That said, I wonder why you don't see Ukraine and Russia doing this more -- "saving up" for massive clouds of long range strike drones every couple weeks, instead of sending out a couple hundred every night. It feels like the latter strategy would be more effective, saturating air defenses and what have you, but it doesn't seem to be used much. Maybe launching that many drones at roughly the same time is really hard?
Otherwise, what stopped them from saving up all the bullets, artillery, or bombs and sending them out in brief pulses in prior wars...
Naval anti-ship drones have been around for many decades. This is a highly evolved area of military technology with a long history of real-world engagements upon which to base design choices.
The standard naval anti-ship drones are Harpoon, Exocet, and similar. These are qualitatively more capable than a Shahed and you still need a swarm of them to get through.
The model you are talking about was basically how things worked in the 1970s. Technology has improved a lot over the last half century.
how much ammunition did the US navy use to shoot down incoming drones, and what are the cost of those vs the attacker's cost?
Yes, aircraft carriers aren't nearly as unstoppable as they were in WWII, but they are still the most versatile mobile platforms the world has for projecting force around the globe.
$13 billion dollar military toybox?
Let’s think.
EMP.
Nets.
Defensive Drones.
Superdome.
Finding the solution isn’t hard - choosing and implementing it takes time when you’re a stumbling behemoth.
Finding a solution isn't hard until your adversary adjusts their tacts slightly and bypasses it a week later
So, better be Agile, and have segmented groups doing really different things in different regions,
not taking 10-25 years to develop new overpriced platforms
while World Wars are being fought on DJI.
Melty-laser systems look cheap, compared to losing that even once.
See also: NFL stadiums in the US