Conflict simulation, peacebuilding, and development

Connections US 2018 report


This year’s Connections US professional wargaming conference—the 26th since the series began—was held at National Defense University on July 17-20. It was the largest meeting yet, with some 280 or so registered, and around 210 attendees. Amongst those registered for the conference were several students (Keiko Ivinson, Kia Kouyoumjian, Jason Li, Caroline  Wesley) from my game design seminar at McGill University this past term, and some of them will be posting their own perceptions to PAXsims in the newr future.

The conference started off with a welcome from Vice Admiral Fritz Roegge, the President of National Defense University. He discussed the use of gaming for education and outreach at NDU, notably through the activities of the Center for Applied Strategic Learning. Later, COL Voris W. McBurnette, the Director of CASL, also welcomed the group and said a little more about the work of the Center.

This was followed with a general presentation by Matt Caffrey (Air Force Research Laboratory) on wargaming. He argued for the utility of wargaming, offering a brief overview of defence wargaming in the US and elsewhere and historical examples of when and where wargaming had made a clear difference. He then went on to define key elements of wargaming.

The conference next split into a choice of conference sessions—I attended a session on wargaming counterinsurgency, by Brian Train (slides). Brian noted the relative paucity of commercial wargame designs exploring insurgency and irregular warfare, despite this being the most common form of armed conflict since WWII. He also noted the ways in which counterinsurgency game designs typically differ from wargames about more conventional kinetic operations.

“Conventional” wargame “Irregular” wargame
binary, zero-sum opposition multiple though frictional points of view/ factions
ordered situations chaotic and nonlinear game states
rigid treatments of time, space and force flexible, malleable scales, non-representational units
symmetry of information, methods and objectives asymmetry of these

After lunch, fellow PAXsims coeditor Tom Mouat (Defence Academy of the UK) and I offered a seminar session on matrix gaming (slides). We provided an overview of the technique and had two games set up (ISIS CRISIS and HIGH NORTH) to illustrate game processes and take a few sample actions.

The rest of the afternoon was devoted to game demonstrations. In addition to having AFTERSHOCK: A Humanitarian Crisis Game and the Matrix Game Construction Kit on display, Brian and I also showed off We Are Coming, Nineveh!, an operational-level game of the 2017 battle for West Mosul, which we are developing with my former students Juliette Le Ménahèze and Harrison Brewer.

Day 2 started with a keynote presentation by John Lillard on US naval wargaming during the interwar period, drawing upon his book on the topic. He highlighted how lessons were learned from both regular wargaming at the Naval War College and the fleet exercises. Games were either STRAT (chart), TAC (board), or QR (quick reactions). Games either focused against a superior (Red/UK) in opponent in the Atlantic, or a weaker (Orange/Japan) opponent in the Pacific. The opposing side was always played by students, as well as Blue. John took us through a series of game vignettes from the 1920s and 1930s, showing how the games addressed new issues and innovations, including air and submarine assets, amphibious operations, logistics, and allied operations. The games included such experiments as different approach routes, formations, amphibious operations, anti-submarine escorts, airships, chemical warfare, floating drydocks, and converted (cruisers, merchantmen) aircraft carriers.


John Lillard discussing interwar wargaming at the US Naval War College.

After a coffee break, attention turned to game design. Goor Tsalalyachin (Dado Center, IDF General Staff) offered an overview of strategic wargaming in the Israeli Defence Force. He emphasized the need to move beyond simplistic, two-dimensional thought. In the IDF, wargames are used for critical decision support and knowledge development, as well as training—often involving personnel at the most senior levels. He highlighted the dynamic and changing nature of the strategic environment (which demands a constantly agile conceptual/operational design process), and the value of wargames in developing and challenging new ideas and concepts. The method he described was largely that of a two-move seminar game. In many cases, I suspect, this produces structured scenario discussion rather than a great deal of iterative gaming.

Becca Wasser (RAND/Georgetown) and Stacie Pettyjohn (RAND/Johns Hopkins) addressed this issue in their presentation on “Beyond the BOGSAT: the case for structured strategic games.” Becca argued that despite the reinvigoration of wargaming, POL-MIL gaming has somewhat languished, in part due to the “squishiness” of the political dimension. Given this, it could be useful to add greater structure to such games by drawing from design features of some commercial games. Unstructured games can tend to opinion-discussion, rather than making choices with consequences. Stacie suggested that too many designers default to free-form seminar games, which end up being non-game BOGSATs or low-quality games. The latter involve unfocused team deliberations, hasty decisions, and an abundance of semi-relevant background materials that are often ignored by players.  Game validity excessively depends on ad hoc, undocumented “expert” adjudication. Seminar games are thus often overly dependent on players and experts, adjudication is based on unexpressed, undocumented mental models, and games are difficult to replicate. They also produce too few innovative ideas.

Stacie suggested drawing upon commercial game mechanisms, pointing to Brian Train’s game designs and PAXsims’ very own AFTERSHOCK as good examples, as well as the Countering ISIL game developed by RAND (based on a design first prototyped at a MORS wargaming workshop). There are, she noted, challenges to developing such games, such as the up-front challenge of producing game models and rules. Such games often do not scale well to large audiences. They might also face problems of acceptability (with strategic boardgames looking too “gamey”).


Becca and Stacey discuss adding more structure to strategic games.

Frans Kleyheeg (TNO) presented on “gaming and VR technology as a game changer,” noting the utility of virtual simulation for both wargaming and design development/experimentation.

Subsequent discussion explored issues of communication and information in games, the use of formal planning tools, and what to do about indecisive players.

After lunch it was time for the Connections Game Lab. Essentially, this consisted of a series of topical issues, questions, or design challenges which had been identified based on input before the conference. Each of these was assigned a table and facilitator where they could be discussed more fully. I led a session on gaming unpredictable adversaries (and allies), which touched on the sources of unpredictability (is the behaviour genuinely pathological and irrational, or does apparent unpredictability simply stem from inadequate information and models?), as well as why and how this might be embedded in a game design. My thoughts on the topic can be found in an earlier piece at The Strategy Bridge, as well as my presentation last year at Dstl.

I also attended a very useful session on wargaming mass atrocity, where we discussed how and why this might be undertaken. I mentioned the work that Kia and Keiko are doing on gaming the Darfur conflict, the purpose of which is to teach about the logic of mass atrocity as well as possible mechanisms for mitigation and prevention. A session later in the conference focused on wargaming and the analysis of human decision-making. Participants noted that not only can games be used to examine human factors and decision-making dynamics, but also that understanding (and manipulation of these through structure, information, and player engagement) is essential to good game design.

Back to panel presentations in the main auditorium, Hyong Lee (NDU) talked about “educating future cyber strategists through wargaming.” He talked about several different cyber games they have used at NDU and elsewhere. He identified several challenges: an excessive sponsor focus on technical solutions (when there may be useful no tech/low tech solutions); incorporating technology issues at a strategic level; gaming the non-technical effects of cyberattacks (political impact, etc.); and gaming the “shaping” (left-of-boom) phase of cyber.

Phillip Reiman (USAF) and Colby Sullins (USAF) discussed the training of attorneys in strategic decision-making through gaming. Rush to Judgment is a boardgame they developed to train Air Force lawyers on moving cases through to resolution using the Armed Services Board of Contract Appeals. They found the game improved knowledge but also underscored to participants the challenges involved in dispute resolution.


Who knew that military contracting dispute resolution could be so fun?

Finally, Patrick Schoof (US Army National Simulation Centre) presented on “Battle for Atropia: Army Division Level Wargaming.” Battle for Atropia is a 45 minute-hex-based wargame inspired by Battle for Moscow. The design for this then sparked a request for an enlarged and revised version, Land Power, for classroom use. The game uses 6 hour turns, with command points limiting activity, and a card system for enablers and supports.

Asked what surprises they had encountered, panelists noted the importance of playtesting, and the surprising directions players sometimes take a game (especially in a matrix game). One audience question raised the issue of information flows in games. There was also discussion of assessing educational utility.

The evening was Connections game night, with an opportunity to play in a variety of games. I ran a game of Reckoning of Vultures, from the Matrix Game Construction Kit.

Thursday began with a panel on “perspectives and tools.” My presentation was the first up, on “In the Eye of the Beholder? Cognitive Challenges in Wargame Analysis” (slides), based on the DIRE STRAITS experiment at the Connections UK 2017 wargaming conference. In this, we asked three separate analytical teams to provide independent assessments of game methodology and findings. Their finds were quite divergent—suggesting that wargame analysis might depend as much on the analysts as the game. I went on to suggest ways of addressing this, including red teaming game analysis and attention to cognitive bias training.

This slideshow requires JavaScript.

David Ross (US Air Force Research Lab) discussed the challenges of effectively integrating emerging technologies into wargames He noted that it is not enough to simply present the technology to participants—it helps to also highlight how it operates as a system, interacts with other systems, and to suggest some preliminary concept of employment.

Ben Connable (RAND) made an excellent presentation on “The Will to Fight: Adding Brutal Realism to the Military’s Games and Simulations.” The emphasis here was not on immersive realism, but rather the modelling of willpower, morale, and cohesion effects. He noted that commercial/hobby wargames generally do a much better job of representing morale effects on battle performance than military wargames, where this key psychological aspect is often completely ignored—resulting in units fighting to the death or failing to react to battlefield events or context. Together with colleagues at RAND, they are in the process of developing models that can represent willpower effects in both strategic and tactical/operational levels. He mentioned the Close Combat series of the digital games as one of their inspirations for their own modelling. In later comments I also pointed to This War of Mine as doing a superb job of representing morale and psychological factors, albeit in a game about civilian survival rather than war-fighting.

After the morning coffee break, attention turned to the importance of narrative in wargame. Anja van der Hulst (TNO) made a presentation on wargaming hybrid warfare, in which she warned about the excessive assumptions of rationality and strategic behavior in most games. Anja noted the importance of grievance in conflict and emphasized that this has an emotional as well as rational or material component. Emotional reactions tend to be less rational/strategic, and anger and humiliation can be a powerful driver of behavior. She discussed how she had explored such issues in a variety of game settings, including a modified version of the Baltic Challenge matrix game.

J. Furman Daniel III (Embry-Riddle Aeronautical University) presented on “fiction as a wargame.” His research (with Paul Musgrave) found that fiction has measurable impact on policymakers and hence policy. He explained this, in part , with reference to Peter Perla and ED McGrady’s narrative-based argument on why wargaming works. He also suggested that fiction can help wargamers better design games that engage participants, and is a useful way of encouraging reflection and analysis.

John Derosa (George Mason) and Lauren Kinney (George Mason) explored narrative analysis of wargaming, based on an experiment that they conducted at the 83rdMORS Symposium, where they performed actant analysis of three parallel games of Drive on Metz. Their findings focused such issues as tactical vs strategic preoccupations, more active or passive approaches to the enemy, and differences in inter-group dynamics.

Last year, Connections presented its first lifetime achievement award to Peter Perla for his important contributions to the art of wargaming. This year, Connections founder Matt Caffrey was the well-deserved recipient of this honour, which came complete with a banana-trophy. (Last year, late arrival of the trophy had forced the organizers to present Peter with a banana instead).


Matt receives his award, with a picture of last year’s award ceremony in the background.

Working Group sessions followed next. I took part in the group on in-stride adjudication, defined as “adjudication performed simultaneously with play over a period of hours or days.”

It was quickly evident that experience with in-stride adjudication varied widely, in terms of resource availability, time and decision pressures, and (of course) game purposes.  In-stride adjudication is a key part of many of my own games, including the Brynania peace operations game as well as most megagames. Matrix gaming also involves another type of in-stride adjudication.

There were eight presentations, based on eleven short papers that had been written for the session before the conference. Among these was a very good presentation on “player perspectives” by Jason Li, developed from ideas discussed earlier in the year by members of the McGill student team.

My own written piece represented a few hastily-written reflections, but I departed from these in my verbal comments, and instead focused on the notion of adjudicators as both game technicians and theatre directors. In the former capacity they are responsible for making sure the game runs smoothly and stays on-course to achieve its objectives. In the latter capacity they are responsible for keeping the players engaged in the game narrative—maintaining the immersive illusion of a fictional or “what if” universe. It is important not to “break the fourth wall” by having players think more about adjudication than the emerging narrative and the embedded choices it presents.

This issue of directness (or indirectness), intensity, and social dynamics of interaction between players adjudicators—and the impact of these on trust, buy-in, and potential player alienation/grievance—emerged as a frequent theme during the subsequent discussions at my working group table.

The final day of the conference featured a terrific keynote address by Volke Ruhnke on “(War)gaming for complexity.” He started by noting the differences between complicated phenomenon and complex phenomenon. Warfare (and politics) falls into the latter category. Forecasting the latter is difficult because of systemic shocks and other non-linear effects. Model-building—purposeful (but inevitably imperfect) simplification—is essential to understanding and forecasting. If you construct a dynamic model, you can then see how the system might behave under differing circumstances.  Constructing a model also provides an opportunity to see what might be missing, and assists in moving from analysis to synthesis of many interacting parts.

Volko also discussed the impact of diversity (and the “wisdom of crowds”) on analysis. External model-building (and hence game design) provides a mechanism for explaining and synthesizing implicit mental models. He used the CIA training game Kingpin (a game about high value targeting) to show how a collaborative process of game design helped to identify gaps in, and facilitate refinement of, the underlying model. Modelling also forces you to make decisions about what is causally important.


Volko says many clever things about games and analytical modelling, with an image of the Kingpin game in the background.


He compared various modelling approaches (graphic, role-play, manual gaming, computational) across five criteria (ease of use, accessibility to the underlying model, dynamism, iteration, and granularity). Each approach has different strengths and weaknesses. Computational modelling , for example, is powerful, but it may be difficult to understand and assess the underlying (largely black-boxed) model.

For game designers, he emphasized the importance of keeping game designs simple, warned against falling in love (too quickly) with your game model, and you ought to be prepared to combine modelling media. For educators, he noted that it teaching systems thinking requires teaching models. Students, he noted, make great playtesters—and you should invite student critique. For analysts, you might want to pre-coordinate (and illustrate) your models. Feel free to tabletop it—it is a relatively cheap way to explore model. Everyone should consider their mix of both approaches and talent/participants (“diversity is your friend”).

In the subsequent Q&A period, Volke noted the value of teaching game-making. Building the game (even if you don’t get to running the game) can produce valuable insight into issues.

Working Group reports followed, for which I wrote down notes as quickly as I could:

  • The wargaming in education group offered a variety of insights:
    • Regarding cyber education, the group noted that cyber information is often (unnecessarily) classified or highly technical—both of which impede accessibility and hence broader education. While technical experts are best supported by existing resources, managers/policymakers and those working on general purpose games/scenarios are less well served.
    • Looking at wargaming with technology, it was noted that education “does not have deep pockets.” Transparency can be an issue in computer games. Web-based games have advantages in development and accessibility.
    • On the topic of gaming, education, and low-intensity conflict, it was noted that there are several interesting current games at NDU, NPS, and elsewhere. Several gaps and shortcomings were noted, including information operations, cultural misunderstanding, and the existence of ineffective games and exercises. Finally, the group asked how willing we are to game these issues (especially with political or cultural sensitivities). They noted that less structured games (like matrix games) may be especially useful for low-intensity and hybrid warfare topics.
    • The working group offered some thoughts on teaching game design. The noted the value of having students play a game (perhaps even a bad game) and suggest modifications, and generally encouraging critical game play. A “petting zoo” approach can be useful in quickly demonstrating different game mechanisms and approaches.
    • Finally, there was brief mention of CWAR—Collaboration for Wargames in Academics Research—a best practices and information sharing group for those within the Department of Defense and US government. The contact person for this is Scott Chambers (NDU).
  • The working group on linking game purpose to game design was based on PhD dissertation working currently being conducted by Ellie Bartels (RAND) on analytical and discovery games. She noted that attention tended to focus on the topic of a game and game design/format, but there was usually less attention to how to link these two. Her draft framework suggests that the information generated can be fitted into four general archetypes: understanding the problem; structured comparison; innovation; and evaluation. Each type, she suggested, are differentiated by several distinguishing characteristics. The working group tried applying her framework to hypothetical games, to assess its utility in aiding a designer. She noted the challenges posed by differing understanding of terminology (validity, verification, confidence, etc.) across the community.
  • The report from the working group on in-stride game adjudication summarized the process we had used, invited additional feedback on the papers, and encouraged the submission of post-conference reflections too. Revised papers and synthesized analysis from the working group session will be available online (via PAXsims) later in August. Watch this space for further details.
  • The working group on wargaming as a catalyst for innovation was the final one to report. It started by noting the erosion of America’s (military, economic, and technological) edge, and the dynamics of competitive innovation in warfare. This group also had several formal papers/presentations to spark discussion (eventually to be posted on the Connections website). Wargames can expose hidden assumptions and provide a forum for trying new ideas. Difficult gaming challenges may help to force innovation by players. They noted the need to make games flexible and user-friendly. There is a need to bring more cognitive diversity to bear on problems (“not just middle-aged white guys”). There is also a need for more rigour in some areas. More use might be made of alternative political futures. (There was a lot more here too, but I could only type so quickly as the brief-back slides flashed by!)

The final session was a conference hot-wash. Connections 2018 had been flawlessly organized, and there had been a great many very valuable presentations, so feedback was overwhelmingly positive. Some of the audience comments and suggestions included:

  • Game lab feedback and reflections are welcome and will be included in the online conference proceedings. I thought the small, participant-suggested, topic-oriented game lab sessions had worked very well.
  • More time for game demonstrations/play.
  • How best to engage students and early career professionals, especially outside the military?
  • Hosting a panel of wargame consumers/clients, to better understand their perspectives.
  • There is now a Connections US Facebook page.

Before the conference closed for a year, I also reminded everyone of the Connnections conferences to come over the next year:

Connections North will be held at McGill University in Montreal on 16 February 2019, so that everyone can enjoy our balmy winter weather.


3 responses to “Connections US 2018 report

  1. brtrain 20/07/2018 at 10:26 pm

    What a great week!
    Thanks for all your help, and see you in London in, uh, six weeks !?!

  2. Brant 23/07/2018 at 12:44 pm

    “Given this, it could be useful to add greater structure to such games by drawing from design features of some commercial games. Unstructured games can tend to opinion-discussion, rather than making choices with consequences. Stacie suggested that too many designers default to free-form seminar games, which end up being non-game BOGSATs or low-quality games. The latter involve unfocused team deliberations, hasty decisions, and an abundance of semi-relevant background materials that are often ignored by players. Game validity excessively depends on ad hoc, undocumented “expert” adjudication. Seminar games are thus often overly dependent on players and experts, adjudication is based on unexpressed, undocumented mental models, and games are difficult to replicate. They also produce too few innovative ideas.”

    Unfortunate that GEMSTONE didn’t get a better mention as a possible attempt to solve exactly the problems he mentioned:
    — unexpressed, undocumented mental model
    — unfocused team deliberations
    — excessively depends on ad hoc, undocumented “expert” adjudication
    — difficult to replicate
    — tend to opinion-discussion, rather than making choices with consequences

    … all of which are addressed in the underlying structures of the GEMSTONE system

  3. John Curry 09/08/2018 at 11:51 am

    very useful write up. Many thanks

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: