Game theory and nuclear brinkmanship

[mr friendly guy]

I need some help in explaining the application of game theory in nuclear brinkmanship, because on another board someone tried to explain it, but essentially seems to obfuscate the issue using buzzwords without actually explaining anything.

The amateur war game situation is this - essentially according to them (ie according to game theory), if China were to suddenly say any conventional attack (ie non nuclear attack) against it will face nuclear retaliation on China's part, no one including the US will dare to attack. Thus China should scrap a "no first use" policy in regards to nukes.

My first thought is what happens if a nation, lets say the US tried to call that bluff and saying I will attack with convention weapon and see if China is willing to escalate with nukes. The outcomes are thus, a) China institutes nuclear armageddon (lets be generous and assume that even with ABM capabilities both sides render each other no longer great powers) and b) China doesn't escalate, keeps it conventional and loses. From a rational point of view it seems b as undesirable as it sounds is the best choice. The US would of course be able to see that and calls out China's bluff.

However according to them, that in game theory the US would fold, thus China doesn't need to really have to make the hard decision about whether to escalate to nukes. When I asked why that is so, I get a link and people saying that the game is not static and that both sides are making decisions at the same time (does that mean both sides have decided to attack at the same time?).

This seems a bit strange to me, because if any side believed that, why would they spend gazillions of dollars on conventional weaponry at all? Why would China invest money into anti aircraft carriers weapons such as attack submarines, studies on a hypothetical ASBM, if all they needed to do was declare we will use nukes first?

Can someone explain the application of game theory to nuclear brinkmanship for me?

[madd0ct0r]

you know the classic prisoner's dilemma?

nuclear war is 'both go to jail option' - everybody looses. Due to the time the missiles take, there is also no 'We nuke them, but they don't nuke us' option.
Looking at the cuban missile crisis.

USA wants to prevent missiles on cuba.
USSR wants missiles on cuba.

Being rational actors, neither of them will nuke the other if they don't get what they want, and both know the other person is thinking that. Basically, nuclear war is too big a threat to be credible.
So how do you use the nuclear war threat to force the other to back down?

Brinkmanship: letting the situation get closer and closer to nuclear war, letting the situation slip slowly and carefully out of control. What is a general goes rouge? what if there's a missile malfunction? what if a pilot accidentally crashes into the wrong ship? Any of these could be the spark.
What you are doing is allowing the chance of nuclear war to slowly rise, and hope the other guy backs down first.
He's rational, you're rational, and having missiles on cuba is worth less to him than NOT having them is to you. When the risk becomes too great, he should back off. He knows you know this, and ordinarily might try and hold on a little longer, but with the chance of nuclear war triggering steadily increasing, holding on longer is a mugs game.

For your situation, look at pakistan vs india. Pakistan can't beat india in a ground war, but can use the threat of the risk of nuclear war (by slowly letting the situation slip out of control) to cause india to back off. 2005 i think.
The same would apply for china.

[mr friendly guy]

So it becomes a giant game of chicken? So you're saying its a matter of which side is less risk averse being the side to back off. Which begs the question, if one side is so confident the other side will back off, why bother investing in conventional weaponry aimed at the other side (as opposed to conventional weaponry to counter a non nuclear state). Theoretically "playing the game" and being more risk averse should be sufficient.

[Dr. Trainwreck]

Because you never know when you have to smack down somebody where nukes aren't involved and, realistically, shouldn't be (say, Iraq).

Against nuclear states it's less useful but, if you have enough hardware, you might get to use some of it as soon as nuclear arsenals are depleted. That's not a strictly rational thing to do, but at that point people would only think about giving the enemy a final fuck you and stomping them as deep in the ground as possible.

[madd0ct0r]

Conventional weaponry is also useful as a smaller stick to threaten people with over smaller stuff (like tiny bits of border).

If you threatened a neighbor you'll blow up the town if he keeps stealing your lettuce, he probably would take that less seriously then if you threatened to give him a beating with a (literal) small stick.

Pakistan would never credibly threaten India with nukes over kashmir, but as a response to a full India invasion of the rest of Pakistan it's much much more credible.

That's one way the UN has been quite successful, they very very very rarely recognize a change in a country's borders, so it reduces the drive for the small land grabs that don't deserve nuclear annihilation. If you can't keep, why bother stealing it? (and so less bushfire wars)

[Algebraist]

While two nuclear powers have yet to get into direct war with each other they can still exert influence by helping/invading other countries in a conflict using conventional weapons. This can be done by purely supplying the conventional weapons or getting troops on the ground as well.

Obviously this happened a lot during the cold war - examples such as Vietnam/ Afganistan where US/Soviets put troops on the ground but the Soviets/US respectively helped the opposition by supplying weapons. What both powers however avoid is having their own troops directly confront each other and the support of an opposition can be covert. Thus you are involved in opposite sides without having to declare war on each other to escalate the situation.

[mr friendly guy]

I am not talking about using conventional weapons against a non nuclear state. So if China builds an aircraft carrier with the view that it might need to slap down a third world shithole sometime in the future, I get that.

However some of the weapons such as attack submarines and a hypothetical ASBM is mainly against aircraft carriers, and its likely to be an anti-US weapon rather than against a non nuclear state. Again why would you do that if you believed nuclear brinkmanship is itself sufficient?

Sure you can say that they will still develop conventional weapons to fund some other nation which might oppose the US interest. But if that was the case, then you wouldn't expect them to equip their military with anti aircraft carrier weapons, because nuclear brinkmanship should be sufficient (if they believed game theory).

It seems like people argue nuclear brinkmanship is enough but want to have other options in case. Now there is nothing wrong with having more options, but by implication it means that they have doubts about game theory and nuclear brinkmanship.

[Algebraist]

I am not talking about using conventional weapons against a non nuclear state. So if China builds an aircraft carrier with the view that it might need to slap down a third world shithole sometime in the future, I get that.

However some of the weapons such as attack submarines and a hypothetical ASBM is mainly against aircraft carriers, and its likely to be an anti-US weapon rather than against a non nuclear state. Again why would you do that if you believed nuclear brinkmanship is itself sufficient?

Sure you can say that they will still develop conventional weapons to fund some other nation which might oppose the US interest. But if that was the case, then you wouldn't expect them to equip their military with anti aircraft carrier weapons, because nuclear brinkmanship should be sufficient (if they believed game theory).

It seems like people argue nuclear brinkmanship is enough but want to have other options in case. Now there is nothing wrong with having more options, but by implication it means that they have doubts about game theory and nuclear brinkmanship.

Yes I doubt they are completely confident of nuclear brinkmanship - given the stakes some doubt is not unreasonable. Although you could also end up with some limited conflict with conventional weapons as part of that brinkmanship. Also not all weapons developements are completely logical.

As you mentioned attack submarines - that's one area where a war game between the US and USSR was ongoing but without actually resulting in direct conflict. The idea being that if you had the an overall advantage with nuclear submarines its possible one nation could decide it had a sufficient advantage to launch a nuclear attack to obliterate the opposition without suffering major damage itself. Thus the ongoing game to keep tabs on each others subs - they would hunt each other. The reality is however that both sides were still pretty much assurred of mutual destruction on the event of full out nuclear war.

[xthetenth]

I am not talking about using conventional weapons against a non nuclear state. So if China builds an aircraft carrier with the view that it might need to slap down a third world shithole sometime in the future, I get that.

However some of the weapons such as attack submarines and a hypothetical ASBM is mainly against aircraft carriers, and its likely to be an anti-US weapon rather than against a non nuclear state. Again why would you do that if you believed nuclear brinkmanship is itself sufficient?

Sure you can say that they will still develop conventional weapons to fund some other nation which might oppose the US interest. But if that was the case, then you wouldn't expect them to equip their military with anti aircraft carrier weapons, because nuclear brinkmanship should be sufficient (if they believed game theory).

It seems like people argue nuclear brinkmanship is enough but want to have other options in case. Now there is nothing wrong with having more options, but by implication it means that they have doubts about game theory and nuclear brinkmanship.

Nuclear brinksmanship by its very nature is a tool to be used sparingly and only in times when letting the risk of nuclear war get to where the other side backs down is less of a risk than them not backing down. Nuclear weapons are pretty much all or nothing existential sorts of things, so if you just have the nukes fall most things die option, then what happens if someone goes after your interests enough to cause harm but not enough to go nuclear over. That's where we get the proxy wars of the Cold War from, things that each country decided warranted force but most certainly didn't warrant nukes or even nuclear sabre rattling. For example, we got Korea and Vietnam from the domino theory and US fears that if communism spread to enough countries the USSR could assemble enough force to outstrip the US and be able to actually defeat it (very simplified). So that meant the US needed conventional forces (actually conventional forces of two flavors, stuff for lower intensity fights in the third world and stuff for a high intensity fight in Europe since there was a danger that the USSR could decide that the US wouldn't risk nuclear war over Europe and take it over). When MAD became more and more of a thing that could very well deter the US, a lot of Europe wasn't happy because they started to wonder whether the US wouldn't let them get overrun, and that's a major part of the genesis of the French nuclear deterrent.

Now what happens if the US goes after Chinese interests that aren't worth a nuclear war with conventional forces? Stuff like you're talking about is a conventional deterrent to the sort of attack that is a threat but isn't big enough to make brinksmanship worth the costs as far as I can tell. If it's not a huge danger the way ICBMs in Cuba are, the other side may well call the bluff, but with conventional capabilities to make such action not worth it hopefully they won't start in the first place.

[madd0ct0r]

for instance - if china is 'intervening' in a civil war in africa to make sure their man gets appointed, and they get access to the minerals in that country, they might need enough conventional high tech muscle to dissuade the US from intervening to prevent china intervening, and so on.

[Irbis]

This seems a bit strange to me, because if any side believed that, why would they spend gazillions of dollars on conventional weaponry at all? Why would China invest money into anti aircraft carriers weapons such as attack submarines, studies on a hypothetical ASBM, if all they needed to do was declare we will use nukes first?

Because nuclear weapons are strictly defensive in nature? You can't invade island on the middle of Pacific with them, nor do virtually anything but bully people into not invading you. Take note that once China acquired nukes they started to delete huge number of poorly trained militia units - since they were useful only in defence, too, they become redundant thus pointless to keep.

Now, the other things you mention, like carriers or submarines are used for power projection - enhancing your diplomacy and allowing you to slowly change status quo without ever coming close to nuclear threshold.

However some of the weapons such as attack submarines and a hypothetical ASBM is mainly against aircraft carriers, and its likely to be an anti-US weapon rather than against a non nuclear state. Again why would you do that if you believed nuclear brinkmanship is itself sufficient?

Again, change in status quo - USA was once able to park SSBN right next to Shangai and send the whole Chinese leadership pining for fiords before anyone could react. Now that China has defensive measures, status quo changed into USA thinking twice before sending ships into area to which China can now project power. Similarly with carriers - having conventional measures against them not only gives you options other than WW III once one comes closer to your shore, it makes USA less likely to support Taiwan (and allows China to more effectively protect its interests in Africa), thus lessening the risk of war.

[Alyrium Denryle]

I need some help in explaining the application of game theory in nuclear brinkmanship, because on another board someone tried to explain it, but essentially seems to obfuscate the issue using buzzwords without actually explaining anything.

The amateur war game situation is this - essentially according to them (ie according to game theory), if China were to suddenly say any conventional attack (ie non nuclear attack) against it will face nuclear retaliation on China's part, no one including the US will dare to attack. Thus China should scrap a "no first use" policy in regards to nukes.

My first thought is what happens if a nation, lets say the US tried to call that bluff and saying I will attack with convention weapon and see if China is willing to escalate with nukes. The outcomes are thus, a) China institutes nuclear armageddon (lets be generous and assume that even with ABM capabilities both sides render each other no longer great powers) and b) China doesn't escalate, keeps it conventional and loses. From a rational point of view it seems b as undesirable as it sounds is the best choice. The US would of course be able to see that and calls out China's bluff.

However according to them, that in game theory the US would fold, thus China doesn't need to really have to make the hard decision about whether to escalate to nukes. When I asked why that is so, I get a link and people saying that the game is not static and that both sides are making decisions at the same time (does that mean both sides have decided to attack at the same time?).

This seems a bit strange to me, because if any side believed that, why would they spend gazillions of dollars on conventional weaponry at all? Why would China invest money into anti aircraft carriers weapons such as attack submarines, studies on a hypothetical ASBM, if all they needed to do was declare we will use nukes first?

Can someone explain the application of game theory to nuclear brinkmanship for me?

The person you are talking to is an idiot who does not understand how to construct a Nash Equilibrium.

Unless a US response to chinese expansion (in this case), poses an existential risk to China, there is no reason to use nuclear weapons. Even then, the application of nuclear weapons in defense--once under a MAD scenario--is strictly spiteful in nature. There is no payoff for you, only preventing the other side from obtaining their payoff. Before MAD, when the number of nuclear weapons was such that they could be used as strategic or even tactical weapons (to kill russians swarming across the Elbe, for example), this was not the case... but this very thing is what created the nuclear arms race in the first instance.

In all possible cases once the number of nuclear weapons on both sides is sufficiently high, the risk of using nuclear weapons massively outweighs the risks of not using them, and there is no payoff to counteract the risk. No one wins in a nuclear exchange--not even the survivors of the initial blast. In fact, being among the first to die IS being the winner of a nuclear exchange.

So, why was the cold war an issue? Why did we come so close to nuclear apocalypse on so many occasions?

People are not perfect Nash Equilibrium calculators . There are other drivers in the system. Cognitive biases that drive a desire to "win". A tendency to escalate social tensions, combined with the fact that our brains did not evolve under in a state of mutually assured destruction, and we dont accept it very well. There was a faction that thought we could "win" a nuclear war with the soviet union. All manner of insane things like "duck and cover" were invented that served as rationalizations. Bomb shelters, warning systems, missile defense, response planning. All of it was a rationalization that created the illusion that there was a meaningful future for us after nuclear war.

But... this also creates an incentive to have an escalation policy. It is not an instant game of nuclear chicken, it is a slow one. The payoff matrix has more than two options. Say the US and China have competing interests in the Indian Ocean. Diplomacy fails. Force is initiated. Both countries go through proxies at first. The signal being "we are fighting you, but we have no intention of fully committing, and we are pointedly not posing an existential threat to you.". And it goes from there. Each side pushing the other until the payoff to national interest equals the risk of a nuclear strike--and thus MAD. The Cuban Missile Crisis was an example of precisely this. Both sides backed off, but it was a VERY close thing. Had someone made a mistake in their back of the napkin risk calculation, civilization as we know it, would have ended. And because we are not perfectly rational game theory players, those mistakes are very easy to make.

[Simon_Jester]

I will say that it's not as... pure-deductively simple as saying "defenses are an illusion." People have done the math; it makes a difference. Warning systems reduce the risk of a sneak attack. Bomb shelters help to ensure that small towns survive the fringe effects of an attack (which can make the difference between "50 years to rebuild" and "200 years to rebuild"). Defenses have a HUGE effect, example:

-British nuclear deterrent consisted of a modest force of missile submarines. By targeting each missile on a different city, the British could kill a huge number of Russians and devastate the Soviet heartland. But the core of the target for the deterrent was Moscow.
-Now the Soviets deploy a ring of ABM sites around Moscow. They're not perfect, but they MIGHT shoot down several missiles headed for Moscow, and it's hard to be sure.
-Suddenly, if you want to hit Moscow reliably, you need to launch enough missiles to put enough warheads through the defenses that you can land several in and around the city. That is a lot of missiles, missiles which are NOT targeted at other Soviet metropoli.

The net result is that a British nuclear deterrent that could have knocked the Russians back to the economic equivalent of 1920 if not 1870 is now turned into a "blow up Moscow" deterrent. Now, blowing up Moscow would be a huge deal, but the Soviets can survive it, and it totally changes the game.

Similar logic applies to building ABM sites around military installations. It makes the installations very, very hard targets to kill reliably, which vastly inflates the amount of firepower the enemy has to throw at hardened bunkers in remote areas instead of at your cities and transportation infrastructure. So while it doesn't make nuclear war 'winnable,' it tends to change the profile from "post-apocalyptic nightmare for all" towards "living in post-Black Death Europe."

[phongn]

People are not perfect Nash Equilibrium calculators . There are other drivers in the system. Cognitive biases that drive a desire to "win". A tendency to escalate social tensions, combined with the fact that our brains did not evolve under in a state of mutually assured destruction, and we dont accept it very well. There was a faction that thought we could "win" a nuclear war with the soviet union. All manner of insane things like "duck and cover" were invented that served as rationalizations. Bomb shelters, warning systems, missile defense, response planning. All of it was a rationalization that created the illusion that there was a meaningful future for us after nuclear war.

Lets not get carried away here: nuclear war is entirely winnable (and not as in "we have two people and they have one", as in "we have an intact civilization and their nation is a smouldering ruin"). Both sides in the Cold War extensively gamed this and knew very well things were winnable (the USSR, in particularly, knew that until the mid-late-1960s or so the US would win, though at the cost of much of industrialized Eurasia destroyed).

But... this also creates an incentive to have an escalation policy. It is not an instant game of nuclear chicken, it is a slow one. The payoff matrix has more than two options. Say the US and China have competing interests in the Indian Ocean. Diplomacy fails. Force is initiated. Both countries go through proxies at first. The signal being "we are fighting you, but we have no intention of fully committing, and we are pointedly not posing an existential threat to you.". And it goes from there.

Kahn's escalation ladder pretty much is a distillation of this. Both sides knew very well that it was entirely possible to stumble into a nuclear war (and took pains to try and avoid that, e.g. the hot line).

Each side pushing the other until the payoff to national interest equals the risk of a nuclear strike--and thus MAD. The Cuban Missile Crisis was an example of precisely this. Both sides backed off, but it was a VERY close thing. Had someone made a mistake in their back of the napkin risk calculation, civilization as we know it, would have ended. And because we are not perfectly rational game theory players, those mistakes are very easy to make.

As a nitpick, the Cuban Missile Crisis wasn't that close - the USSR fully knew their strategic weakness. They just thought Kennedy was so weak he'd be pushed into acceptance (and his earlier performance at various summits only reinforced this idea in the Soviets' min).

But no, life will go on. It will be less pleasant, there will be many dead, but civilization will continue, minus perhaps a couple hundred years of progress.

[Esquire]

Plus, there's always the chance somebody will develop a reliable ABM system, at which point relying entirely on nuclear deterrent will become incredibly stupid. Current conventional military spending serves a valuable purpose as an investment in future capability.

[Sea Skimmer]

Effective ABM just rules out killing people in 40 minutes after a 3 minute countdown. It doesn't rule out nuclear deterrence via aerodynamic weapons; and indeed war technology of all sorts has a way of becoming self countering. The lightweight materials and compact low cost computing and other gear which has helped make ABM possible (without using nuclear tipped interceptors) has also helped make much better air launched cruise missiles feasible for a much wider spectrum of the world, and the field of hypersonic weapons is just opening up. We've also simply not seen a bomber yet with better then 1980s technology. All existing designs are from that era and yet some massive improvements have taken place in all forms of aircraft technology. Both Russia and the US have serious projects for new bombers, China is unknown as far as new projects go, but making new versions of the H-6 Badger, which as old and small as it might be is still capable of attacking CONUS with a single inflight refueling.

Defending against these sorts of weapons is actually more demanding and expensive then ABM in most respects; ICBMs are kind of a simple brute force solution to a delivery system.

[Esquire]

Reliably effective anti-insert-delivery-method-here systems, then. All I'm saying is that it would be silly to assume what's a world-ending, unbeatable weapon now will continue to be indefinitely, and that therefore you can't just say "I have nukes, therefore all other military spending is useless forever."

[Simon_Jester]

Yeah, you still have to keep investing in delivery systems.

Although it's a lot easier to stay an effective nuclear threat if your aims are defensive, because all you have to do is make an attack on you unpalatable. Having tactical nuclear weapons in your arsenal makes you more or less immune to invasion by the Big Menacing US Ground Army, unless you give the US a damn good reason, because we're not willing to lightly risk losing numerous billions of dollars and thousands of lives for each shot that gets through our defenses.

And having a strategic nuclear arsenal that has trouble penetrating enemy defenses BUT can at least get enough through that one or two key cities could be hit is probably enough of a deterrent to keep the enemy from invading you, even if it's not enough to let you get away with playing imperialist games in their backyard.

[Guardsman Bass]

I thought the point of ABM was that it specifically blunted ICBMs, one of the most destabilizing set-ups to base your nuclear arsenal around (since they arrive so fast that you either need a ton of nukes to sit it out and have second strike capability, or you need to launch on a hair-trigger). It doesn't protect you against all nuclear warfare, but it's not supposed to - it just pushes you towards delivery mechanisms where an accident or misunderstanding are much less likely to lead to a nuclear exchange.

Although it's a lot easier to stay an effective nuclear threat if your aims are defensive, because all you have to do is make an attack on you unpalatable. Having tactical nuclear weapons in your arsenal makes you more or less immune to invasion by the Big Menacing US Ground Army, unless you give the US a damn good reason, because we're not willing to lightly risk losing numerous billions of dollars and thousands of lives for each shot that gets through our defenses.

You still need to keep the nuclear weapons' locations secure against a preliminary strike, but otherwise yes. Nukes let you scale way back, particularly if you don't have to really get involved in a bunch of operations that require an active troop presence.

[Sea Skimmer]

I thought the point of ABM was that it specifically blunted ICBMs, one of the most destabilizing set-ups to base your nuclear arsenal around (since they arrive so fast that you either need a ton of nukes to sit it out and have second strike capability, or you need to launch on a hair-trigger). It doesn't protect you against all nuclear warfare, but it's not supposed to - it just pushes you towards delivery mechanisms where an accident or misunderstanding are much less likely to lead to a nuclear exchange.

Kind of depends on the details. ABM blunts ballistic missiles, but ICBMs tend to be awful numerous. A large ABM system creates a risk that it creates a situation in which launching a first strike on the enemy ICBM farms can leave the enemy with too few submarine launched missiles to overwhelm your ABM. Oh... and we can knock out ICBM silos now with conventional warheads. That creates some interesting possible situation. Fear of this happening is more or less what is driving the Russians mad right now.

Also by nature of high cost bomber and submarine weapons will only ever be based at a few sites (counting submarine at sea as a site) at any given time which can leave them very vulnerable to a bolt from the blue attack if the submarine technology is not flawless. A shitload of ICBMs in discreet silos simply can't be knocked out by any cheap tricks like a nuclear missiles fired from a few airliners might decimate bomber bases. Only a massive nuclear attack or massive ABM system can negate it. That is in and of itself stabilizing. Not in a very good manner, but its worked so far.

At some point in the future were bound to see hypersonic vehicles using existing ballistic missiles as boosters, and also air launched via various manners. That will then require yet another totally new defense system largely independent of existing SAM and ABM weapons. See HTV-2 for what such things might look like (HTV-2 tests are even using old MX missile boosters!); such concepts aren't at all new for strategic weapons, its just never been necessary to fund them to production before. Of course if we felt like it, hardened shelters for nuclear bombers are totally feasible. Build one at every airport in the country....

[xthetenth]

As far as I understand, the way defense plays into the whole deterrence balance is by making a country's second strike capability (as in how much damage you can do after getting hit) more resilient and therefore enough to deter attack without heading into MAD levels of first strike capability. There's still plenty of potential for damage to work against any vaguely rational actor, but it's in the French model of way too much cost for the benefit rather than the US style of ending the other country as a functional thing. Another main benefit is that things like the rogue colonel scenario are easier to deal with because there's a good chance a few missiles/bombers can be intercepted.

I'm not sure whether a thick defense that can stop a full out enemy attack from doing some damage in a Cold War scenario is possible because with two reasonably close competitors, the defenses to stop a Moscow criterion style attack on a single target are basically going to have to take on the entire arsenal of the enemy and even with optimistic estimates of counterforce strikes, that's going to be a lot more materiel needed for such a defense than an offense to overwhelm it in one place. However a thin screen to protect the country as an entity and to defend against marginal players like North Korea are viable and generally a good idea.

That sound about right?

[Starglider]

We've also simply not seen a bomber yet with better then 1980s technology. All existing designs are from that era and yet some massive improvements have taken place in all forms of aircraft technology.

Such as? Airliner range has increased about 50% for the same fuel/cargo load from the A300 to the B787, so presumably a B-1B equivalent could be made now with 50% more range and relatively affordable F-22 style stealth. That doesn't seem like a game-changer, just a more cost-efficient version of the B-2 (lower maintenance, less need for tankers). Would you expect supercruise as well, and is that really a game changer when mach 3 supercruise was feasible in the late 60s?

[Simon_Jester]

From what I know...

Fast enough supercruise would in some ways STILL be a game-changer, since it's a problem most existing air defense networks are not well designed to handle, especially when combined with high altitude. If that could be combined with stealth it would make the problem even thornier; low RCS means they can't see you until you're close, high speed means they have only a few minutes to engage you before you shoot right past them, and high altitude means any ground based weapons or orbiting fighters will have to spend that time climbing just to get anywhere near you.

That is a much trickier combination to handle than a bomber which is hard to see but only hard to see, and is trivially shot down if you can localize it and lock a weapon on it.

[Sea Skimmer]

Think about an aircraft able to hold three hundred small diameter bombs, each with a 40kt warhead and able to boost-glide 50-100km while striking with 30m accuracy completely via inertial guidance, before you think any further about what it might feature. A bomber weapon system isn't just about the plane. In fact this need not even be a new aircraft necessarily but an application of new technology.

End result is a single plane is now able to end a entire national infrastructure, or a national military, or a huge portion of both completely on its own, and has a serious capability to saturate even the densest air defenses while it does so.

[Starglider]

Think about an aircraft able to hold three hundred small diameter bombs, each with a 40kt warhead and able to boost-glide 50-100km while striking with 30m accuracy completely via inertial guidance, before you think any further about what it might feature.

Does it even make sense to load that many expensive nuclear weapons onto one platform that can be taken out by one missile or serious mechanical failure? I recently saw an estimate of $28M each for the B61-12 and that's using existing fissile (although a small production run). You probably have a better idea of unit cost of nuclear cruise missiles than me, but if correct that's twenty ish munitions for the adjusted unit price of a B1B. Even with operational expenses considered, over a hundred nuclear warheads on one platform does not seem sensible (for a penetrating aircraft that is bound to take some losses).

Fast enough supercruise would in some ways STILL be a game-changer, since it's a problem most existing air defense networks are not well designed to handle, especially when combined with high altitude. If that could be combined with stealth it would make the problem even thornier; low RCS means they can't see you until you're close, high speed means they have only a few minutes to engage you before you shoot right past them, and high altitude means any ground based weapons or orbiting fighters will have to spend that time climbing just to get anywhere near you.

Is mach 3 'fast enough'? Because in the 1960s it wasn't considered 'fast enough' to give reasonable defense against near-future Soviet capability (admittedly the quality of that analysis can and has been challenged), hence cancellation of the B-70. Any faster and you're into hypersonics and vastly expensive technology that is still in the experimental stage. Unless there is some synergy between high-supersonic speeds and modern electronics & guidance (e.g. greatly extended glide range for gravity weapons) that has yet to be exploited in a bomber?