I'd be curious to see a more complete data chain (including all human actors) threat analysis (or sketch). Franklin's advice "Three can keep a secret, if two are dead" is perhaps true at this level (directed state-level actors trying hard to stop you). I'd think that the larger risk is counterparty compromise, which applies to any transmission mechanism.
Post-compromise (when they know about and are looking to catch/capture you), then dead drops are much riskier, as you have to physically be there.
Tails and Tor Project are willing to stick out their neck and provide recommendations and security for how to use Tor.
Just remember that they also make a ton of silent assumptions about your threat model, and have strong incentives to compromise to get adoption (because the overall system security depends on having many diverse users).
Why I don't usually recommend dead drops
Disclaimer
Summary
Main
What are you smuggling?
Here's some random dark web comment on how to setup a dead drop. onion link to this comment
It covers following steps:
This is hard to execute correctly in practice
I will stick my neck out a bit here and admit that I tried to set up dead drops too, but realised how difficult this would be in practice.
I currently think that for most circumstances, the probability that govts have successfully broken Tor and will use this capability to attack you specifically, is lower than the probability that you will be caught while attempting a hard disk dead drop.
Note that once govts tip their hand and use an attack, everyone else becomes aware that they did this attack. Parallel construction of evidence can only work so many times before the world finds out. Unless you are their highest value target (example: you're a nuclear spy from a foreign govt), it seems unlikely they'll use this capability on you.