Connections 2019 report

This year saw the Connections wargaming conference move to the US Army War College, in Carlisle, Pennsylvania. The event was just outside the base in the Army Heritage and Education Center, which was particularly helpful in that the administrative burden to get permission to attend for people like me (non-US citizen) was made significantly less onerous.

The event took place from Tuesday 13 Aug 19 to Friday 16 Aug 19.  Details of the programme are here.

I elected to fly over two days early because, thanks to changes in the Uk Ministry of Defence regulations, we are not permitted to drive having flown across the Atlantic until the following day. Since flying in to Washington and driving to Carlisle was considerably cheaper than trying to get closer by air (and gave me a car all week), it was an easy decision to make – and since the Battlefield at Gettysburg was on the route, I was able to get a little bit of military history on the way.

This year’s theme was: Futures of Wargaming – to which the obvious question was: was this to be about the “future of wargaming” or about “wargaming the future?” – to which the answer was “Yes!”.

Day 1

Colonel Russ Griffin welcomed us to the Heritage and Education Centre, explaining the purpose of the Center, along with the usual senior officer’s politically correct joke – but this time it was one about the fact that at the Gettysburg Cemetery, President Lincoln spoke his “few appropriate remarks” in less than 2 minutes, after former Secretary of State Edward Everett’s 2 hour oration. History remembers Lincoln, but few remember Everett – so his remarks were going to be like Lincoln’s – short and memorable, rather than long and forgetful. A fine ambition for a military officer and an anecdote that I shall steal for myself…

The programme this year was far more complex than usual, reflecting the high attendance, but the number of concurrent sessions caused frustration from the start, with Dean Cheng’s presentation on PLA Wargaming clashing with Mark Leno’s Computer-Assisted Wargaming. Since the topic of computer assistance was one that my superior headquarter was particularly interested in, I felt that duty came before personal preference and went to Mark’s session. I was not disappointed, however, as Mark gave a very interesting presentation on the practical aspects of computer assisted wargaming in a clear and frank manner, covering three separate game designs: An Information Warfare Game, a Defense Management Game and a Build the Force game.

The Information Warfare Game was a 7-Sided, blind, resource allocation game with a game map representing the cognitive space of the conflict with the aim of training the participants to think in a different way about the problem, allocating resources to discover information and apply influence. The challenge being to provide a gameboard capable of being centrally updated and delivered to different locations.

The Defense Management Board game was used to replace a written assignment as part of an existing exercise in order to make it more challenging. The requirement was intended to be part of a distance learning package using asynchronous turns, so players needed to manage their own time in planning and delivery. In essence the different roles represented by the participants had to agree 45 current and 16 future programmes going forward.

The Build the Force game was to intended to get the participants to generate a balanced force, based on political and intelligence estimates, to meet and defeat any crises that develop during the game, at the minimum cost. The types of crises were graded from “most likely” to “least likely”, and used a mechanism to re-order them each turn. I was particularly interested in the fact that the students who elected to do excessive homework for the game doing some heavy math preparation tended to do well in the early stages of the game, but poorly later. That spoke to me of a well-balance design, where experience and intuition were also important. Managing the design to use real world force elements, with “ball-park” representation of costs, and still keep it unclassified, was also quite an achievement.

Mark was frank about the challenges of project like this, operating on military systems, using software tools (mostly Microsoft Office applications) that were universally available – rather than trying for bespoke solutions that would be extremely difficult to distribute and maintain. This meant that graphics would be limited, but the training burden would be lowered due to familiarity with the products. He also emphasised that it was essential to try not to be too ambitious and to keep solutions simple. A particularly helpful presentation intended to provide practical advice, rather than trying to make the team look clever.

For the second session, I elected to attend Joe Saur’s session on Combat Modelling, rather than Operational Wargaming in Korea or a session on Logistics. As a logistician by Arm of Service, I was tempted by a session referring to Logistics as the “Red-Headed Stepchild of Wargaming“, but in the end went with the Combat Modelling.

Joe is an extremely experienced instructor and his clear exposition on Attrition Models and Lanchester were very helpful, moving forward in complexity, covering factors in the modelling, with examples like the Madden NFL video game, through deterministic and stochastic models, and the sources of data.

I particularly like Joe’s discussion about the tendency to measure what is easy to measure and the problems associated in implementation between the Coders who make things happen in a simulation, versus the subject matter experts who have a more holistic view about the effects that changes in the variables might have.

This connected directly for me with the assumptions that designers make in their models, that may consciously or unconsciously reflect their biases, and the example provided in the UK’s Defence Wargaming handbook of the the “Gulf Strike” commercial boardgame during the First Gulf War (page 41). The designer, Mark Herman had developed a simple hobby wargame model, that was far more accurate than the multi-million-dollar simulations used by the DOD, as it was based on more realistic assumptions.

This was followed by a choice between Ed McGrady on Games as a form of Play, Merle Robinson on Megagame Design and Development, or Matt Caffrey with a book talk about On Wargaming. Since I have been playing Megagames for over 40 years, I was interested in Merle’s talk but felt I couldn’t really justify it, and Matt was to be presenting at UK Connections where I could see his talk, I elected to listen to Ed’s talk about Play.

Ed’s proposition is that play is underrated and that understanding it properly will give us insights into the process of learning, understanding and make for better games. He covered the elements that make up play, the huge range and diversity of play, relating them to games.

The “Magic Circle” was also covered. In games, the “magic circle” is the space in which the normal rules and reality of the world are suspended and replaced by the artificial reality of the game world. There is a boundary with an “inside” and an “outside”, but the boundary is porous, with the inside having its own unique rules but reflecting characteristics of the real world and the player can used this different cast on the real world, in order to help understand it and learn from it.

I was especially interested in his comments outlining the fact that surprise within the scenario is important in practicing the player in reacting to change (the Israeli Dado Center for Interdisciplinary Studies view this as extremely important for building adaptability). Also, the important psychological importance of the difference between an adjudicator rolling the dice, permitting the player to roll the dice, and requiring the player to roll his own dice. This is closely linked in my mind to the relationship between dice and understanding risk – and more importantly differentiating the difference between the understanding of a calculated risk and merely gambling.

The essential point being that play is serious and has a significant impact on a wargame and removing all aspects of play from a game is a mistake – but also we can learn from a study of play about the essential elements that encourage “flow” (the mental state of operation in which a person performing an activity is fully immersed in a feeling of energized focus, full involvement, and enjoyment in the process of the activity), such as movement, haptics, surprise, challenge, etc. It was also illuminating to consider how computer games suppress many of these elements, while excessively indulging in others.

This was followed by a panel session on Wargaming Methods.

Jeremy Sepinsky started out with Programming with People about the organisation of the human component within the wargame, pointing out that whatever answer you get from a game will be meaningless unless you really understood the question the wargame was intended to answer in the first place. Players need to be organised in a way that supports answering the question and the more complex the question, the greater the number of people needed to address the question – and the greater difficulty you have in ensuring that the question is really understood.

He also pointed out that players require active management – you won’t get a second chance (unless you are really lucky), so they need to be kept focussed on the goal – but take care not to go too far to the extent that game play, or the people themselves, are adversely affected.

This was followed by Kenneth Sawka and Incorporating Structured Analytic Techniques into Business Wargaming. This was interesting – the method’s aim is to make the analytic process conscious and transparent, thus reducing the probability of errors caused by cognitive biases that go unchallenged in more usual unstructured and intuitive analysis – and the business use of the technique helped illuminate difference between that and the intelligence community. I especially liked the focus on assumptions, critical assumptions, checking assumptions and indicators of change and events that would change or invalidate assumptions.

Lastly Jim Markley talked about US Army War College Wargaming. This was the Strategic Wargaming Department and the use of games and exercises for education. What particularly impressed me was the range of diverse games and techniques (and my jealousy of their 3D printer and large-scale plotter). Chris Engle matrix games were singled out for a mention, which was nice to see as an accepted “mainstream” technique after all this time.

This was followed by another split session with the choices of Using Design Thinking for Designing Wargames by Yuna Wong, War Cry – Combat Force Cohesion and Capability Disintegration with Uwe Eickert or Gaming Urban Terrain and Megacities with Ed McGrady. Since I have been specifically asked to look at urban warfighting I elected to go with Ed again.

He covered a lot of ground in his presentation, pointing out that running this as a game was an abstraction, hoping to gain insights that would be useful, rather than an attempt to slavishly represent all aspect of urban combat.

There were a huge number of elements in urban combat that were of particular note in city environments, as well as the fact that every city presented unique terrain challenges due to the extreme diversity and that sweeping generalisations were not likely to be helpful. The terrain of course, was not limited to the surface physical domain, but covered sub-surface, cognitive and electromagnetic domains as well.

The issue was extremely challenging at all levels of representation and there are no easy answers. Many attempts at manual game structures tried in the past have been barely adequate and most computer simulations seeking to address the problem were extremely poor.

The day finished off with a Wargame Testing and Interactive Demonstration social session at the Desperate Times Brewery, which was very pleasant (but the background noise due to the building construction made communication difficult). I played Paul Vebber’s game about research and development in the production of the next generation submarine.

Day Two

Day Two started with a Keynote from Dr Steven Stoddard, Deputy Director for Force Development. He elected to give a direct view on wargaming with the assertion that wargames inform change, and change needs leaders to make decisions. He chose to illustrate this with three examples: a NEO operation in the Pacific, a game for the Resolute Support HQ in Afghanistan and a game about near-peer warfighting prior to an Army modernisation programme. As he put it – telling war stories about wargames…

He gave a whole series of fascinating insights into designing games for senior leaders and the design process when you have high level games involving very senior officers and large numbers of them. The requirement for targeted rehearsals for mentors, facilitators and the Generals’ own staff; modifying the games based on the feedback received; all intended to ensure buy-in from the participants. I really liked the way in which success could be measured by the amount that the junior Generals and staff took over the debrief to present points to the Seniors.

Dr Stoddard was very open and frank about what worked well and what didn’t, and it was an outstanding presentation.

This was followed by a panel session on Futures Gaming.

Dr John Hanley talked about the CNO Strategic Studies Group Gaming, covering new concepts, requiring new models and the problem of proper validation. He took a historical view moving forward and finished with a clear view on what worked well and what did not.

Deon Canyon looked at Matrix Gaming at the Daniel K. Inouye Asia-Pacific Center for Security Studies, with some compelling insights into the current standoff with the DPRK and how the issue was examined in one of their courses. There were a number of observations raised on the technique, and modifications to make it work with the specific circumstances of the course that I found especially helpful. There was also a desire to come up with a system for indicating at a “winner” for the game, but I was please to see that this did not end up distorting the open-ended Matrix Game play.

Finally, Dr Kiran Lakkaraju talked about Experimental Gaming at the Nuclear Threshold. This was looking at experimental wargaming for National Security related to nuclear weapons, breaking the “nuclear taboo” and come up with some quantitative evidence from a large number of games. The technique used both manual wargaming in order to perfect the design, but also an on-line version. The Project on Nuclear Gaming (PoNG) is available to play at specific times as announced on the website. Some users have mentioned difficulty in running the game, but it is generating useful data. The manual board-game version was available to play in the demonstration sessions.

We then went into the Game Lab sessions.

Again, as I have been given homework to do about urban operations, I went to the Future Combat in Megacities session. We were fortunate to have a diverse group discussing the topic, very capably led by Ed. The summary of the session that I gave to Yuna Wong afterwards really reflected the topic – it is hard to do – very hard. This was something of a relief to me as I had been struggling with the subject for some time. There are some good approaches, specifically the excellent We Are Coming, Nineveh! game from Juliette Le Ménahèze and Harrison Brewer at McGill (modified by Brian Train and Rex Brynen), but this was a level of abstraction that was unsuitable for many circumstances, and the equally interesting Urban Operations from Sébastien de Peyret, was seen as too complex for other circumstances.

We covered issues of gaming at the strategic, operational, tactical and detailed sub-tactical level; we looked at the difficulties of depicting structures, elevations, tunnels, line-of-sight (the game designer Jim Wallman physically modelled parts of a city using large quantities of foamboard and lasers for line-of-sight for one game); we also considered communications, cyberspace, casualty rates, logistics (drone delivery anyone?). The issue of hidden information and how that could be represented in a game was a complete topic in its own right.

An interesting point that had come out of last years’ Connections in Yuna Wong’s session on AI, was the possible use of AI “bots” not physically deployed in the city, but instead deployed “on-line” using algorithms to track down insurgents and enemy combatants from their communication interactions.

A very good point was made about the unique difficulties of representing civilians and their effect on operations at different scales – and the fact that a disproportionate amount of planning and game design made huge and often totally invalid assumptions about their behaviours and effects.

One of the fundamental issues is why a military force would wish to be involved in an urban location in the first place, since it was operationally so difficult. There are a number of possible reasons for intervention, such as vital route clearing or the destruction of enemy indirect fire assets using the terrain as cover, and it seems to me that each mission may require a slightly different approach to a possible game.

There was quite a lot of experimentation done in the UK about fighting in urban areas in the 1970s, some of the results of which are quite counter-intuitive, and are contained in the excellent book The Stress of battle: Quantifying Human Performance in Combat, by David Rowland, but it is out of print and very difficult to get hold of at a reasonable price.

In the afternoon we had the Gaming Showcase and Demonstrations. I put on a demonstration of a modification of the DSTL Cyber game that is used for training senior officers and students on Cyber courses offered at the Defence Academy of the UK. This involved the players attempting to carry out a Cyber attack on a piece of vital national infrastructure from the enemy point of view.

Following dinner, the demonstrations continued and I was asked to repeat the Cyber game demonstration until we had to depart.

Day Three

The following morning started off with Commander Phil Pournelle, Retd, from ONA, on The Challenge Before Us. He observed that Secretary Bob Work had left the building and some of the impetus behind wargaming had slackened, but that the community was extremely vibrant and it was possible that we could “keep the nose up” on our flightpath and continue to deliver what was needed.

He observed that the nature of warfare had changed, needing new operational concepts that we didn’t have yet. We are not prepared for the new threats; our models are overly simplistic and simply not good enough for the current challenges. However, the analytical community and the wargaming community were actually starting to cooperate at last, so all was not lost. It was therefore essential to maintain momentum and made a plea to support the MORS Conference and other initiatives.

We then had a panel on Modelling and Wargaming.

This started with Jon Whetzel on Online serious gaming: Developing wargames for the crowd. Jon was part of the team from Project on Nuclear Gaming (PoNG), mentioned above and it was refreshing to have a younger member of the community presenting on the subject. He covered the types of applications used to create the Signal game and provided a number of key take-aways:

• It is possible to create a complex game with a small team using open source software.
• There is an underlaying dichotomy between the need for complexity to provide the analytical results and the need to make it entertaining in order to to get people to play the game (especially if you were not going to pay them).
• Fidelity across mechanisms was important and part of the development process (Boardgame version vs the on-line game).
• The technical infrastructure for an on-line game is at least as important as the game.

He was followed by Michael Robel with a presentation Toward Automating the Course of Action Generation and Staff Wargame. This is a key area of interest for the UK currently and was of particular interest to me. The presentation was very interesting – especially as the approach is a little different to the direction we are taking and seemed to support my view that the computer assistance is not there to select courses of action, but to highlight the ones that were most likely to fail and could therefore be discounted.

Brian Kirkpatrick followed with War Game Networks for Digital Distribution and Collaboration. He highlighted obstacles to technology transfer, from the concepts to the actual deployment, and the time it takes. In order to do digital distribution effectively, you need common platforms, services and data (matching Mark Leno’s comments from Day One).

Then we had Karl Selke with Modelling the player: A requisite for structured wargaming. The aim was empowering wargames through the automation of much that it was possible to automate, in order to get to a point where the wargames could generate robust analytical data.

We went on to the next group of seminar sessions: Data Collection and Analysis with Chris Weuve, A Wargaming Approach to Computational International Relationswith Karl Selke, and How to Improve your Communication Skills with Dana Lombardy.

I chose Chris, rather than Dana (despite, as an educator, wanting a view on improving my communication skills), because I felt that since the OR function in the UK MOD was now wholly the responsibility of the Defence Science and Technical Laboratory (Dstl), I had not had enough exposure to Data Collection and Analysis and this was an area of weakness.

Chris started with the comment that “A Wargame is often an act of Political Theatre” and carried on in a refreshingly candid style. He pointed out that wargames were full of stated and unstated assumptions, where best at posing questions rather than definitive answers and needed to be approached with realism rather than cynicism.

He also pointed out that the main purpose of the game report was normally to limit the claims made about the game results—so, in designing data collection and analysis, it was essential to plan for failure. You cannot analyse data that you didn’t capture and you cannot capture data that you didn’t generate.

Analysts need to be involved with the design of the wargame from the beginning as attempting to suddenly “bolt on” analysis afterwards was bound to end in failure. Data collection should be a mechanism of the game if it was not going to become intrusive and distort the game play (in the same way as Senior Officer “observers” often do) (interestingly there was a paper offered at the IITSEC Conference demonstrating that the physiological effects of fear and the stress caused by having your actions closely watched by someone sufficiently senior, are practically identical).

He stressed in analysing a game that you should be sceptical, highlight the subtle, caveat the limitations and ruthlessly quash the irrationally exuberant. In designing the game and the data collection plan, always get the sponsor to agree in writing, be realistic and plan for failure, document and critique the process and always have lots of office supplies…

He even provided a reading list:

• Wargaming Pathologies
• The Art of Wargaming
• Rules of Play 

I was extremely impressed with the presentation. Chris was very good at explaining how things were supposed to be, while all the time being totally comfortable in providing real world examples of why they were not, with the easy confidence of someone with a huge amount of experience in the field.

The next session was working groups on: The Future of Wargaming with Ed McGrady and Mike Ottenberg, Wargaming the Future with Stephen Downes-Martin and Wargaming for Future Leaders with Mike Dunn and James Morningstar.

I elected to take a look at the Future of Wargaming, since this was also an area where I was asked to look at specifically.

The session featured Pete Swan from VT Mak, which was a truly awful propaganda presentation (delivered on-line) about their product (VR Forces). Since VT Mak are part of the team involved in the US Army Synthetic Training Environment (STE), I can understand why they are asked to contribute, but equally having listened to BG Gervais explaining the ambition of the programme that went from exciting, to very challenging, directly into magical unicorn territory (given the timelines) (it gave me flash-backs to the FCS programme for those of you who remember that), I would have appreciated some indication (however optimistic) as to how they thought they were going to deliver.

However, this was followed by a truly excellent presentation from Lucien Parsons, Director – Mixed /Augmented /Virtual Reality Innovation Center (MAVRIC). He provided a candid and engaging insight into the use of VR/AR/MR (now called “XR”) that was of direct used in the courses that I deliver at the Defence Academy.

The insights into the changing demographics of the gaming industry were especially useful as was the example of Maj Travis Sheets and Matthew Elmore in their paper Abstract to Action: Targeted Learning System Theory Applied to Adaptive Flight Training  in low cost pilot training.

There was an awful lot more in his engaging and frank presentation, including the dark side of the technological advances being made (such as privacy, eye tracking revealing hidden emotions, etc), and I would recommend checking out the slides when it is put up on the Connections website.

This led directly into the breakout session where we split into 4 sub-groups in order to examine the possible Futures of Wargaming through the mechanism of using quad charts with suitable axes such as low cost to high cost vs manual to digital techniques.

We elected to go for low cost to high cost against barely effective to very effective, but I am ashamed to say that was probably because I was influenced by the RAND paper on Collective Simulation-Based Training in the US Army that used the same axes and highlighted the problem of effective but hugely expensive and overvalued solutions, while failing to identify the cheap and effective alternatives (such as Majors Sheets and Elmore proposal for pilot training, above).

However, one of the other teams came up with the novel and interesting: Traditional to New Methods vs Physical to Cognitive. This generated quads of Physical/Traditional (Kriegsspiel), Traditional/Cognitive (Freudian) (like traditional Information Warfare and Pysops), to Physical/New Methods (Ender’s Game) and the most interesting Cognitive/New Methods (Minority Report/Inception).

This exercise was both fascinating and useful, and I shall ruthlessly steal the technique in the future.

Day Four

The final day started with an excellent keynote from Ed McGrady, The Future of Wargaming – It’s up to you!

This was a call to arms for wargaming and a stern admonition that we need to stop feeling sorry for ourselves and start making better choices. Games are about understanding a problem, about narrative rather than data, they are not a science experiment – the story is champion in the experience. He pointed out that if the wargaming community fails to properly define its value to Defence, then someone else will…

He suggested that in order to be influential, influential players were needed – but care was necessary. They need to believe in the game and you need to give them a good experience. We need to concentrate our efforts on the right subjects – precisely those that are the current, hard topics, such as social media, influence, grey zone conflict, people, etc., and get the right people (decision makers) to take part.

We should also not be concerned with sniping from the margins about wargames not being repeatable – practicing decision making, for which wargames are ideally suited, is obviously repeatable – whereas Desert Storm was the very definition of a one-off event (that really wasn’t instrumented properly) (and people took a great many questionable conclusions about the future of war from it).

The game might be unique, but themes emerge, and these provide the sorts of valuable insights that senior leaders need in order to shape the direct of the Armed Forces for the future.

Finally, we had outbriefs from the other sessions, closing remarks and a quick hotwash.

In conclusion I would say this was a truly excellent Connections Conference. My disappointment with multiple sessions running in parallel (meaning I missed out on a couple of things I wish I had experienced) was offset by the depth of the sessions,

The presenters gave the impression of being very open and candid – as if they were among friends who were interested in means and methods, rather than formally trying to impress anyone. This made them more accessible and the presenters approachable. I firmly believe that this made a significant contribution to the value of the event.

In addition, the administration for the week was simply excellent. The venue was ideal and the US Army War College were wonderful and approachable hosts. The catering arrangements were outstanding and something, as one of the organising team from Connections UK, I looked on with jealousy as it is not something we could hope to compete with in central London.

In short – something of a triumph – and a note of respect to Matt Caffrey, Tim Wilkie, the US Army War College and the rest of the team.

Tom Mouat