Metals on asteroids are easier to extract. Some asteroids, like 16 Psyche are almost solid chunks of metal. Asteroid ice (oxygen and hydrogen) can be turned into rocket fuel with electricity from solar panels.
Single asteroid close to the Earth can supply all the needs on earth for centuries.
16 psyche is one of the 10 most massive objects in the belt. The belt is more than twice as far away, and transfers to it are much more variable over the years than the fairly regular Mars windows.
Finding the “Goldilocks” object that supplies all our needs, can be regularly reached economically, etc etc isn’t an easy problem.
Also, water gets you hydralox propellant, which is a bit of a pain in the ass. You really want organic compounds too so you can do methalox, which is one more constraint on Goldilocks.
[0] https://m.phys.org/news/2013-05-curiosity-mars-rover-drills....
I don’t have the expertise to really compare asteroid mining to Mars, but I’m suspicious of the fixed infrastructure costs. Mining and refining takes a lot of mass. Replicating that mass for each object you want to mine or flying it from rock to rock both seem like significant costs. Perhaps we’ll find a big Goldilocks rock that’s near Earth, will stay near Earth for a long time without insanely expensive capture maneuvers, and contains all the materials we want, but I think Mars has a better shot.
Mars also has much friendlier conditions for propellant manufacturing, which is an important factor in these cheap launches.