>comparatively enormous infrastructure investment
Any infrastructure that can remain on the ground and doesn't have to be on the rocket is worth whatever the investment cost.
You note of course that instead of legs, which have mass and have to have a structure in the stage which distributes the loads the SH has those parts with pins which were locked eventually with Mechazilla's arms, and those parts also have mass and need to have corresponding distribution of loads.
How different those consoles with pins are from possible landing legs, and how much savings they provide is an interesting question. It's quite possible they provide some savings - but it would be nice to know some details.
Why so? Pin parts need to withstand similar loads - and if amortizing rails of Mechazilla may soften the contact, the direction of loads for pin parts is less favorable than for legs. Legs don't need to be big or too numerous - effectively legs are those pin parts moved to the engine compartment and turned for an angle.
Compression and tension are quite different loads. There have been rockets in history that would collapse under their own weight unpressurized. Neutron's second stage is a hung tank for similar reasons. Bucking is a pain. Super Heavy can obviously support its own weight, but tension is always going to be the easier load path.
Because the pins can be much shorter. Take a look at the falcon 9 legs. They are enermous both in absolute terms and relative to the whole rocket. They need to be that long to provide a stable platform and enough clearance for the nozzles and the residual plume as the engines shut down.
Don't forget that between certification and catch attempt that catching infrastructure is subject to the launch of the most powerful rocket man has ever created. It seems that the consideration about another part to fail is not valid here as the parts to fail have not disappeared but rather moved to the tower. They could still fail - in fact it seems that there are now many more recovery-critical parts.
That is, unless the falling rocket could abort a tower catch and move to a secondary nearby tower if a failure is detected in time.
I think it is very much valid if the entire context is taken into account.
The tower is used for various stacking and craning operations between launches. There is a better chance to detect any developing anomalies outside the launch context.
Also, being on Earth and flying nowhere, it can be sturdier and heavier than any flight hardware. Much like Roman aqueducts, it can be overbuilt a bit to ensure some extra resilience.
Plus, more towers at the same site, as you say. If one malfunctions, another one can act as a backup. In contrast, every single landing leg is a mission-critical component and cannot be replaced in-flight by another one.
> and move to a secondary nearby tower if a failure is detected in time.
They will have a tower in the Cape. It’s conceivable they could land there depending on return trajectory and save some mass for payload with that maneuver. I am also quite sure they will be a dozen towers in Boca Chica and I wouldn’t be surprised if they build a couple in California for Southward launches.
They do checks of the tower systems before using it, and have abort contingencies in case something goes wrong during final approach. I'm not sure if they intend (or have fuel budget) for last-second aborts to other towers, or if they just ditch in the ocean (remember there are no humans on the booster).
I'm curious how late in the catch sequence they can still abort.
This seems a bit like removing landing gear from aircraft and telling airports to shoulder the added cost of accommodating them. You've simply shifted complexity elsewhere. I understand that people are dazzle-eyed over the science fiction appeal, but IMO this feels like a distraction. The rocket's already reusable and already the largest rocket ever built, this doesn't add any fundamentally new operational capability while also burning a lot of engineering cycles and adding complexity and uncertainty.
What you're missing is the rocket equation. The less weight, the less fuel you need and the larger your payload. We should trust the decisions of these experienced engineers who are deeply familiar with the tradeoffs involved in spaceflight more than our own intuition.
Yup, the rocket equation is truly brutal. Anyone who thinks legs are superior to a tower hasn't played Kerbal Space Program--and remember that stock KSP is easy mode. You don't need anything like the mass ratio that Earth rockets need.
What you're describing with airplanes already happened. Large airplanes used to land on the water, which incurred a mass and aerodynamic penalty for the airplane but was very cheap to operate airfields ("fields"?) for; it only required a flat lake or harbor which was already there. The switch to landing gear allowed airplanes to be more optimized but requires more infrastructure expenditure for large aircraft.
Removing unnecessary systems that have mass is a big part of making reusable rockets work. It's why propulsive landing is superior to landing with wings, for example.
It is not just about shift of complexity from A and B. Anything that stays on Earth permanently can be built without particular regard to its weight, e.g. much stronger, much more resilient, with bigger safety factors etc.
With any flight hardware, you need to make painful tradeoffs between reliability/sturdiness and weight.
If anything out of the ordinary happens, massive steel chopsticks can take a lot more strain than a landing leg which needs to be carried to the edge of space and back.