital in Milne Bay Province, Papua New Guinea, where people live and work beneath coconut palms for hours each day. Generalizing from the global average to your risk ignores everything that actually determines whether you’re in danger.

The Real Math

Once you account for exposure, the arithmetic becomes more interesting. A mature coconut palm produces roughly 75 coconuts per year (the FAO puts the range at 70 to 100), which works out to about one coconut falling every 5 days per tree, or a fall rate of 0.0086 coconuts per hour per tree.

Now the geometry. A coconut palm’s canopy spans 6 to 10 meters, and wind drift during a 20 to 30 meter fall widens the drop zone to roughly 60 square meters. Seen from above, which is the only angle a coconut cares about, a person covers about 0.15 square meters of head and shoulders. That gives a hit probability of about 0.25% per fall, or 1 in 400, for someone standing under the canopy when a nut lets go.

If a coconut does hit you, the outcome depends heavily on where it lands. No study reports deaths per impact, so this step is our estimate. A 2025 retrospective study at a tertiary trauma center in coastal India (PMC11829407) found that about 21% of the 29 patients who reached the emergency department had severe head trauma, and none died. Barss saw no in-hospital deaths either, but recorded two people killed outright in their villages over four years. Most impacts are glancing blows to shoulders and backs that never reach a hospital at all. If roughly 1 in 20 impacts is serious enough for an emergency department and about 4% of those prove fatal, the fatality rate per impact comes out near 0.2%.

Chain those probabilities together (fall rate, hours of exposure, number of trees, hit probability, fatality rate) and one hour a day under a single palm works out to an annual death risk of about 1 in 64,000. As a sanity check, that rate would need around 10 million people with that level of exposure to produce 150 deaths a year, which is a plausible count of farm workers and villagers who spend an hour a day directly beneath a canopy.

Your Actual Risk

The calculator below lets you adjust your actual exposure. The key variables are how many hours per day you spend under palms, and how many trees are in your vicinity. A resort worker walking beneath a row of decorative palms all day has a fundamentally different risk profile than a tourist who passes under two palms on the way to the beach once a year. At the default setting of one hour a day under one tree, the annual risk is about 1 in 64,000. Push it to eight hours a day under ten trees, a plantation worker’s day, and it reaches roughly 1 in 800, ten times the risk of dying in a car crash.

How It Compares

For scale: the annual fatality risk from a car crash in the United States is roughly 1-in-8,000, a risk almost everyone accepts without much thought. Lightning strike runs about 1-in-16 million per year for the average American (roughly 20 deaths a year, per the NWS). Bee sting death sits around 1-in-4.6 million (about 72 deaths a year, per the CDC).

At low exposure, passing under one palm for a few minutes a day, coconut risk falls well below driving risk, though still far above lightning. But for agricultural workers in Papua New Guinea or Kerala or the Philippines who spend entire workdays beneath palms, the calculated risk climbs past driving and into ranges that qualify as an occupational hazard.

The comparison to sharks also deserves scrutiny. Sharks kill roughly ten people per year globally, not 150. If coconuts are “more deadly than sharks,” that reflects two distortions at once: we underestimate coconut risk by using the wrong denominator, and we overestimate shark risk from media coverage. The fix is the same for both figures: divide by the exposed population, not all eight billion humans.

Sources

• Barss, P. (1984). Injuries due to falling coconuts. Journal of Trauma, 24(11), 990–991. The original study conducted at Alotau Hospital, Milne Bay Province, Papua New Guinea, documenting head and spinal injuries from coconut falls.
• Palm biology references: Coconut palm (Cocos nucifera) annual yield data from the Food and Agriculture Organization of the United Nations (FAO), Coconut: Post-harvest Operations, 2007.
• Canopy geometry estimates based on published palm morphology data. Canopy diameter of 6–10 meters per Dimensions.com; effective drop zone ~60m² accounting for wind drift at fall heights of 20–30 meters. Site estimate.
• Human top-down projected area (head and shoulders) of ~0.15 square meters, giving a hit probability of 0.25% per fall. Site estimate.
• Fatality rate: India 2025 retrospective study (PMC11829407), 29 patients at Kasturba Medical College, Manipal. 20.6% severe (intracranial trauma), 0% in-hospital mortality. Barss (1984): 0 in-hospital deaths, 2 deaths in villages over 4 years. The 0.2% per-impact fatality rate used here is a site estimate (~1 in 20 impacts reaching an emergency department, ~4% of those fatal); no published per-impact case fatality rate exists.
• Lightning comparison: NWS, 10-year average of 18.6 US lightning deaths per year, 2015-2024. Bee sting comparison: CDC, ~72 US deaths per year from hornet, wasp and bee stings.
• The “150 deaths per year” figure: traced by The Straight Dope and Snopes to a 2002 Club Direct travel insurance press release, not to any peer-reviewed source. Barss (1984) reported no global death estimate.