It's also not something that you can just try out. I'm sure it will "work" in some sense at small scales - you can probably buy a few worn batteries off a used car and connect them to the grid and perhaps make a profit.
But the problem is that this can't be relied on by a whole country as a solution to ensuring power generation. This type of storage is by definition "best-effort" : you don't know at what rate you'll be able to increase storage capacity (how many car batteries will be retired and with how much remaining capacity?), you don't know how much capacity you actually have (what is the actual remaining capacity of these batteries, and what is the risk each will malfunction in some way in the next charge/discharge cycle?).
Not to mention, the risk of fires while amassing worn-out lithium batteries in close proximity to each other is gigantic, and there is no simple way of putting out a lithium battery fire. It could even be that this alone is enough to make such an operation unprofitable when accounting for insurance costs, even if it were allowed to run and connect to the grid.
I'm not going to spend days or weeks researching and coming up with a decent statistical model for something I don't even believe makes sense on intre the face of it for a comment thread.
I'd also note that something being infeasible from first principles explicitly implies that there's no need for calculations or experiments. For example, if someone comes up with a machine whose claimed performance makes it a perpetuum mobile, I don't need to calculate anything: I know from first principles (conservation of energy) that they are wrong or lying. Of course, this is nowhere near as cut and dry, but I did give my reasoning for why this doesn't seem to me to make any sense.
