War-Gaming Taiwan: When Losing to China Is Winning

What military planners learn when they simulate a Chinese attack.

By , the lead war game designer at CNA, and , a research analyst on CNA’s gaming and integration team.
taiwan-china-us-war-games-nicolas-ortega-illustration
taiwan-china-us-war-games-nicolas-ortega-illustration
Nicolás Ortega illustration for Foreign Policy

In 2030, Beijing reaches a deadly decision: reunification with Taiwan at any cost. Cyberattacks plunge the island into darkness, confusion, and fear. A rain of missiles overwhelms Taiwanese defenses, destroying military bases and communications infrastructure. At sea, the Chinese navy surrounds the country as hundreds of landing craft speed across the Taiwan Strait. As U.S. warships close in, they are overwhelmed by missiles, torpedoes, and drones. Videos of the Chinese blitz and sinking U.S. ships flood the internet. The demoralized Taiwanese decline to raise arms against the Chinese forces storming their shores.

This is the kind of narrative most people imagine when they think of military war games—scenes in the bowels of the Pentagon, units fighting digitally on electronic maps, commanders pondering their next step in a fast-moving crisis. Victory in the simulation, so the popular imagination goes, shows how to win a real-life conflict. Defeat in a war game, on the other hand, is an acknowledgement that any actual conflict will likely be lost.

Contrary to the popular imagination, however, this is not how war games work. Rarely is a war game designed to predict the future or develop a single definitive strategy. Instead, a war game helps military planners and analysts explore and understand a complex problem, regardless of the outcome. Win or lose, the purpose isn’t to define a strategy for the U.S. military but to help it better understand the capabilities it has, what it can already do, and what it needs.

In 2030, Beijing reaches a deadly decision: reunification with Taiwan at any cost. Cyberattacks plunge the island into darkness, confusion, and fear. A rain of missiles overwhelms Taiwanese defenses, destroying military bases and communications infrastructure. At sea, the Chinese navy surrounds the country as hundreds of landing craft speed across the Taiwan Strait. As U.S. warships close in, they are overwhelmed by missiles, torpedoes, and drones. Videos of the Chinese blitz and sinking U.S. ships flood the internet. The demoralized Taiwanese decline to raise arms against the Chinese forces storming their shores.

This is the kind of narrative most people imagine when they think of military war games—scenes in the bowels of the Pentagon, units fighting digitally on electronic maps, commanders pondering their next step in a fast-moving crisis. Victory in the simulation, so the popular imagination goes, shows how to win a real-life conflict. Defeat in a war game, on the other hand, is an acknowledgement that any actual conflict will likely be lost.

Contrary to the popular imagination, however, this is not how war games work. Rarely is a war game designed to predict the future or develop a single definitive strategy. Instead, a war game helps military planners and analysts explore and understand a complex problem, regardless of the outcome. Win or lose, the purpose isn’t to define a strategy for the U.S. military but to help it better understand the capabilities it has, what it can already do, and what it needs.

Whether it’s Taiwan or any other potential conflict, the scenario is rarely the focus of the war games we at CNA design for the U.S. Defense Department. Instead, war games are about better understanding how the U.S. military can build deterrence, what technology gaps could hobble its forces, how an adversary’s capabilities might evolve in response to U.S. capabilities, and how all that might impact what Washington should invest in today. Fundamentally, war games strive to explore and distill the fundamental nature of the problem itself—which rarely leads to definitive scenarios or solutions.

In fact, using war games to craft a clear-cut strategy is impossible. Done right, war games are a plausible method of providing a brief and limited glimpse into a possible future—a single future in a multiverse of possibilities. Trying to imitate victory in a war game, on the other hand, means trying to align both sides’ future decisions in a complex conflict with the scenario that played out during the game. Obviously, these decisions are numerous and mostly beyond one’s control.

When military leaders are unwilling to accept failure, they drive the entire Defense Department toward stagnation and safe bets.

Trying to avoid the failures identified in a war game presents a much richer avenue to explore. Did the U.S. side fail because of logistics shortfalls? Was its demise due to a missing capability—or an incorrect assumption about the adversary? War games can be designed to focus on just a single aspect of a conflict, such as the U.S. Navy’s deployment at the start of a Chinese attack on Taiwan, and to stress that aspect to the breaking point. Finding the conditions under which the U.S. side fails is invaluable, because it creates an opportunity to address a weakness. Success, on the other hand, makes it very hard to know where to invest and even harder to convince anyone the investment is worth making. Failure identifies the critical pieces that need attention, and the war game narrative highlights the cost of neglect. Through multiple iterations, a war game can steadily offer clarity on the way forward.

In 2030, Beijing reaches a deadly decision: reunification with Taiwan at any cost. Cyberattacks plunge the island into darkness, confusion, and fear. A rain of missiles overwhelms Taiwanese defenses, destroying military bases and communications infrastructure. At sea, the Chinese navy surrounds the country as hundreds of landing craft speed across the Taiwan Strait. U.S. warships provide critical missile defense to threatened U.S. regional bases. Long-range bombers and guided missiles decimate the landing craft approaching Taiwan, giving the country a brief window to rally its defenses and gather strength against the remaining Chinese troops who reached their shores.

For a war game to be useful, it’s largely irrelevant which potential conflict is played out. The game might use a Chinese invasion of Taiwan as the backdrop—but only as a use case for the military forces being examined. The real takeaways aren’t so much about victory or defeat in a particular conflict but about what’s exposed along the way. Why did deterrence fail in the scenario? Which of the adversary’s capabilities limited the United States’ ability to project force? Fine-grained analysis of questions like these drives progress in the U.S. military and allows it to increase its tactical, operational, and strategic proficiency. It forces hard examinations of information sharing, command and control, processes and procedures, and even organizational structures. None of these things is really about a Taiwan conflict. But if the United States did go to war—over Taiwan, a Baltic state, or any other country—the answers to these questions would help it succeed.

In the words of U.S. Air Force Gen. John Hyten, the U.S. side “failed miserably” during a 2020 war game that reportedly involved a Chinese assault on Taiwan. (The exact scenario remains classified.) But that result wasn’t the disaster it was made out to be by the press. Failure in a war game is a common feature, not a bug. Hyten’s remark proved the game had been a success. “[W]e have to make sure that we fail, and we fail quickly, and we learn from our failures and move fast,” he added in the same speech. When military leaders are unwilling to accept failure, they drive the entire Defense Department toward stagnation and safe bets, opening a window of opportunity for adversaries. The war game is a safe environment for failing fast—and learning to succeed.

In 2030, Beijing reaches a deadly decision: reunification with Taiwan at any cost. Cyberattacks plunge the island into darkness, confusion, and fear. Equipped with U.S. capabilities, Taiwanese defenses stave off the initial wave of Chinese missiles. At sea, the Chinese navy is attacked by swarms of unmanned submarines. As hundreds of landing craft speed unprotected across the Taiwan Strait, they are decimated by this unseen force. Over successive weeks, U.S. warships escort a steady stream of reinforcements. Supported by sensors from U.S. Marines and Special Forces, sophisticated missiles launched by long-range bombers sink the remaining fleet approaching the island. Taiwan begins repairing its infrastructure, mobilizes its reserves, and hunkers down for a long standoff.

The United States can certainly learn from winning a war game, but the lessons are usually narrow. What worked in a single war game has limited utility—it worked against a specific adversary making a specific set of decisions using a specific set of game rules that may or may not accurately reflect the world. Failure, on the other hand, doesn’t require the game to be a perfect simulation. We often hear complaints from players that our war game rules make the adversary “10 feet tall.” But it is better to stress U.S. forces more than to give the adversary too little credit and not stress U.S. forces enough. Stressing the capabilities of the U.S. forces to their breaking point from all sides allows analysts and researchers to identify vulnerabilities and what might be needed to fix them.

In 2030, Beijing reaches a deadly decision: reunification with Taiwan at any cost. Cyberattacks plunge the military ports on the Chinese coast into darkness, confusion, and fear. Hundreds of landing craft, unable to load at the ports, wait for further orders. Most of the Chinese missiles targeting Taiwanese military installations are intercepted by defensive batteries. At sea, the Chinese navy falls prey to advanced mobile mines and hypersonic missiles. U.S. warships lead regional allies and partners in shuttling critical munitions, supplies, and capabilities to the island. Although damaged, Taiwan’s infrastructure succeeds in rapidly transporting troops and supplies to defensive positions, readying the nation to repel the next wave.

There are many reasons why none of these narratives will come to pass. All it takes is one decision to shift the entire story. And in any war game—not to mention in real-life conflict—there are an untold number of decisions. Just as war games aren’t about victory or defeat, they’re also not critically determined by the so-called road to war, the initial circumstances that drive the game’s conflict. Instead, the dynamic heart of a war game is the way the players leverage and adapt current military doctrine, strategy, and capabilities—and apply them to the problems they face. Did predictable actions yield predictable results? Were there moments of insight or flashes of inspiration when players improvised? Novel decisions lead to new experiences, and new experiences can identify doors that lead to different and better outcomes.

So, in a war game, pay no attention to who won or lost. War-gaming is about the process, not the result—and analyzing that process is what will allow the U.S. military to turn losing into winning.

In 2030, Beijing reaches a deadly decision: reunification with Taiwan at any cost. The rest of the story? It’s up to you.

Jeremy Sepinsky is the lead war game designer at CNA and a former CNA field representative with the U.S. Marine Corps.

Sebastian J. Bae is a research analyst on CNA’s gaming and integration team and teaches war game design at Georgetown University and the U.S. Naval Academy. Twitter: @SebastianBae

Join the Conversation

Commenting on this and other recent articles is just one benefit of a Foreign Policy subscription.

Already a subscriber? .

Join the Conversation

Join the conversation on this and other recent Foreign Policy articles when you subscribe now.

Not your account?

Join the Conversation

Please follow our comment guidelines, stay on topic, and be civil, courteous, and respectful of others’ beliefs. Comments are closed automatically seven days after articles are published.

You are commenting as .

More from Foreign Policy

Soldiers of the P18 Gotland Regiment of the Swedish Army camouflage an armoured vehicle during a field exercise near Visby on the Swedish island of Gotland on May 17.
Soldiers of the P18 Gotland Regiment of the Swedish Army camouflage an armoured vehicle during a field exercise near Visby on the Swedish island of Gotland on May 17.

What Are Sweden and Finland Thinking?

European leaders have reassessed Russia’s intentions and are balancing against the threat that Putin poses to the territorial status quo. 

Ukrainian infantry take part in a training exercise with tanks near Dnipropetrovsk oblast, Ukraine, less than 50 miles from the front lines, on May 9.
Ukrainian infantry take part in a training exercise with tanks near Dnipropetrovsk oblast, Ukraine, less than 50 miles from the front lines, on May 9.

The Window To Expel Russia From Ukraine Is Now

Russia is digging in across the southeast.

U.S. President Joe Biden and Secretary of State Antony Blinken participate in a virtual summit with the leaders of Quadrilateral Security Dialogue countries at the White House in Washington on March 12.
U.S. President Joe Biden and Secretary of State Antony Blinken participate in a virtual summit with the leaders of Quadrilateral Security Dialogue countries at the White House in Washington on March 12.

Why China Is Paranoid About the Quad

Beijing has long lived with U.S. alliances in Asia, but a realigned India would change the game.

Members of the National Defence Training Association of Finland attend a training.
Members of the National Defence Training Association of Finland attend a training.

Finns Show Up for Conscription. Russians Dodge It.

Two seemingly similar systems produce very different militaries.