Colin Marston (Dstl) passed on to me some slides (public domain identifier PUB098428) presented at the recent MORS wargaming special meeting which address the range of wargaming approaches and methodology. Given the growing interest in wargaming—what it is, what it can do, and how it might do it—I thought they would be of interest to PAXsim readers. I’ve also inserted a few thoughts of my own.
You’ll find the full set of slides here here (ppt) and here (pdf).
The first set of slide suggests that wargames can be differentiated by the level of analysis (strategic vs operational, vs tactical), by the nature of the problem (bounded and clear, or wicked and messy), and by the type of adjudication used (open/free versus rigid and rules-based). I would have probably listed the adjudication issue last, because the choice of appropriate methodology can really only be made once you are clear on what sorts of question(s) you are trying to answer.
The slides don’t say much about purpose. Elsewhere, Graham Longley-Brown does so, noting the divide between analytical and training/education games:
While that differentiation is useful because it points to important differences in purpose and hence design, I’ll admit that I’ve been increasingly interested in the extent to which we might be able to develop hybrid games—that is, wargames that serve an education/training function, but in which participants are also generating data that is of analytical value too. My own Brynania civil war/peacebuilding game at McGill, for example, is designed for educational purposes but has now been used to generate data for two PhD theses (one on terrorist violence, the other on educational gaming). While there’s a risk of compromising analytical rigour or educational effectiveness in doing this, it could also provide a useful way of stretching limited resources.
The Dstl presentation goes on to discuss which game approaches are often of value in which contexts:
Here they comment:
On this slide the top blue line represents the different levels within the problem space. The red, middle line represents types of adjudication. The bottom green line indicates the different levels of complexity. On top of these axes we have the types of wargames that we employ in Dstl and across the MOD. Please note that these techniques are not limited to their positions on the axes. We find that the techniques on the left of the spectrum generally provide more opportunity for original thought and creativity (imagination). In addition, methods at this end of the spectrum generally provide an opportunity for doing lots of Courses of Action with little depth – so essentially short games that might last a couple of hours to a day. The methods on the right can provide increasing depth, but are often slower to set up and run. These methods generally employ more rigorous and precise techniques – although that does not necessarily mean that they give more accurate outputs. All of these approaches have their merits, some being better at trying to answer certain questions than others. So, when appropriate, we try to use a combination of different approaches.
They also identify some “essential elements” of a wargame:
Now, the type of game that we use is just one part of the process. This slide highlights the other factors that we need to consider. There’s no fixed order in the way we tackle these – it’s an iterative process and depends on the question.
The wargame is not the simulation. The simulation is but a subset of a wargame.
Effective communication and transparency are crucial throughout the whole of the wargaming process and it is crucial that everyone – from the players to the customers – are involved at the relevant stages.
The optimal approach to providing decision-support is often to fuse the information pertaining from both human-in-the-loop and non-human-in-the-loop techniques.
There are many different types of wargames and careful consideration should be given as to which type, or types, of game are most appropriate for a particular problem. Also wargaming should often NOT be used in isolation but as part of a broader analytical tool and / or iterative process that incorporates a range of different techniques.
Feel free to add your own thoughts in the comments section.