There is a way to objectively distinguish moves from the same category though (drawing ones or losing ones). It's similar to Kolmogorov complexity. Let's say mo

There is a way to objectively distinguish moves from the same category though (drawing ones or losing ones). It's similar to Kolmogorov complexity. Let's say moves A and B both draw but the shortest algorithm that draws vs A is much longer then one that draws against B. We can say A is objectively a better move. In practice instead of formal definition we could use a benchmark engine. What's minimum CPU time/RAM requirements for an engine to draw the resulting positions (or convert them to a win in case of losing moves). A move that requires serious hardware to defend against is better than one a 10 years old laptop can hold a draw against.