The logic goes like this: since both sides know, and they know that the other side also knows, the game theory conclusion for Russia to have any chance of something other than defeat is to massively attack with nuclear weapons the moment enough NATO countries demonstrate they are seriously committed to fighting and enough forces are assembled.
Since it’s understood that the Russian armed forces can’t possibly win an exclusively conventional battle if NATO commits, as the industrial capacity of North America + Europe is so much greater.
Although obviously the losses would be horrendous, with less than a fifth of the population, a 5 or 6 to 1 exchange in favour of Russia would cause vastly asymmetric damage. Thus giving a ‘relative victory’ to Russia, which may be perceived as better than a conventional loss.
This then produces the paradox that the more committed the NATO countries are to fighting, the less safe they become. And the more resources spent on their military, the less safe they become, as the conventional gap grows wider.
Whereas conversely the stronger and more confident Russia becomes conventionally, the lower the chance of them resorting to rapid nuclear escalation, thus making the world safer. (assuming nuclear forces remain constant on both sides)
In economic terms, every extra dollar of wealth generated, and every extra bit of productivity by the NATO countries above and beyond what Russia can produce, actually makes everyone less safe as they decrease the confidence that Russia can resist a conventional loss.
It sounds totally bizarre, is there a better interpretation?