Skip to content

Cookies 🍪

This site uses cookies that need consent.

Learn more

Zur Powderguide-Startseite Zur Powderguide-Startseite
mountain knowledge

The pitfalls of statistics: are only other winter sports enthusiasts really killed by avalanches?

Summary of a study by Krister Kristensen, Manuel Genswein and Werner Munter

by Tobias Kurzeder 12/27/2013
The authors attempt to quantify the statistical risk of being killed by an avalanche in the course of a winter sports enthusiast's lifetime. The starting point of their study is the problem that, despite greatly improved avalanche training and risk management methods, well-trained and experienced winter sports enthusiasts are repeatedly killed in self-inflicted avalanche accidents.

The authors attempt to quantify the statistical risk of being killed by an avalanche in the course of a winter sports enthusiast's lifetime. The starting point of their study is the problem that, despite greatly improved avalanche training and risk management methods, well-trained and experienced winter sports enthusiasts are repeatedly killed in self-inflicted avalanche accidents. Why do athletes who should know exactly what they are doing take such high risks? Part of the answer lies in the fact that the risk of dying on a potential avalanche slope as a result of a single wrong decision is very low. However, if an accident occurs, the consequences are often fatal. If you add up the residual risks of all the ski touring and freeriding days in a winter sports enthusiast's life, the probability of death suddenly becomes frighteningly high. However, this cumulative risk is recognized by very few winter sports enthusiasts and suppressed by most.

Why do we keep putting ourselves in great danger despite our knowledge of the potentially fatal consequences?

In recent years and decades, the number of avalanche accidents has fallen significantly in relation to the increasing number of winter sports enthusiasts in open ski areas. However, despite increasing awareness of the danger of avalanches, accident patterns have remained constant: A striking number of well-trained winter sports enthusiasts die in avalanches. Kristensen, Genswein and Munter investigate the question of why ski tours and freeride ventures repeatedly end in avalanche accidents, even though the winter sports enthusiasts were able to correctly assess the avalanche risk thanks to their expertise. Mere coincidence or bad luck and residual risk cannot sufficiently explain the comparatively high number of accidents in obviously dangerous situations.

I am also part of the statistics - the key to further reducing avalanche accidents

Kristensen, Genswein and Munter argue that the key to further reducing fatal avalanche accidents does not lie in improving knowledge about the risk of avalanches. Instead, they call for an increase in winter sports enthusiasts' awareness of how high their personal individual risk of death is.

If people were to make exclusively or predominantly rational decisions, then the current accident statistics (for experienced winter sports enthusiasts) could be seen as the level of risk that ski tourers and freeriders are prepared to take: Avalanche deaths would then be the consciously accepted price for the freedom to move around in the mountains on one's own responsibility. The authors put forward the theory that many winter sports enthusiasts take a very high risk of death without realizing how high the probability of dying in an avalanche is. People can achieve (almost) nothing in life without taking risks. In theory, we therefore fall back on the ideal figure of the "rational agent". This agent is prepared to take risks if the expected benefit exceeds the risks and the greater the expected gain, the greater the willingness to take high risks. Psychological studies show that risk perception is influenced more by experience and emotions than by knowledge of (accident) probabilities. A decisive problem in risk perception appears to lie in the frequent lack of ability to recognize the abstract, statistical risk of death as a possible consequence of one's own actions and decisions. People are always particularly bad at assessing probabilities when they receive little or no feedback. And this is the kind of situation we are dealing with when making decisions in avalanche terrain: even in critical situations and when making the wrong decisions, avalanches rarely occur. We humans think of probabilities in such a way that they are the quantity or relative frequency of events over long periods of time. Formal, statistical decision-making methods, on the other hand, are a very new tool and run counter to this sense of probability. One problem with statistical statements about accident or even death probabilities is that the statements are necessarily highly abstract: This means that when you read statistics, you almost always assume that others are or will be affected, but that you are not. And so the advantage of statistics becomes their disadvantage: due to the abstraction, they lose part of their "perceived" reality.

Lifetime Risk

In "3x3 Avalanches", Werner Munter investigated the probability of death in various winter sports. According to Munter's assessment, one fatal accident occurred on 36,000 ski tours in Switzerland in the 1980s. He considered such a risk to be far too high and instead postulated his risk threshold of 1:100,000, which implies one avalanche fatality per 100,000 ski touring days. This threshold value is expressed in Munter's reduction method as residual risk 1 or RM 1. If the fatal avalanche accidents of the 1980s are calculated using the formula of the reduction method, they occurred with an average residual risk value of 2.2. Werner Munter also advocates the so-called "limits": these limits are intended to avoid residual risks of RM 2 (1 fatality per 50,000 tours) and higher. The American avalanche expert Bruce Tremper calls this limit the "Stupid Line". The extreme risk limit 4 used in the table corresponds to the average residual risk that was taken in the particularly serious accidents (with at least 5 fatalities) in the 1980s. RM 4 means 1 fatality per 25,000 ski tours. Munters recommendation, on the other hand, is to stay below or around the risk limit of 1 and to accept the limits (RM 2) as an upper limit. However, these limits should only be accepted in special situations and circumstances. The RM 1 limit value includes an individual risk of 1:50,000 to 1:200,000 due to the statistical spread. The fact that the RM 1 residual risk value of the reduction method appears to be a limit value accepted by most winter sports enthusiasts (at least in the touring sector) is supported by a long-term study by the Summit Club (DAV): The average risk taken by mountain guides during 5000 guided tour days corresponded to the residual risk RM 0.8. As it can be assumed that most winter sports enthusiasts found these tours to be enjoyable mountain experiences, this result supports the recommendation of the risk limit of RM 1, which is shown as the green area in the diagram.

Most trained winter sports enthusiasts and especially group leaders or mountain guides are now able to recognize problematic decision traps. However, understanding and accepting what the statistical probability of fatal accidents means for the individual still seems very difficult for many winter sports enthusiasts. The problem with making decisions in avalanche terrain is still that the probability of triggering an avalanche is low, but the consequences of a possible avalanche accident are dramatically high. To address this problem, the authors call for the development of training methods that provide immediate feedback on whether decisions are right or wrong and how high the risk is. This could help to prevent winter sports enthusiasts from the particularly risk-averse frequent skier category from practising their sport with particularly high probabilities of death without realizing it (see Table 1). In the medium term, this could also lead to a further reduction in the individual probability of death when ski touring and freeriding.

"Code of Honor" - Recommendation to reduce the likelihood of fatal accidents

Elementary precautions:- Always carry full avalanche emergency equipment
- Pay attention to alarm signs: Whamming noises, fresh avalanches, remote triggers are criteria for aborting the tour
- Maintain safe distances if there is any doubt about the stability of the snowpack. Accept the upper risk limit "Limits":- In case of high avalanche danger (warning level 4), ski on terrain that is less than 30 degrees steep.
- In case of considerable avalanche danger (warning level 3), ski on terrain that is less than 40 degrees steep.
- In case of moderate avalanche danger (warning level 2), avoid shaded slopes that are 40 degrees or steeper if they are untracked.

Here you can discuss the article or the study... Here you can download the original article...

ℹ️PowderGuide.com is nonprofit-making, so we are glad about any support. If you like to improve our DeepL translation backend, feel free to write an email to the editors with your suggestions for better understandings. Thanks a lot in advance!

Show original (German) Show original (French)

Comments

mountain knowledge
presented by