Neither. Both the Clarke scale and the Elo scale were essentially arbitrary with one of the objectives being to avoid negative grades or ratings. Back in the day, they would compare notes. On the usual conversions, Clarke's zero was higher than Elo's.Chris Goodall wrote: ↑Thu Oct 24, 2019 12:13 pm

Is a grade of 0 supposed to represent the weakest person capable of playing a game under FIDE Laws, or a random move machine?

The underlying mathematical statistics of both systems makes an assumption that chess strength is stable if unknown. Arguably an assumption that breaks if you try to apply it to extremely weak players.