The safety mindset

Ask a new student where paragliding gets dangerous and they usually point at the sky, high above the hill. The accident reports point somewhere else: at the ground. Most reported accidents happen during the launch, just after it, on the approach and in the landing itself.

That is good news, in a way. Those are the moments you practise most in training, and most accidents grow from small, ordinary mistakes you can learn to catch. This lesson is about the habits of mind that catch them.

Where accidents happen

In the German federation’s long-term average, about two in three reported accidents happen in the four phases near the ground: the launch, the seconds just after lift-off, the approach and the landing¹. Austria’s 206 hang-gliding and paragliding occurrences in 2024 show the same 64%². Landing is usually the largest single phase, in European data and in older US reports alike³⁴.

Figure: Accidents reported by German pilots, by flight phase (DHV long-term average). Launch and departure make up 30% and approach and landing 34%, together about two thirds.

Why near the ground? There is little height to recover and plenty to hit. Launch and landing are also the busiest moments, with the most decisions in the least time, and once your feet leave the ground you can no longer abort⁵. That is why training spends so long on ground handling, aborts and the landing approach⁵⁶.

What the numbers can’t tell you

Read accident statistics with care.

• They are shares, not rates. The bars show where reported accidents happen, not how likely an accident is on a flight; only the Portsmouth study below gives a rate. Launch and landing last only moments, so minute for minute they are probably even more demanding than the bars suggest¹.
• Many injuries go unreported. Swiss statistics count just under 1,000 people injured paragliding and hang gliding each year, but only about 100 reports reach the Swiss federation⁷.
• Deaths follow a different pattern. Of 48 German paragliding deaths from 2018 to 2022, nearly half happened in flight, while thermalling, gliding or ridge soaring⁸. Across European data, fatal accidents are mostly loss of control after a collapse in turbulent or thermic air, and they often involve experienced pilots⁹. The high part of a flight is not the safe part; it only gives you more time.

So how risky is paragliding overall? A University of Portsmouth study combined a survey of pilots’ flights in 2019 with a UK incident database. For UK pilots it estimated about 1.4 deaths and 20 serious injuries per 100,000 flights, roughly twice the risk of general aviation and skydiving¹⁰. That risk is real, and much of it is in your hands: equipment failure was involved in only 0.6% of incidents, and the rest came down to control or decision errors¹¹.

How small errors line up

Accidents rarely have one big cause. Reviews of serious incidents keep finding a pile-up of small factors: the wind a little off the hill, a bit stronger than expected, a rough site, a tired or stressed pilot¹². Each alone would probably have been fine; together they were not.

A common way to picture this is the Swiss-cheese model. Each slice is a layer of defence, such as a sound go/no-go decision, calm preparation, a full check or the willingness to abort. Every layer has holes, because people are human. An accident happens when the holes line up.

Figure: A chain that keeps appearing in accident reports: after an aborted launch the conditions worsen, the pilot hurries, the repeat check is skipped, and they launch with a knot in the lines. Closing any one hole would have stopped it.

Many holes are opened by distraction: chat from other pilots, questions from spectators, someone filming you. In a group, a buddy check adds another layer, but it never replaces your own¹⁴. Any one intact layer can stop the chain, so keep every layer as solid as you can: you never know which one will be the last. Start the check again from the top after any interruption or aborted launch¹³. Abort if the wing is not right; an abort is good airmanship, not a failure⁵.

Holes also grow slowly. A shortcut that worked a few times starts to feel normal; this drift is called normalisation of deviance, and it sounds like “it always worked before”¹⁵. Hazardous attitudes and human factors goes deeper.

The decision is made on the ground

Once your feet leave the ground, your options narrow and the time to decide shrinks. You can still land early or turn away from trouble, but the big decisions, whether to fly at all, whether your equipment is right, where you will land, belong on the ground, where you can take your time and change your mind.

Pilots have an old rule for this: a doubt is a no. The German federation says most accidents could be prevented by keeping to it, and the Swiss federation puts it simply: if your gut says no, listen¹⁶. During training your instructor makes the go/no-go call. Your part is honesty: say when you are tired, unwell or not ready, and when something about your kit looks wrong.

Pressure is the enemy of this routine: a long drive, friends already in the air, spectators at launch, a group chat buzzing about the day¹⁹. Go/no-go and personal minimums shows how to set limits before the pressure arrives.

Currency: skills fade

Skills and judgement rust without practice; staying in practice is what pilots call currency. A BHPA analysis found that pilots who fly more are less likely to have an emergency, and handle one better²⁰. If weeks pass between training days, expect to repeat a step before you move on, and re-read the lessons first. Accidents, currency and the logbook covers returning after a break.

The conservative pilot

The German federation calls over-confidence the greatest danger: a little-trained beginner, or a pilot who rarely flies, in thermals or wind too strong for them²¹. SafePro warns that once pilots can fly actively, some start to believe they have no limits, and that accidents also happen when pilots step up to a new stage²².

The conservative pilot is not timid. They fly a lot, on good days, within their limits, and are still flying years later. Staying on the ground on a doubtful day is a good decision, not a timid one.