How Dangerous Is Skydiving, Really?

Sixteen Deaths, Three and a Half Million Jumps

In 2025, sixteen people died skydiving in the United States. The year before, it was nine. The United States Parachute Association (USPA), which has tracked fatalities since 1961, has not recorded more than twenty in a year since 2017. For a sport whose whole premise is leaving an aircraft at 13,000 feet, those totals look almost implausibly small, and drop zones are happy to say so. Their websites are full of reassurances that you face more danger on the drive to the airfield than on the jump itself.

The counts are accurate, but they are only half of a fraction. A death toll tells you how many people died, but the number that decides your risk is how many jumps produced those deaths, and how many of those jumps will be yours.

The Denominator Exists. Use It.

Skydiving is unusual among the risks on this site because the denominator is published. USPA surveys its members every year and estimates how many jumps they made. In 2025 that came to 3.47 million jumps; in 2024, 3.88 million. Add up the five years from 2021 to 2025 and you get 65 deaths in 18.5 million jumps, or one death for every 284,000 jumps.

That rate has fallen a long way. In 2008, thirty people died in 2.6 million jumps, a rate more than three times today’s.

The rate also bounces around more than the headlines suggest. In 2024 it hit a record low of 0.23 deaths per 100,000 jumps. In 2025, with fewer jumps and seven more deaths, it doubled to 0.46. Neither year means the sport suddenly became twice as safe or twice as deadly. When the annual toll sits in the teens, a handful of accidents swings the rate either way, which is why the five-year average is a better guide than any single season.

Lifetime Odds Against a Single Jump

The comparison drop zones reach for is driving. One New England drop zone’s safety page, a typical example, quotes the National Safety Council’s lifetime odds of dying in a motor vehicle crash (1 in 101) next to USPA’s per-jump fatality rate, and concludes that skydiving is safer than driving.

Those two numbers measure different things. One is the chance of dying on the road over an entire lifetime of commuting, errands and road trips. The other is the chance of dying on one jump. Put them on the same footing and the picture changes. US roads killed 1.20 people per 100 million vehicle miles in 2024, according to NHTSA, or one death every 83 million miles. Divide that by 284,000 and a single average skydive carries about the same risk as 290 miles of driving. That road rate counts everyone, pedestrians and motorcyclists included. Count only people inside cars and trucks (about 0.83 deaths per 100 million miles in 2023) and the equivalent stretches to roughly 420 miles, which only widens the gap below.

A skydive lasts roughly six minutes from exit to landing, about a minute of freefall from 13,000 feet and five minutes under canopy. Even at a steady 60 mph, 290 miles is nearly five hours behind the wheel. Minute for minute, the jump is far more dangerous than the drive. The reason skydiving looks safe is that almost nobody spends very many minutes doing it.

Tandem, Licensed, BASE

The average of 1 in 284,000 blends very different jumps. Tandem students, strapped to a rated instructor who handles the parachute, do better than average. USPA has put the tandem rate at about one student death per 500,000 jumps, averaged over ten years. In 2025 no student died at all.

All sixteen of 2025’s dead were licensed jumpers. USPA’s fatality summary describes deliberate low turns close to the ground, canopy collisions, and malfunctions where the jumper never pulled a handle. Strip out the tandem jumps and the licensed jumper’s per-jump rate sits somewhat above the overall average.

BASE jumping is a different sport altogether. Jumpers leave from cliffs, antennas and buildings with a single parachute and no reserve. A study by Søreide and colleagues in the Journal of Trauma (2007) went through eleven years of records at the Kjerag massif in Norway, one of the world’s most popular BASE sites. Over 20,850 jumps there were nine deaths, one for every 2,317 jumps. Per jump, that is about 120 times the current US skydiving average.

The Chain of Probabilities

The calculator below has only two links, and that simplicity is the point. Your annual risk is the number of jumps you make this year multiplied by the death rate per jump. The rate is fixed at 0.0000035, the USPA five-year average rounded to two significant figures (1 in 286,000).

As a sanity check, that rate should reproduce the real death toll. USPA says its members and first-time students make about 3.6 million jumps a year. At 1 in 286,000 per jump, 3.6 million jumps produce about 12.6 deaths, against an actual average of 13 per year from 2021 to 2025.

Your Actual Risk

At the default of one jump, the calculator gives about 1 in 286,000. That is the bucket-list tandem, and for a tandem specifically the true figure is closer to 1 in 500,000. Either way, the one-off jumper takes a risk comparable to a long weekend road trip.

The licensed jumper is in a different position, because the risk adds up with every load. At 85 jumps a year, roughly what you get by spreading USPA’s 3.47 million jumps across its 41,000 members, the annual risk is about 1 in 3,400, more than twice the yearly risk of dying in a US car crash. A committed weekend jumper doing 200 jumps a year reaches 1 in 1,400. Instructors and competitors who log 500 jumps a year land at about 1 in 570, roughly fourteen times the car-crash figure, and that is before adding the drive to the drop zone.

None of this is an argument against jumping. A risk of 1 in 286,000 per jump is genuinely low, and USPA’s numbers have improved year after year. The point is that the annual death count describes the sport, while your risk depends on how much of it you do.

How It Compares

The yearly chance of an American dying from lightning is about 1 in 16 million. A single tandem jump, at 1 in 500,000, carries the risk of about 32 years of lightning exposure. Bee, wasp and hornet stings kill about 1 in 4.6 million Americans a year, so one average skydive equals roughly sixteen years of sting risk.

Against the activities people usually class as extreme, skydiving holds up well. A climber on Everest faces about a 1% chance of death per expedition, the equivalent of nearly 3,000 average skydives. BASE jumping sits between them. A BASE jumper making 50 jumps a year at the Kjerag rate would face an annual risk of about 1 in 47, worse than a single Everest expedition.

The drop zones are right that skydiving is safer than it looks. They are wrong about why. A jump is not safer than driving. It is short, and most people only do it once.

Sources

• United States Parachute Association. How Safe Is Skydiving? (uspa.org/discover/faqs/safety). Annual US fatalities and estimated jumps, 2004-2025: 2021, 10 deaths in 3.57 million jumps; 2022, 20 in 3.9 million; 2023, 10 in 3.65 million; 2024, 9 in 3.88 million (0.23 per 100,000, record low); 2025, 16 in 3.47 million (0.46 per 100,000); 2008, 30 in 2.6 million (1.15 per 100,000).
• United States Parachute Association. About USPA (uspa.org/about/about-uspa). “41,000-plus members” who, with first-time students, make “roughly 3.6 million jumps per year” at almost 200 affiliated drop zones.
• United States Parachute Association, quoted in Fox News, “Skydiving instructor death: What to know on the sport and safety,” 21 February 2022. Tandem: “one student fatality per 500,000 jumps on average over the past 10 years.”
• USPA Skydiver’s Information Manual, Chapter 5. Freefall from 13,000 feet lasts about 60 seconds belly-to-earth, as little as 40 seconds in freefly positions. Canopy time of about five minutes per drop zone guidance.
• United States Parachute Association (April 2026). “A Need for Vigilance: The 2025 Fatality Summary” (uspa.org/uspa-news). 16 deaths in 3.47 million jumps; no student fatalities, only the second such year since USPA began keeping records in 1961; four deaths from intentional low turns, two from canopy collisions, two malfunctions where the jumper never pulled either handle.
• Skydive New England, “Is Skydiving Safer Than Driving?” Example of the drop-zone comparison of NSC lifetime motor-vehicle odds (1 in 101) with per-jump skydiving rates.
• NHTSA (2025). Traffic Safety Facts, 2023 Data: Passenger Vehicles, DOT HS 813 723. 23,959 passenger vehicle occupant deaths; occupant rates of 1.13 (cars) and 0.66 (light trucks) per 100 million VMT. The 0.83 combined figure is a site calculation (occupant deaths divided by the 89% of VMT driven by passenger vehicles).
• NHTSA (April 2025). Early Estimate of Motor Vehicle Traffic Fatalities in 2024, DOT HS 813 710. 39,345 deaths; 1.20 fatalities per 100 million vehicle miles traveled.
• Søreide, K., Ellingsen, C.L., Knutson, V. (2007). How dangerous is BASE jumping? An analysis of adverse events in 20,850 jumps from the Kjerag Massif, Norway. Journal of Trauma, 62(5), 1113-1117. 9 fatal and 82 nonfatal accidents, 1995-2005; one death per 2,317 jumps.
• Everest comparison: Huey et al. (2020), PLOS ONE, death rate near 1% per expedition for members. Lightning comparison: NWS, 18.6 US deaths per year, 2015-2024. Bee sting comparison: CDC, about 72 US deaths per year.