Ongoing defeats in Ukraine have reopened discussion of a potential Russian nuclear response. Vladimir Putin surprised everyone by restarting his invasion of Ukraine in the first place, so it would be dangerous to put anything past him. Yet there are a number of reasons to expect that Putin will not “go nuclear.”

The video at this link is a pretty good summary of the arguments among the learned. Allow me to offer another: these are old weapons that have not been tested for decades in a country where everything military has been hollowed out by neglect, corruption, and the corrosive authoritarian culture of ass-covering.

We have seen just how bad Russia is at conventional war. So how bad are Russian nuclear forces?

Consider just one potential target, the Azovstal factory tunnel complex in Mariupol. Readers will by now be familiar with this fortress, a deep redoubt tying down Russian troops. Despite constant pounding from the heaviest artillery and missiles in the Russian Army, the Azov Battalion remains defiant and has fought off multiple attempts to storm their defenses.

Whereas that might seem a perfect target for a nuclear strike, the tunnel complex is in fact a Cold War nuclear bunker designed to resist a direct hit.

Nevertheless, let us set aside the caveats and imagine that a frustrated Russian Army uses a tactical nuclear weapon against the Azovstal fortress tomorrow. Putin would have to accept international consequences in advance, not least of all NATO intervention. This is only the “sticker price” of a nuclear strike, though.

Bear in mind that Russia cannot make a nuclear move without drawing the immediate attention of US strategic intelligence systems. Given the level of intelligence-sharing with Ukraine in this war, we would then expect the Air Force and air defenses to do something about it. They have already intercepted an impressive number of Russian missiles so far in this war.

Given all the above, here are five possible outcomes for Putin to consider:

• The bomb might explode and end resistance

• The bomb might explode, but not end resistance (oops)

• The bomb might miss the target (embarrassing)

• The bomb might get intercepted (very embarrassing)

• The bomb might be a dud (dead of embarrassment)

Just one of these outcomes has anything in it that Putin might want. Four of the five are humiliating — and none of them is far-fetched.

In fact, the nuclear dud has an entire chapter in the technological history of the Cold War. Warhead reliability was such a real concern during the 1960s that it prompted the development of long-range nuclear missiles capable of throwing a dozen warheads or more at one time, shotgun-style, to ensure the destruction of a hard target such as a nuclear silo.

The Soviets never matched the accuracy of American rockets, so they had even more reason to adopt this approach. Criticized as “overkill,” these MIRVs (multiple independent reentry vehicles) helped raise Soviet and American nuclear warhead stockpiles into the tens of thousands.

Planners made a rough calculation that one in three warheads might malfunction or get intercepted on its way to the target. A rational risk assessment in the Kremlin might conclude the chances of some sort of failure are at least that high, and therefore unacceptable, now.

If the only way to ensure success is to use more than one warhead, increasing the risks and costs commensurately, a tactical nuke becomes a much less attractive option.

Events are also overtaking our scenario. The Russian Army reportedly balked at giving the Azov Battalion safe passage from the fortress in negotiations this week, then resumed trying (and failing) to enter the fortress or seal it off. Today, they are again negotiating the evacuation of wounded fighters via a Turkish ship. Perhaps a few thousand dead Russians from now, Putin’s army will reassess what a “win” looks like, and just let them go.

While I am not a nuclear strategist or a “game theory” mathematician, at this point, the probability of some other outcome at Azofstal seems much more likely to me than a Russian nuclear strike.

Now let us look afield for an alternative target. It is difficult to find any place on the battlefield where a tactical nuke would have desirable effects. Ukraine has mostly dispersed forces to prevent their destruction.

Like Mariupol, the site of the Azofstal factory complex, most of the places where Ukraine has significant force in the eastern half of the country are also places that Russia wants to have. Putin is hardly going to make his new real estate radioactive.

That leaves the western half of Ukraine, where the problems noted above are magnified. Ukraine enjoys air superiority here, raising the chances of a shootdown.

Inaccuracy becomes more problematic with distance. The yield of a tactical nuclear payload varies, but is usually smaller than the 1946 Test Baker (see photo). So if the Russian objective is actual destruction of a factory, bridge, marshalling yard, airstrip, etc, then it actually does matter whether the bomb hits close to the target, absurd as that sounds. Many of these targets are just as vulnerable to regular long-range munitions.

Even a nuclear dud would be worse in western Ukraine, for it would be in enemy hands, whereas a dud in Mariupol might at least be recovered by Russians.

Again, to use even a small nuke would risk direct western intervention. Putin cannot realistically imagine defeating NATO in a conventional war at this point. Nuclear weapons are supposed to be the final backstop for Russian national security. Expose even this institution as hollow, and Russia’s enemies will lose all fear of Moscow. Putin dares not take that risk.