Wild Cards - Preparing for the Unpredictable

By: Karlheinz Steinmueller     Posted on: 01/03/2011     Last update: 2255 days ago by Mindcom


Wild Cards – Preparing for the Unpredictable

Karlheinz Steinmüller

Z_punkt The Foresight Company, Berlin/Germany


Wild Cards in the Framework of Risk Assessment and Horizon Scanning

In recent years, risk assessment and horizon scanning have gained increasing relevance for decision makers. This is not accidental. In a closely connected, utterly complex, globalized world, unpredictable change has become a constant phenomenon. Innovations drive economic development, they create opportunities and risks for economies, societies and the natural environment, and in most cases their impacts are at best uncertain. It may even be argued that all spheres of life are accelerated due to unceasing technological and social innovations. Under these conditions, conventional wisdom built on the experiences of former generations is often depreciated, likewise governance has become increasingly difficult - and risk management simply a necessity.

But there are risks and risks. Some risks are in a way well behaved, their probability can be calculated, their impacts are known. Other risks are elusive, they betray our efforts to calculate their probability, their impacts are mostly unknown, they are almost without precedence, and they hit us as a lightning bolt out of the blue. They are difficult to identify and even more difficult to manage, exotic and sometimes rather implausible as they seem to be. Such risks, which are termed wild cards, pose a challenge for horizon scanning and for risk assessment.


Wild Cards: Fundamentals

The concept of wild cards was introduced in 1992 with a joint study of the CIFS (Copenhagen Institute for Futures Studies, Denmark), BIPE Conseil (Issy-Les-Moulineaux, France) and the Institute for the Future (Menlo Park, California/USA). The three institutes proposed a definition that was focussed on business: “A wild card is a future development or event with a relatively low probability of occurrence but a likely high impact on the conduct of business.” (BIPE et al. 1992, p. v) Petersen extended it to social systems; his definition is now widely accepted: “A Wild Card is a low-probability, high impact event that is so large and/or arrives so fast that social systems are not able to effectively respond to it” (Petersen 2000)

The concept of wild cards rather quickly gained acceptance in the futurist community, since it was a much needed addition to the methodological toolbox, a concept to counterbalance the rather deterministic use of trends and crosscuts, a means to stimulate thinking out of the box, to challenge common wisdom and established assumptions about the future. Since the seminal work of 1992, several researchers have elaborated on the concept and developed wild card methodologies: Rockfellow (1994), Petersen (1997a, 1997b, 2000), Steinmüller (1997, 1999, 2004, 2007), Mendonça et al. (2004) – to mention only a few. Several catalogues of wild cards have been proposed since that time, e. g. by Peterson (1997b) and by Steinmüller & Steinmüller (2004).

As a rule, neither the likelihood nor the impact potential of a wild card is sufficiently known in advance; but both have to be crudely assessed when identifying an event as a wild card. With respect to probability, a qualitative evaluation will, as a rule, be sufficient, determining that the event in question is indeed rather improbable but not entirely impossible. In much the same way, a qualitative estimate of the impact is needed for the identification of wild cards: wide-ranging impacts should be expected. An in-depth evaluation of probabilities, of likely impacts and consequences usually takes place later (see below).

Hiltunen (2006) has recently argued that labeling wild cards as low probability events is misleading since in the end even the – apparently – most likely future may be in fact rather unlikely. One has to concede that the most plausible mainstream future is as a rule not very consistent, and that “objective” estimates of probabilities can not be achieved, not for wild cards and not for future scenarios in general. But low (a priori, perceived) probability and surprising character are interlinked characteristics of wild cards. Dropping one, implies dropping the other. With hindsight you can easily claim that a certain wild card simply had to happen and therefore was rather very likely, and that only deficiencies of perception led you to regard the event as improbable or even impossible. The main point on wild cards is however to take blind spots as granted – in order to overcome it at least partially.

At first glance a wild card is something surprising, perhaps even shocking, something which happens unexpectedly. Surprise is, however, even a still more subjective category than probability. It depends on one’s world view and is therefore not very well suited as a criterion. Nevertheless the question “What might surprise you?” is a good starting point for a wild card brainstorming session. The element of surprise frequently disappears during closer analysis.

In a way, the definition given above actually plays down the real value of the notion of wild cards. Characterizing them by low probability and high impact misses one central point: Wild cards are shocking not only because they have really large impacts on business or what else, they shock us since they do not fit into our usual frame of reference, they run counter our perception of the ordinary normal way things develop, they challenge the concepts through which we regard the world, perhaps even ridicule them.

Wild cards change our frame of reference, our mental map of the world. A point in case is the emergence of words with new meanings after a wild card has occurred: super-terrorism, climate protection, or – to take some older ones – HIV/AIDS, stagflation, and glocalisation. Therefore, wild cards do not only change reality but also, and perhaps even more deeply, they change our perception of reality and the concepts we apply to organize all the data about the world around us. It is often observed that wild cards force us to re-write the future, but this is only part of the truth. They entice us even to re-write the past. After a wild card occurred, we look with different eyes on past developments. Did they give rise to the wild card? What trends provided an environment favourable for the wild card? Which weak signals already hinted at it?

Take the Chernobyl disaster as an example. This catastrophe was not only one more reactor accident (of a beforehand unknown dimension), nothing like, say, the disaster of Harrisburgh (Three Miles Island). It was without any precedent for a second reason: it changed the way most people now think about the “peaceful use of the atom”. – If the future is the space of our hopes and fears, our wishes and plans, or, more generally, our expectations, wild cards are shocks to this space. They are “futurequakes” changing all of the landscape of the future.


A Short Typology

There are different types of wild cards: A break-through in high-temperature superconductivity has other causes and other effects than a radio-smog panic, a new pandemic disease evolves in other ways than a disruption of the Gulf Stream, political upheavals follow other patterns than the outbreak of a GMO. Different aspects may be used to systematize wild cards (Petersen 1997b, Mendonça et al. 2004, Steinmüller & Steinmüller 2004).

1)      Topic: The subject of the wild card, e. g. the STEEP sector in which the wild card originates, or upon which it has direct impact. We have to distinguish technological wild cards like fusion power or robots that become self-conscious from political wild cards like terrorist attacks, revolts or assassinations.

2)      Reach: Is the impact of a wild card restricted to one specific sphere of life, one industry, one region or not? We have to distinguish global wild cards – e. g. an impact of an asteroid – from regional ones like unexpected aspects of climate change in a territory, industry specific ones like the identification of new hazardous substances (after the pattern of the asbestos crisis) and wild cards affecting all of a national economy.

3)      Plausibility: Wild cards are by definition rather unlikely but we have to assess whether a wild card is highly improbable or simply not very probable. Another distinction carries greater psychological weight: some wild cards are plausible because they fit – although perhaps only after a preliminary analysis – like most natural disasters into our worldview. Other wild cards are not plausible; they go against intuition and common sense, without, however, being absolutely impossible. Seen from a methodological perspective, it might make sense to take even “impossible” wild cards into account because the demarcation line (often fuzzy at best) between the possible and the impossible is based on the knowledge available at the moment and on one’s personal view of the world.

4)      Time scale: One can also distinguish wild cards which have immediate impacts like most disasters and wild cards that have a certain gestation time and influence medium-term or longer-term developments like scientific breakthroughs.


Wild Cards, Trends and Weak Signals

Wild cards erupt suddenly, but like a lightning bolt they are not without causes. Frequently, wild cards result from trends, especially from trends with a non-linear characteristic (and trends are rarely linear). Possible catastrophes resulting from global warming are a point in case. If the temperature rise surpasses a certain point – the “tipping point” – global atmospheric or maritime circulation pattern may suddenly change with tremendous impacts on the whole climate system (e. g. new El Niño phenomena). The gradual, frequently unnoticed processes that can give rise to wild cards are sometimes termed “creeping catastrophes” (Böhret 1990) in contrast to acute, catastrophic events. After the event one may ask with Hiltunen (2006): “Was it a wild card or just our blindness to gradual change?”

Perhaps more frequently, wild cards are caused by the interaction of trends (crosscuts). If it is difficult to predict the future evolution and possible impacts of a trend, it is even more difficult to forecast the outcome of complex trend interactions. In most cases, no sophisticated models are available and trend interaction methods usually rely on an intuitive understanding of interactions – which is prone to blind spots or wishful thinking.

On the other hand, wild cards may wreck a trend, or turn it in another direction, or put a temporary halt to it, or give rise to completely new trends. After Chernobyl, trends in the construction of nuclear power plants were disrupted. After 9/11, markets for biometrical and surveillance technologies exploded. In some cases, a wild card produces a real torrent of secondary wild cards – like the break up of the Soviet Union and resulting wars in the Caucasian region.

Another interesting point is the relation of wild cards and weak signals. Sometimes the two concepts are confused, as Hiltunen (2006) noticed. It has to be stressed, that weak signals are what their name implies: signals, i. e. signs, symptoms, indications, not more. They hint at something – be it a new trend, a new issue, or an “approaching” wild card. Weak signals therefore may herald the advent of a wild card, or make them more plausible or help us to assess their probability and impact. Wild cards in difference stand for themselves as factors shaping our future. But naturally, even an event with severe impacts can be interpreted as sign of something else, perhaps still bigger. So the US sub-prime loan crisis was a strong hint to a possible earthquake of the whole global financial system. It depends therefore on the frame of reference whether you interpret an event as a wild card (since it occurred surprisingly and with important impact) or as a signal (since it might be a precursor of still more powerful events). The term “weak signal”, however, should be avoided: Wild cards are “strong” by definition.


The Life Cycle of a Wild Card

Wild cards evolve in a specific way, follow what could be termed a life-cycle. For a while, they prepare in a hidden, latent form. Then, suddenly, they erupt, become manifest. After that, one is confronted by their impacts. In principle, these three stages form the life cycle.

1)      Latency: During that period, the wild card “grows” in the hidden, be it due to some trend, some crosscut, to human plans and preparations for their execution or otherwise. In that stage, the wild card may produce first weak signals, but usually these are either not detected or disregarded because of the “noise” of other weak signals or of an unwillingness to receive their message. Sometimes a specific group – scientists, artists – are already discussing the wild card, but the general public, opinion leaders, managers or politicians do not take it serious.

2)      Eruption: The wild card happens, suddenly manifests in a disruptive event. If it is seen as a signal, than it is a really, really very strong one. Now, everybody is talking about it: How could it happen? Why is nobody prepared? Who is to blame for it? What are the consequences? As the general public, decision makers in government or business are taken by surprise. Frequently, e. g. in case of industrial disasters, there is a short period of denial, followed by non-systematic, inadequate and inappropriate reactions which are primarily “for show”. Overreactions, panics or hypes are common, controversial interpretations are given (e. g. conspiracy theories).

3)      Normalisation: For a while, the impacts of the wild card spread through the system in question producing shocks of second and higher order. Measures taken become (hopefully) more and more adequate. With time going by, decision makers and the general public get used to the new situation. Most stocks – to take an example – come again close to pre-event values. As the wild card is integrated into the common worldview, a standard interpretation is generally accepted, mostly with a paradigm change with respect to issues close to the wild card.

In some respects, the concept of wild cards is a counterpart to the concept of chaos in the theory of dynamic systems. Like chaos, wild cards place limits on both forecasting and planning. Like chaos, they are the result of the inherent complexity of the system being analyzed and of its environment. Like the bifurcations in the chaos theory, they mark the beginning of new developments, diverging evolutionary paths.

One lesson of chaos theory is that non-linearity can lead to counter-intuitive system behaviour. Therefore, a basic rule for futures studies is not to depend on intuitively convincing and plausible theories, but rather to take counter-intuitive system behaviour into account. This is possible by supplementing a study with a wild card analysis.


Towards Wild Card Management

At first, wild cards were introduced within the framework of scenario studies. As one of the steps of a scenario process, wild card analysis is used to test the stability of scenarios with respect to external shocks or internal disruptive factors which had been neglected or disregarded before. More generally, wild cards can fulfil several functions in a scenario process:

·        They can, as mentioned, be used in order to estimate the resilience of a scenario to external disruptions.

·        They can be used to compensate for potential weak points in the conceptual framework (mental map).

·        They can help the team who constructs the scenarios to recognize alternatives and to be open-minded to unexpected developments.

·        Ultimately, they can also be used to counteract certain widespread faults – such as a shortage of imaginative capacity, the predominance of wishful thinking or a fixation on catastrophic scenarios (“hyper worst case thinking”).

Apart from that, wild cards are sometimes used directly to test the robustness of a strategy or to increase the risk awareness of decision makers. Similar to risk management, wild card management can be broken down to several steps: starting with the identification of potential wild cards and an assessment of their probability and impact and leading finally to (counter-) measures.

·        Bounding: In the first step, the scope of the analysis is defined.

·        Identification: Within the second step, wild cards are identified and selected according to specific criteria (relevance for the problem investigated etc.). As result a portfolio of wild cards is established.

·        Evaluation: After that necessary prerequisites, favourable and unfavourable conditions for the occurrence of the selected wild cards are assessed as well as possible impacts (e. g. along the “Arlington Impact Index”, a set of criteria given by Petersen 1997b)

·        Strategy Building: Finding strategies to minimise risks and to seize opportunities is the next step. It includes usually questions like: How can we prepare, either by preventing the wild card, by taking advantage of its occurrence or by alleviating its impact? Are there weak signals that foreshadow the occurrence of the wild cards?

·        Implementation: In the end, measures have to be implemented and strategies have to be communicated to relevant groups of persons.

Quite generally, wild card management implies thinking in advance. Sometimes wild card studies lead to establishing an early warning system (Nikander 2002) that systematically identifies weak signals for a specific group of wild cards. It is, of course, impossible to prepare for every conceivable wild card. But the discussion of wild cards in the course of decision-making can reduce the element of surprise when a real wild card does occur and can in general increase flexibility of response.


Some Practical Points

What are the criteria for selecting a suitable set of wild cards? There is no all-embracing answer to this question, and one can indicate only a few general rules based on experience. Firstly, the wild card must be appropriate to the problem. A wild card need not necessarily stem from the central topical area of the study, but it should nonetheless be associated with it. Wild cards that would be entirely without consequences will not help to detect any additional information. Secondly, a wild card should be as original as possible, should be something which has not already been taken into account in another form; its consequences should not be immediately apparent. Thirdly, one should also think about wild cards that (in accordance with conventional thinking) are at the far edge of that which is just barely possible.

One can add some simple rules:

·        The analysis should not be limited to a too small number of wild cards. Taking only three or four would imply that these (rather arbitrarily chosen) events receive too much attention. The quality as well as the plausibility of the study could suffer. A sufficient number of sufficiently distinct wild cards is a condition of success. As a rule one can find to every trend some wild cards…

·        “Negative” wild cards, those that presumably would not support the scenario constructed, but would rather undermine it, should be given priority consideration (as a test for the stability of the scenario). A closer analysis may, however, prove that “supportive” wild cards can also have interesting counter-intuitive consequences. It makes therefore sense not only to look for “weird” risks, but also for unusual opportunities.

·        In addition to wild cards with a strong contextual reference to the problems studied, it is also advisable to consider some wild cards that could cause disruptions in the external environment of the system in question.

·        In order to avoid potential prejudices, it may be useful – especially when identifying wild cards – to incorporate outside expertise into the study, either through interviews or by means of a workshop. An internet forum or even a competition may be useful in collecting a large number of potential wild cards (Rodenhäuser/Steinmüller 2006).



Foresight has experienced a number of remarkable developments in the last decades. Futurists have moved away from the planning optimism that characterized their earlier days and from far-reaching forecasts of the future. Thirty years ago, Herman Kahn sought to describe the coming two hundred years; by contrast, futures research today seeks to identify in a quite pragmatic fashion feasible roads into a future that will be worth living. But futurists have learned yet another lesson: There is always a chance that dramatic events change the whole map of the future, they way we think about it, the concepts we use and even the aims we try to achieve.

Each individual wild card has a low probability, but there is a large number of possible wild cards and as time goes by this number is increasing – the trends we established become more and more uncertain, their interactions less predictable, human actors pursue new plans, and each of these uncertainties, interactions, new lines of action can give raise to wild cards. If we look into the far future, the combined probability of all wild cards tends to 1. This implies that the probability of realization of the standard or mainstream scenario approaches zero. In the long run, wild cards will shape the future.

Integrating wild cards in risk assessment can not only provide us with insights into possible disruptive changes, it can as well lead to a better understanding of existing trends and challenges, their preconditions and delimitations. Wild cards open up new perspectives and provoke fresh thinking about future options, strategies and measures. They are a new, but powerful concept that can help horizon scanning to go beyond a perhaps too narrow horizon.



BIPE Conseil,  Copenhagen Institute for Futures Studies,  Institute for the Future 1992. Wild Cards: A Multinational Perspective. Palo Alto/CA.: Institute for the Future

Böhret, Carl 1990. Folgen. Entwurf für eine aktive Politik gegen schleichende Katastrophen, Opladen: Leske und Budrich

Hiltunen, Elina 2006. Was It a Wild Card or Just Our Blindness to Gradual Change? -  Journal of Futures Studies, Vol. 11 Issue 2, pp. 61-74

Mendonça, Sandro,  Pina e Cunha, Miguel,  Kaivo-oja, Jari  and Ruff, Frank. Wild cards, weak signals and organisational improvisation -  Futures, Vol. 36, Issue 2, March 2004, pp. 201-218

Nikander, Ilmari O. 2002. Early Warnings. A Phenomenon in Project Management, Dissertation Helsinki University of Technology

Petersen, John L. 1997a. The ‘Wild Cards’ in Our Future: Preparing for the Improbable. -  The Futurist, July-August 1997, pp. 43-47

Petersen, John L. 1997b. Out of the Blue, Wild Cards and Other Big Surprises. Washington: The Arlington Institute

Petersen, John L. 2000. Out of the Blue, Blue How to Anticipate Big Future Surprises. Lanham: Madison Books, 2000

Rockfellow, John D. 1994. Wild Cards - Preparing for ‘The Big One’ -  The Futurist, Jan-Feb 1994, pp. 14-19

Rodenhäuser, Ben  and Steinmüller, Karlheinz 2006. Lass dich überraschen. Über das Management von Wild Cards -  Research & Results, no. 3-2006, pp. 34f

Steinmüller, Angela and Karlheinz 2004. Wild Cards. Wenn das Unwahrscheinliche eintritt. Hamburg: Murmann

Steinmüller, Karlheinz 1997. The Future as Wild Card - Towards a Methodology of the Unpredictable. -  Kreibich, Rolf,  Steinmüller, Karlheinz  and Zöpel, Christoph (eds.). Beyond 2000. Challenges for Futures Studies. Documentation of the SFZ Summer Academy 1996, SFZ-WerkstattBericht 20, Gelsenkirchen 1997, pp. 98 - 104

Steinmüller, Karlheinz 1999. Wild Cards. Eine neue Entwicklung in der internationalen Zukunftsforschung -  Zukünfte no. 27, pp. 51-53

Steinmüller, Karlheinz 2004. The future as Wild Card. A short introduction to a new concept -  Brockett, Susan and Dahlström, Margareta (eds.). Spatial Development Trends. Nordic Countries in a European Context. Stockholm: Nordregio 2004 (Nordregio Report 2004: 6), pp. 193 - 202

Steinmüller, Karlheinz: Thinking Out of the Box. Weak Signals and Wild Cards for European Regions -  Futura no 2/2007, pp 22-29

van Notten, Ph. W. F.,  Sleegers, A. M.  and van Asselt, M.B.A. 2005. The Future shocks: On Discontinuity and Scenario Development. -  Technological Forecasting and Social Change. no. 72, pp. 175-194

, , , ,