Nature: Two decades of pandemic war games failed to account for Donald Trump

Before the current COVID-19 pandemic, American pandemic preparedness had been the subject of numerous serious policy games. At Nature this month, Amy Maxmen and Jeff Tollefson ask why the US response has been much worse than those games anticipated.

Event 201 was one of dozens of simulations and evaluations over the past two decades that have highlighted the risks of a pandemic and identified gaps in the ability of governments and organizations around the world to respond.

The exercises anticipated several failures that have played out in the management of COVID-19, including leaky travel bans, medical-equipment shortages, massive disorganization, misinformation and a scramble for vaccines. But the scenarios didn’t anticipate some of the problems that have plagued the pandemic response, such as a shortfall of diagnostic tests, and world leaders who reject the advice of public-health specialists.

Most strikingly, biosecurity researchers didn’t predict that the United States would be among the hardest-hit countries. On the contrary, last year, leaders in the field ranked the United States top in the Global Health Security Index, which graded 195 countries in terms of how well prepared they were to fight outbreaks, on the basis of more than 100 factors. President Donald Trump even held up a copy of the report during a White House briefing on 27 February, declaring: “We’re rated number one.” As he spoke, SARS-CoV-2 was already spreading undetected across the country.

Now, as COVID-19 cases in the United States surpass 4 million, with more than 150,000 deaths, the country has proved itself to be one of the most dysfunctional. Morhard and other biosecurity specialists are asking what went wrong — why did dozens of simulations, evaluations and white papers fail to predict or defend against the colossal missteps taken in the world’s wealthiest nation? By contrast, some countries that hadn’t ranked nearly so high in evaluations, such as Vietnam, executed swift, cohesive responses.

The scenarios still hold lessons for how to curb this pandemic, and for how to respond better next time. Deadly pandemics are inevitable, says Tom Frieden, a former director of the US Centers for Disease Control and Prevention (CDC). “What’s not inevitable is that we will continue to be so underprepared.”

Part of the answer, as the title of their article suggests, is Donald Trump:

Confusion emerged in most pandemic simulations, but none explored the consequences of a White House sidelining its own public-health agency. Perhaps they should have, suggests a scientist who has worked in the US public-health system for decades and asked to remain anonymous because they did not have permission to speak to the press. “You need gas in the engine and the brakes to work, but if the driver doesn’t want to use the car, you’re not going anywhere,” the scientist says.

However, they also note that—regardless of who occupies the presidency—institutions also failed to respond to insights and warnings that emerged from many of these games.

Perhaps the biggest limitation of simulation exercises was that they didn’t actually drive policymakers to prioritize and fund improvements to the public-health system. Morrison now questions whether it’s even possible to do that through simulations alone, or whether people must experience an epidemic at first hand.

The vulnerability of games (and, by extension, policy response) to idiosyncratic player/policymaker effects, and the difficult of translating game results into real world consequences, is something that was debated in a classic 1964 internal RAND discussion between Robert Levine, Thomas Schelling, and William Jones—which every policy gamer should read. I’ve also addressed the challenge of gaming unpredictable and unreliable decision-makers in the specific context of the Trump Administration, while Devin Ellis has commented on the problems that arise when game insights are not translated into policy or preparedness.

For more on pandemic, see the PAXsims COVID-19 serious gaming resources page.

Rielage: An Open Letter to the US Navy from Red

In the latest issue of the US Naval Institute’s Proceedings (June 2017), Captain Dale Rielage offers some hard-hitting observations about the way in which the US Navy prepares to fight future wars, written from the perspective of the Red (opposing) team.

Dear U.S. Navy,

It is time we talked.

We have regarded each other from a distance for years, but we need to get to know one another better. You see us in every major exercise and wargame. In the outbriefs, we usually are on the back wall, mixed in with the staff. The White Cell and Control talk about us a lot, but usually in the third person. Rarely do we have an honest conversation.

But lately the Chief of Naval Operations (CNO) is talking about high-velocity learning, and there is discussion of a renaissance in wargaming. Maybe this is the excuse we need to start talking.

Across dozens of exercises, live and synthetic, tactical to operational, on both coasts, we have had the opportunity to watch your ways. We sit in every wargame, each one unique. The Blue teams across from us are diverse, representing every type of Navy authority, from students to operational-level commanders, and every warfare community and variety of staff life. We respect the variety and depth of professional excellence they bring to the fight. Nonetheless, regardless of which actual adversary we are representing as “Red,” there are patterns to our interactions that are worth your consideration, both in how you fight and how you train.

Much of what he  has to say is a searing critique of the way the Navy—and, I would argue, many others—approach wargaming. I’ll quote him at length, because it is an absolutely outstanding piece:

Your exercises have become Christmas trees. Time is precious, and the Fleet has fewer days under way and fewer flight hours than ever before. As a result, pressure to “maximize” training opportunities has grown. Doing a field-training exercise? Great, combine it with a staff exercise. Add some outside experimentation. Make sure several echelons are being evaluated and certified at the same time. The result is efficient but requires a high degree of scripting. Anything that throws off the timetable of the exercise results in a cascading series of events that don’t happen.

Enter Red, the adversary who defines success by creating friction and failure in Blue’s world. The only way the Christmas tree keeps all its ornaments is if Red is prevented from imposing too much friction. Have limited range time and need to conduct a strike mission? “White card” the high-end naval surface-to-air missile threat out of existence, because shifting to conduct a maritime strike mission to clear the ingress would throw off the exercise schedule.

Clearly, accommodations are necessary in training, but making them has become your opening assumption. Blue would do well to review why events are being conducted and identify the minimum essential events that must be completed. It may be that less is more.

Your opposing forces often are very good, but you have trained them to know their place. Most fleet training centers have a team capable of presenting a good-to-excellent Red threat. However, our experience is that they have learned to self-regulate their aggressiveness, knowing what senior Blue and White cell members will accept. As one opposing force member recently told us during a “high-end” training event, their implied tasking included not annoying the senior flag officer participating in the event. They knew from experience that aggressive Red action and candid debriefs were historically a source of annoyance. They played accordingly.

Excellence may be where you least expect it. We have consistently seen that the real centers of innovation and excellence are the commands and teams that have only recently started to look at a particular operational problem. As the new folks, they are learning the current baseline, are less likely to make assumptions based on how they “know” the scenario is supposed to go, and are open to what constitutes true “high-end” warfighting.

There are no points for internal excellence. As U.S. Navy professionals, we understand it is essential for the warfare commanders to be aligned and communicating well. The quality of the staff’s standing orders and the clarity of the commander’s intent are important. The experience your planners gain in the training is praiseworthy. As Red, we really don’t care. The bottom line is simple: Did you beat us? There is a time and place for sorting out staff processes. If that is the focus of this training event, great. If not, don’t commend yourself for it.

You must make time to stop, listen, and think. In too many events, the training loop is never completed. Debriefs tend to be cursory, typically at the end of the day when the entire team is tired and wants to move on. Events often are not equipped to capture ground-truth data and feed it back to the training audience quickly. Often months later, a long report is generated. The more honest it is, the narrower its circulation—in many cases never outside the training audience, who by then has moved on to the next challenge.

Be clear what we are doing. There are a number of ways to present Red. Red can be unconstrained, using the adversary toolbox in ways that seem most effective from a U.S. view. Red can be doctrinal, using the adversary toolbox in the way we think the adversary likely would. Most often, however, Red is constrained, asked to perform a specific function to facilitate an event.

Wargamers and exercise planners often recall Millennium Challenge 2002, an experimentation wargame run by Joint Forces Command. Marine Major General Paul Van Riper, playing an unconstrained Red, used innovative asymmetric tactics to shut down Blue in the first move. Blue had asserted that its new concepts would be tested and validated against an unconstrained Red, but when its objectives were threatened, it reset the game and created rules that, according to the final report, boxed in Red “to the point where the end state was scripted.” The entire event generally is remembered as an example of what not to do, perhaps because the game became a public controversy after General Van Riper quit as Red force commander. The reality is that we repeat this experience on a smaller scale multiple times each year.

In one recent event, Red was helping assess a new naval concept. In support of this assessment, Red presented a consistent, accurate, and limited threat to Blue, allowing Blue to work through a series of actions and understand the variables involved. It was the military equivalent of batting practice, with Red serving as the ball machine to put consistent fastballs in the strike zone. It made sense, and doing it well was important and worthy work. The problem developed later. As the results of the event were presented to more and more senior audiences, the briefs grew shorter and more “executive.” The description of the Red role eventually became a list of the organizations that had contributed Red players. By the time the briefing reached the four-star level, the implication was that Blue had validated its concepts in a full game against an unconstrained adversary—which was not the case. Red left the event convinced that, given realistic latitude, it could have stressed Blue’s concept to failure, perhaps even turned it into a costly defeat.

Failure should be an invitation to learn. Generally, when Blue units are killed in training events, they are quickly regenerated. Why? Typically, there are two answers:

• If Blue does not have X, it cannot do Y, and Y is a training objective. This makes sense in some cases, but in more complex exercises there is value in fighting hurt. Yes, if Blue falls below a certain level of forces, it cannot complete its tasking. How about the implied task of preserving surviving forces? Breaking contact, regrouping, and reengaging? These do not appear on the training order and are not normally exercised, but maybe they should be.

• If unit X is killed, it will miss the opportunity for further training. We create negative learning when taking fatal damage is consequence-free. If training demands a unit be regenerated, at the very least, the killed unit needs to conduct an immediate critique to answer the basic question “why did we get hit?” The answer in many cases is that they were balancing risk across a number of mission areas and the die roll came up badly for them. Sometimes, however, there was an avoidable loss of situational awareness or a failure to account for one threat while focusing on another. The cost of coming back into the fight should at least be a back brief to the White Cell. Further, if regenerating units is required for training, senior officers need to stop citing the resulting exchange ratios as evidence of operational proficiency. A 10-to-1 victory isn’t if Blue was effectively missile-proof.

You talk about accepting failure as a way to learn, but refuse to fail. It is instructive to ask a room of senior officers the last time they played in—or even heard of—a game or exercise where Red won. If our collective assessment is that Blue really can best its adversaries every time, we are in a good place. If not, it is time to rethink the process we have created.

For us, the point of playing Red is not to beat Blue. It is to train Blue. At the end of the day, nothing would make us happier than to bring our best game to the fight and get our clock cleaned. At this point, getting there will require a number of uncomfortable conversations and a level of personal and institutional self-honesty that, bluntly, we have not cultivated. But we must, and soon. As the CNO has said, our “margins of victory are razor thin,” and the real adversaries keep improving.

Meanwhile, we are always available to talk. Just look across the table.

(Competitively) Yours,

Red

Hopefully his observations will spark considerable discussion in the relevant sessions at next week’s Military Operations Research Society annual symposium, as well as the Connections US (August 1-4) and Connections UK (September 5-7) professional wargaming conferences. I’ll be at the latter, while other members of the PAXsims crew will be at the first two.

h/t Tom Mouat

Legal Advises You to Choose a Fictional Country…

Jonas Savimbi IRL and in Call of Duty: Black Ops

We don’t talk a lot on the blog about the weirder liability considerations involved in games designed for profit – or even sometimes as part of a public research agenda – but the risk is out there.

The family of infamous Angolan rebel Jonas Savimbi is suing the makers of Call of Duty: Black Ops over the game’s depiction of the warlord. Three of Savimbi’s children, who live in Paris, having taken the company, Activision, to court, demanding 1 million Euros in damages for defaming their father as “a barbarian.” The game designer’s lawyers, meanwhile, have called the portrayal: “favorable.”

Call of Duty: Black Ops 2 shows him rallying his troops with phrases like “death to the MPLA”, referring to the party that has governed Angola since independence from Portugal in 1975.

But his family said they are outraged at the depiction.

“Seeing him kill people, cutting someone’s arm off… that isn’t Dad,” said Cheya Savimbi…

A lawyer for Activision Blizzard, Etienne Kowalski, said the firm disagreed with Savimbi’s family, saying it showed the former rebel as a “good guy who comes to help the heroes”.

OK then. Well the U.S. government had strong currents of support for him at times too, I guess – despite the appalling violence committed by UNITA (including burning suspected witches. Really).

At least in Brynania you can assign whatever despicable behavior you want to the Zaharian Peoples Front (ZPF) without fear of winding up in court. Game writers take note.

Layout wonkiness

Visitors will notice that the layout of PAXsims has suddenly changed, with the sidebar having mysteriously vanished and reappeared as a footer—which doesn’t look right at all.

We’re not sure why this has happened—perhaps its some sign of the approaching Mayan apocalypse—but we have a support request into WordPress to get it fixed.

* * *

UPDATE: And, as strangely as the problem arrived, it seems to be fixed. We’ve even tweaked the font for good measure. Thanks be to the capricious Mayan gods!

ANOTHER UPDATE: Apparently the new font displayed poorly on some versions of IE, so it’s back to the original one.

Serious gaming the challenges of humanitarian preparedness

Pablo Suarez (associate director of programmes at the Red Cross/Red Crescent Climate Centre) was kind enough to drop us a note highlighting some of the work that they have been doing over the past few years using serious games to highlight and address the humanitarian consequences of climate change and extreme weather events. Some of this work has been done in conjunction with the PETLab at  the Parsons—The New School for Design, who have also put together a website (here) devoted to this particular case of “developing public interest games for better crisis-decision-making.”

* * *

Weather or Not is a simple game where participants are given the probability of a major storm, and then must decide whether or not to pre-position relief supplies. If they DO and there IS a flood (or if they DON’T, and there is NO flood) all is good. However if they DO and there is NO flood (or if they DON’T and there IS a flood) they are punished for over-reacting or failing to prepare. The game can been seen in use in the video below, with a graduate class at Columbia University: 

The best game strategy here seems rather blindingly obvious (prepare if the chance of a flood is above 50%), so presumably this would best be used to either familiarize people with probability estimates or to spark a larger discussion of the emergency preparedness.

* * *

Before the Storm is a card-based game where groups of participants are given a series of weather forecasts (at 10 days, 48 hours, and 12 hours) and are asked to select the appropriate preparedness measures from the deck. They can also develop their own ideas, and summarize them on their own card. This seems to me to be a much richer use of a game mechanism, with participants not only encouraged to weigh the pros and cons of various options but also challenged to think of new approaches of their own. In the video below the game can be seen being used in Senegal. In this case, once the game had been played and new various options had been generated, the group visited a flood-prone village to get community feedback on their ideas.

* * *

Spreading the Word is a version of the party/children’s game telephone, used to highlight problems of communication between scientists, relief workers, and local communities. You can see it at work here (at 04:00 to 17:45 in the video) in a workshop in Bangladesh. While the outcome isn’t surprising to anyone who has played the game before, it does seem a very entertaining way of highlighting the point in a lecture or workshop setting.

* * *

Choices in a Changing Climate looks at the twin challenges of flood and drought (longer version here). Again, the game is as important for the way that the game mechanics stimulate and facilitate discussion as it is for the lessons built into the game rules themselves.

* * *

Dengue, Catch the Fever! is designed to teach primary school children (and, secondarily, their parents and other stakeholders) about the risk factors for Dengue Fever, and the way these relate to issues of climate change. You’ll find an overview of the game here, and the game instructions here. Very clever, and it looks fun to play!

* * *

The Red Cross/Red Crescent Climate Change Centre also has links to other serious games used by national Red Cross/Red Crescent Societies:

• Goose Escalera, a Spanish-language snakes-and-ladders type game for children (board, instructions) used to highlight environmental and climate change issues in Colombia.
• Earth Savers, an Arabic-language boardgame on climate change for children, this time prepared by the IFRC for use in North Africa.
• A Syrian computer game on the same theme.

One also shouldn’t forget a couple of other browser-based games with somewhat similar themes that we’ve discussed before at PAXsims, namely Stop Disaster (developed for the United Nations’ International Strategy for Disaster Reduction) and Inside Disaster (an interactive videoclip game on the Haiti earthquake). I’ve used both of these with students with great success.

Overall, there is a lot here to spark ideas as to how similar approaches can be used to address other humanitarian and developmental issues.  Moreover, as the work of IFRC and PETlab shows, you don’t need to make these complicated or electronic to get the basic point across. From a gaming perspective,it  is also easy to think of a number of existing card and boardgame techniques that might be applied to the issue of disaster preparedness. It would be interesting, for example, to design a cooperative card-driven game somewhat akin to Pandemic that whereby event cards generated disaster risks, forcing players to adaptively switch emphasis and limited resources from longer-term mitigation strategies to shorter-term emergency preparedness and response.

(Coincidentally I spent part of the holidays designing and play-testing a disaster response game. On the plus side, it was a hit with my local gaming group. On the other hand it may not be of much practical use, since it involves a future zombie apocalypse. Even without prodding from the IFRC, however, we did work climate change into the basic game setting!)

how not to run a simulation in the Caucasus…

It seems that Georgia’s Imedi TV has done its own version of Orson Welles’ famous 1938 War of the Worlds radio broadcast, Caucasus-style:

As al-Jazeera reports:

“Russian tanks have invaded Georgia and the country’s president has been killed.”

That is what Imedi TV, one of Georgia’s most popular television channels, reported on Saturday night. It also broadcasted video of what appeared to be the Russian army in the country.

In reaction, many Georgian residents rushed into the streets in panic and emergency services were flooded with phone calls. Even a number of heart attacks were reported to have occurred.

But when the broadcast turned out to be untrue, widespread panic across the country turned into anger.

The pro-government channel later apologised, saying it was a mock newscast to show “what the worst day in Georgian history might look like”. It also said it “should have clarified” that the programme was a fake TV report.

You’ll also find coverage at CNN, and elsewhere.