In this eye-opening breakdown of the CSIS Taiwan war simulation, a massive Chinese invasion quickly turns into a nightmare. Two U.S. aircraft carriers sink, 138 Chinese warships are destroyed, and tens of thousands of PLA soldiers end up as prisoners of war.
Despite heavy losses—including all of Taiwan’s navy and devastating strikes on Japan—the simulation shows China’s amphibious assault collapsing within three weeks. Taiwan’s ground forces, shore-based missiles, and timely U.S. intervention create a deadly “meat grinder” at sea.
Drawing from 89 detailed wargames, CSIS reveals the brutal reality: even if China lands troops, it cannot sustain them. The war ends with China’s navy shattered, while Taiwan survives as a democracy—at a horrific cost to everyone involved. A sobering look at why invading Taiwan may be far harder than Beijing expects.
Hi, I'm Dong Xiong. Welcome to Digging into China, where we take a deep dive into what's
really happening in China and share some unique insights. Two U.S. aircraft carriers have sunk
to the bottom of the Pacific. China's navy has lost 138 major warships. Tens of thousands of
soldiers have become prisoners of war. Taiwan's entire fleet of 26 destroyers and the frigates
have been wiped out. And even Japan has come under ballistic missile attack.
What you just heard isn't some feel-good fantasy script or online military fanboy trash talk.
It's the result of a major U.S. defense think tank, CSIS, spending three years and running
89 different simulations. What we are going to walk through today is the most detailed and
realistic version among those 89 runs. They drew on real U.S. military combat data, brought in
retired generals and experts from the Naval War College, and did their best to model what would
actually happen if China ever tried to invade Taiwan. So in the end, who wins? And how on earth
does it play out like this? Let's break it down step by step. The war is set to kick off in
October of some year. Why October? Well, the Taiwan Strait has a nasty temper. Typhoons in summer,
monsoons in winter, thick fog in spring. Out of the whole year, only April and October offer decent
windows. And October happens to fall right in that political low during the U.S. election cycle.
In the weeks leading up to D-Day, satellite images across the strait start showing strange
activity. PLA Type 071 dock landing ships and Type 075 amphibious assault ships begin
massing along the coasts of Fujian and Guangdong province. At the same time, hundreds of civilian
row-on-row of ferries and deck cargo ships quietly slip off their commercial routes and head for
military ports. Rocket-force missile launchers roll out of their mountain tunnels, and long-range
rocket artillery systems move into positions on the opposite shore. This kind of build-up is
impossible to hide. Satellites from every major country see it crystal clear, just like when
Russia massed troops on Ukraine's border in early 2022. The whole world knows something big is coming.
It's Tuesday morning in October, 4.07am. You are jolted away by massive roar. You rush to the
light on the horizon. But it's not the sunrise. It's the glow of fires coming from the direction
of Songshan airport. Thousands of ballistic and cruise missiles are streaking in from the Chinese
mainland. The PLA's target list puts Taiwan's airbase runways at the very top, Hualien, Taitung,
Chayi, Qinchuankang, Songshan. Next come the naval bases, Zuoying, Qilun, Suao.
Taiwan's 26 destroyers and frigates get hammered right at the docks. The CSIS report puts it
bluntly. Taiwan's air force and navy are largely destroyed in the first few hours of the war.
Radar sites, air defense positions, and parts of the power grid came under attack right after.
Meanwhile, the Chinese crossing operation across the strait has already begun,
even as the missiles are still flying. The entire PLA navy amphibious fleet surges out,
around 70 landing ships of various types streaming out from Fujian and Guangdong,
followed closely by requisitioned civilian vessels, those same passenger roll-on-roll-off
ferries you see on regular routes, now packed with infantry fighting vehicles and tanks,
ready to drive straight onto the beaches and unload. According to CSIS modeling,
all this amphibious lift combined can move about 20,000 troops in the first few days.
Sending 20,000 men against an island of 23 million, that's basically crazy. So they are
there to seize a beachhead, then feed in wave after wave of reinforcements,
plain and simple. These 20,000 are the tip of the spear suicide squad, to keep them from becoming
cannon fodder. The PLA has to maintain sea control the whole time. In this simulation,
the Chinese decide to roll the dice. They go for surprise and choose to land in the north,
where Taiwan's defenses are actually the strongest. The main effort is aimed at the Taoyuan
area. Right now, Xi Jinping is pacing back and forth in a bunker at Beijing's western hills.
He is betting on one big thing, that the United States won't jump in immediately.
At the same time, DF-41 intercontinental ballistic missiles silos in Gansu and Xinjiang
finish their fueling checks, and Type 094 ballistic missile submarine slip into the
deep ocean. The message is loud and clear. We are nuclear power. Don't even think about it.
Back in the White House, they are facing one of the toughest decisions in human history.
Intervene, and you risk triggering all-out war with a nuclear power and crashing the global
economy. Stay out, and Taiwan falls, wiping out 70 years of U.S. alliance credibility across the
Asia-Pacific in a single night. In the CSIS simulation, the U.S. chooses to intervene.
The analysts figure the long-term cost of staying out would be far worse than the short-term pain
of getting in. But that decision still takes dozens of hours to execute. And in those dozens
of hours, China is frantically shipping troops across to Taiwan. One line in the report really
hits home. Delay and half-hearted intervention will only increase future American casualties.
Every extra hour you hesitate could mean another ship going down later.
The President gives the order. But help is too far away to arrive in time. Carrier strike groups
are racing in from all over the Western Pacific. It'll take days. U.S. fighters based in Japan
still need Tokyo's permission before they can take off. So the first forces to reach the fight
are the nuclear attack submarines already on patrol in the Western Pacific. Virginia class
and Los Angeles class boats. They can slip through China's defense and get close to the Taiwan
Strait. One MK-48 heavy torpedo can break a 10,000 ton warship in half. But the average depth of
Taiwan Strait is only about 60 meters, far from ideal hunting grounds for the subs. So the American
submarines can't do all that much. But their surprise attacks still throw the Chinese convoys into
total chaos and slow down the crossing. At this stage, both sides are really fighting over just
one thing, time. China is racing to get troops onto the beachhead. The U.S. is racing to bring
more firepower into the theater. Right behind the subs come America's real aces, B-2 stealth bombers
and B-1B Lancers flying straight from the U.S. mainland. They don't need to enter China's air
defense zone because their key weapon is the AGM-150AC long-range anti-ship missile. These
missiles have a range of over 900 kilometers. They are stealthy and they can skim the waves.
You just program the target area and the missile finds the highest value targets on its own.
Picture this scene, a group of B-1Bs sitting 800 kilometers outside China's air defense envelope.
All release hundreds of stealthy anti-ship missiles at once. The missiles hug the sea surface,
almost invisible to radar. Then in the same time window, they come screaming in from different
directions and angles towards the Chinese ship's convoys in the Taiwan Strait. That's the main
reason those 138 Chinese warships get taken out early on. But these missiles have one fatal
weakness. There aren't very many of them. CSIS estimates that by 2027 the U.S. inventory might
still be under 500, and there aren't that many planes that can launch them. So if those 500
run out before China's crossing capability is broken, we are in real trouble. The good news
from the simulation? 500 turned out to be more than enough because the Chinese convoys were so
tightly packed that almost nothing was wasted. Let's hit pause right here for the quicker side.
A lot of people online say Taiwan will be over in three days. You wait for the U.S. president
to decide, then wait for the U.S. planes to arrive. At least 96 hours have passed. The Taiwan
presidential office would already have a Chinese flag planted on it, right? How could they still
be moving troops? I am not here to argue, but the conclusion is pretty sobering. They are still
moving them. Let's do the math on how many trips those ships could make in three days. The Taiwan
Strait is 130 kilometers across. Loading at mainland ports, queuing, embarking troops,
vehicles and supplies, that alone takes at least 4 to 6 hours. The crossing takes 4 to 5 hours.
Then at the Taiwanese beachhead, using ramps to unload onto the sand and possibly taking fire
while doing it, that's even slower than landing. So at least 6 to 8 hours. Empty return ship, another
4 to 5 hours. One full round trip takes at least 24 hours. And that's in ideal conditions where
no ships get hit and the beach isn't jammed. So by the fourth day, the Chinese convoys might have
managed only about 3 round trips at most. According to CSIS, the PLA's initial landing
capability on day 1 is only around 8000 troops. So landing 20,000 in 3 days is already a pretty
aggressive estimate. Taiwan's role obviously can't just be to absorb the first missile barrage and
then wait for the Americans to arrive to the rescue. The US subs won't join the fight until
at least 2 or 3 days in, and the bombers even later. Taiwan's defense though starts the moment
these missiles lift off. Yes, its air force and navy take a brutal hit in that first wave. But
Taiwan's real anti-ship firepower is actually sitting on trucks on the ground. Xiongfeng's
three anti-ship missile launchers can hide inside city factories or under highway overpasses.
They fire one, then relocate a few kilometers away and fire again in 10 or 15 minutes. Even
if the Chinese gain air superiority, they can't do much about it. On top of that, there are the
hundreds of Harpoon shore-based anti-ship missiles the US has sold Taiwan in recent years.
Plus the Xiongfeng II's, Taiwan's west coast is already covered in a dense anti-ship missile net.
The total number of various anti-ship missiles exceeds 1,400. In other words, the last few
dozen kilometers before the Chinese convoys reach shore is the most heavily defended kill zone with
Taiwan's shore-based missiles. The carnage there can only be described as a meat grinder at sea.
Especially those unprotected civilian row-on-row of ships, one order anti-ship missile is enough
to send them straight to the bottom. So in the CSIS simulation, Taiwan's own shore-based
missiles form one of the three critical kill chains. Let's shift the view back to the Chinese
side. Of course, they are not just sitting there taking punches. PLA fighters jets are
dueling with US forces over the strait, J-20s, J-16s, Su-30s, more than a thousand of them.
In the CSIS report, America's 5th generation fighters clearly outperform China's 5th gen
jets and bring in more combat experience. But China has the numbers. More importantly,
China now faces a hellish strategic choice. Do they strike US bases in Japan or not? If they
don't, American F-35s and F-18s will keep launching in endless waves from Japan. Within US operating
range, there are roughly 600 usable airfields. Without hitting those bases, American air power
would be basically unlimited. But if they do hit them, Japan automatically becomes a belligerent.
In the simulation, China chooses to strike. Once their convoys start getting slaughtered,
they have no way out. Not striking would only make them lose faster. If the convoys
are completely wiped out, Beijing's domestic strong nation narratives would collapse.
So Chinese ballistic missiles start falling on Okinawa and Kedona Air Base. The logic in
Beijing at this moment is a lot like Japan's when it attacked Pearl Harbor. They are betting America
won't hit back with everything it has. If China can take Taiwan in three days, it will be stuck
in a very bad spot. The ballistic missiles begin raining down furiously on US bases in Japan.
There is a chilling passage in the original CSIS text. When follow-on US units arrive at Kedona,
what they see is entire squadrons of wrecked aircraft pushed to the side of the runway by
bulldozers. The base hospital is packed with hundreds of wounded, and what comes next is the
shocking centerpiece of the entire Taiwan war simulation, the American aircraft carriers.
To China, US carriers are no longer just military targets, they are political targets.
Since World War II, America has never lost a supercarrier in combat. It's become a symbol
of American global power. So for Beijing, even if they ultimately lose the war, if they can take
out even one American carrier in a big way, the domestic narrative can flip instantly.
China is willing to pay any price. They unleash the carrier killer DF-21D that they've been
preparing for over 20 years. Along with all kinds of anti-ship missiles launched from destroyers,
submarines and bombers, these missiles come in from different directions and attitudes at the
same time, swarming the American carrier group. Even with America's strongest Aegis defense system,
100% interception is impossible. Facing this all-out saturation attack, the baseline scenario
in the SCIS simulation ends with the US losing two aircraft carriers, 10 to 20 large surface
combatants, and over 10,000 sailors in the water. Of course, the other side of the story is that
China's navy is being slaughtered at the same time. In three weeks, China loses 138 major warships
and 155 fighter jets. 90% of its amphibious fleet is destroyed. Now let's swing the camera back to
the beaches of Taiwan. By the second week of the war, China has managed to land about 37 battalions,
roughly 30,000 troops, mostly concentrated on beachheads in the northern Taoyuan area.
It sounds like progress, but it's actually a disaster. 90% of the amphibious ships are now
on the ocean floor. Their sea lines of communication, which they depend on for survival,
have basically ceased to exist. So these 30,000 soldiers are running low on ammunition,
diesel is nearly gone, and there is no way to evacuate the wounded. Even worse, once those
30,000 men hit the beach, they discover there is almost no room to advance. Either they plunge
straight into the urban streets of Taoyuan for brutal block-by-block fighting, or they try to
maneuver in the surrounding rice paddies. Their armored units get stuck, the tanks sink in and
can't get out. Meanwhile, Taiwan's army is closing the ring. Missiles can sink warships
and crater runways, but they can't eliminate every infantry company and anti-tank teams
scattered across the cities. Once Taiwan issues the reserve mobilization order, about 73% of
reservists make it to their assembly points. That means over 200,000 additional troops join the
fight, far more than China's 30,000. So one of the core conclusions in the CSIS report is this,
Taiwan's ground army is the real decider in the whole battle. As long as the ground forces can
contain the beachhead and wait for China's sea logistics to collapse, time works in Taiwan's
favor. Of course, China will do everything it can with long-range strikes to slow down the
Taiwanese reserves, blowing bridges, cutting roads, but nothing can stop 200,000 men from walking or
riding motorcycles along backcountry paths to reach the frontlines. By the third week, the Chinese
landing force is in desperate straits. Ammunition is basically gone. Sure, some people will say,
what about air resupply? Taiwan's air defense was badly degraded at the start, but manned portable
air defense missiles and mobile systems scattered throughout the cities are still deadly, just like
Ukrainian Stinger teams taking down Russian helicopters. In the end, those 30,000 Chinese
soldiers out of food and bullets face two choices, fight to the last man, or surrender. In the original
CSIS text, China's navy is shattered, its core amphibious capability broken, and tens of
thousands of soldiers end up as prisoners of war. The CSIS simulation highlights something we often
overlook, how important Japan is. If the US military couldn't use bases in Japan, Taiwan would
probably fall. China knows this perfectly well. So when American forces start operating from Japan
bases, China is forced to strike Japan. And once Chinese missiles land on Japanese soil, Japan's
self-defense forces join the war fully. In the CSIS simulation, Japan also takes heavy losses,
over 100 fighter jets and 26 warships. Three weeks, 21 days, that's roughly how long the whole war
lasts in the simulation. Some might ask, China has such strong industrial power, how could it
possibly stop after just three weeks? The reason it stops isn't because China's national strength
ran out, it's because once China's sea transport capacity for the crossing is destroyed, the war is
effectively over. No ships means no more troops. No troops means they can't hold the beachhead.
Without the beachhead, the little bit of ground they seized has to be given back. After that,
all that's left is shouting across the water. Final casualty count from the simulation. The US
lost two aircraft carriers and 10 to 20 large surface warships, roughly 200 to 400 aircraft,
and 3,200 troops killed. That casualty figure is already about half the total American deaths
from the entire 20 years of the Iraq and Afghanistan wars combined. For China, 138 major
warships sunk, 155 fighter jets lost, 90% of the NPBS fleet wrecked, about 10,000 soldiers killed,
and tens of thousands taken prisoners. China's navy is essentially gone. For Japan, over 100
aircraft lost and 26 warships sunk. For Taiwan, all 26 destroyers and frigates sunk, army casualties
around 3,500. The worst part is the devastating blow to Taiwan's economy and the massive
reconstruction that lies ahead. But Taiwan survives as a democratic entity. The CSIS
description says Taiwan's military did not collapse, though it suffered serious losses,
and it will have to continue defending from an economically shattered island.
From 2023 to 2025, this top strategic think tank in Washington, CSIS, ran a total of 89 simulations
of a Taiwan Strait war. 24 of them modeled a full-scale amphibious invasion. The conclusion
from those 24 runs is that in most scenarios, China's invasion fails, but everyone pays a
terrible price. China did win in a few cases, but only when certain necessary conditions weren't
met. There are four key ones. First, the US must be able to use bases in Japan. Without access to
bases on Japan's Japanese soil, American fighter squadrons can't maintain a high enough sortie rate
over the Taiwan Strait. Second, you absolutely cannot repeat the Ukraine model. In Ukraine,
the US and NATO could keep pouring in weapons by land after the war started. But Taiwan is an
island. Any idea of slowly shipping supplies after the shooting begins is suicidal. Everything has to
be pre-positioned on the island before China attacks. Third, the United States must stay
outside the range of China's air defense envelope and use bombers and nuclear submarines to fire
large numbers of long-range anti-ship missiles. If the inventory of those long-range missiles is
too low, even if America intervenes, the loss ratio becomes unbearable. And finally, the most
important one, Taiwan cannot surrender. It must hold its lines where it stands. Taiwan does not
surrender is the premise in every one of these simulations. If Taiwan's ground forces collapse
or give up before American help arrives, then all the follow-on reinforcements are pointless.
I have included the link to the full CSRS simulation report in the program description.
Feel free to check it out for the complete strategic breakdown. Thank you so much for
watching. If you find this video helpful, please like it, drop a comment below, and hit the
subscribe button right here. It really helps YouTube suggest my video to others. I'll be back
with more content soon. Until then, be well.