The USSR didn´t die from fright, and certainly not because of the Star Wars system. They died from that f´ed-up communist economy of theirs, if you recall.
That was chronic. Panicked military spending was acute. You can call the cause of death anyway you like, but when I see a guy flat on the sidewalk beneath a open 10th-storey window, my initial diagnosis is not heart disease.
If you think that you´re safe from the enemy, yes it does. Remember that the safety of Mutually Assured Destruction lies in the "Mutual" part. If you think you´re protected, you´re going to take more risk - it´s an elemental and easily proven part of human psychology.
I really don't see that as the fact you seem to, and you haven't addressed the counterexample of overwhelming retaliation, but there we go.
The point also stands that even one teeny-tiny 100-kt nuke going off in the middle of New York, or any other US metropolis would, at the very least, kill hundred of thousands or a million people, cause property damage in the hundreds of billions and economic damage that defies calculation - think of it as 9/11 times 100 at least.
No, the point is, if you will recall stating it, is:
Besides, if LA or San Francisco (or Washington DC, for that matter) get hit by "only" one missile out of three, that´s not that much less of a disaster.
I've shown how that's not really true, and how reducing the number of hits by 65% (to keep with the hypothetical, optimistic though it may be) is, indeed, MUCH less of a disaster.
The overall point is, someone like Kim can´t launch an attack devastating enough to destroy or permanently cripple the US (as the USSR could and Russia probably still can); they can, however, hurt the US far worse (by orders of magnitude) than anyone before ever did, and under real (non-test*) conditions any Missile Defense the US has cannot reliably reduce a potential attack to the point where that´s no longer true.
I am not contesting that the risk of failure is miserably high. However, you have failed to explain why you are happier under a scenario where the risk of failure is miserably 100%.
Again, why is no umbrella better than a leaky umbrella to you? Because we might be more tempted to go out in the rain? News flash, buddy, you can't always control when and where the rain is.
*Intercept rates under real conditions will always be lower than under test conditions, because in reality there will be many more unknowns, and more adverse conditions, than in tests.
True. And irrelevant. You're arguing a strawman because I told you up front that 65% success rate was a hypothetical. Your 0% success counterproposal, however, is completely precise.