The mine shaft things are not obvious losers, but suffer by inability to re-use the expensive part for multiple mineshafts. There are undersea methods that do better, sharing the expensive, onshore equipment among as many simple undersea units as you like.
The scheme described in TFA has the advantage that the tankage needed for an unlimited amount of storage is cheap, with only the total wattage rate in and out limited by initial investment. Tankage underground (e.g. in salt domes) could store the CO2 and the heat in the same place, without losses; earth is excellent insulator. They specifically say in TFA they are storing the heat of compression by pumping it into iron.
The Chilean project storing energy in liquified nitrogen is similar. They also say they are banking heat, even though boiling the nitrogen with ambient air on the way out, in a "warming tower", seems to me more practical.
75% round-trip efficiency is absolutely fine. Pumped hydro is not better, that way.
There is an unfortunate habit in the energy sector of promoting ideas in absurdly expensive form, just because that makes it look more "hi-tech", to be taken more seriously by investors attuned to look for that. The form of each idea actually built and used by utilities will be whichever form is cheapest, which will seem too boring for the press to pay it any attention.
