Slight underperformance in the engines resulted in the the booster not reaching where it was supposed to be and the GNC systems automatically aborted to the gulf for safety.

"Slightly low on fuel."

If I understand correctly, the margin was tight enough not to take the risk of damaging the tower.

Those numbers don't make sense.

It doesn't take the same amount of fuel to land as it does to lift.

Speaking as a Kerbal Space program playing, internet commentor, It's probably <1% of the fuel to briefly counteract terminal velocity as it is to lift off and reach space.

They use the atmosphere to brake, not fuel.

"I heard one person say that carrying fuel to land over the whole flight means it takes 23x more launches to get an equivalent payload to orbit."

I don't think that this is the case when using the atmosphere to shed most of your velocity, which is what we do when landing on Earth. Propulsive braking would be another story, but you can avoid that on bodies that do have some atmosphere. In our proximity, only the Moon and Mercury would need propulsive braking, but Moon's gravity is low and Mercury is hard to visit anyway.

That said, once you use the atmosphere to cushion what is basically a fall, the weight of the necessary heat shield comes to the equation too.

How is it twice? Braking on the way down doesn't take nearly as much fuel as lift off so the "tyranny of the rocket equation" should be paid only once.

The booster doesn't go into orbit, starship ends up going 5x as fast. You need a booster to reach orbit though, so it makes sense to recycle the booster too. Why not do it? It doesn't use that much fuel coming back. They can also eventually fly it again quickly in theory (this is easier for the booster due to less stress).

You can be sure they'll have done the math on reusable boosters and it makes sense, or they wouldn't bother.