"Winning a record number of consecutive international tournaments" can serve as a decently well defined metric if we finetune it a bit (manual selection of tournaments and players) for a tiny selection of players. I agree it can put an interesting order in the chosen selection (although it has some downsides too and I would dare to ask why consecutiveness has to be part of the definition).
My point is it does not statistically "correlate" with "player skill" once we extend the selection of players to a larger selection like "all GMs". Then, according to the suggested metric, all players who never won any international tournament would be equally bad which is not true.
In particular such a metric is useless when it comes to detecting cheaters among GMs or among all players. Therefore it can hardly serve as an example of how complex a good estimation of player skill is and therefore how difficult it is to find cheaters with "inhuman" skill patterns.
Regarding the quoted summary:
I can partly see the point. One could even argue the opposite: the analysis would be most suspicious if any stated idea was exactly one of the top engine lines for a large number of ideas. Humans and engines do not think alike.
However, in order to play such a game vs the world's best player in a classical time control it is reasonable to expect a super gm level understanding of positional ideas and core variations. If his post game analysis seemed to display a strong contrast to such expectation it'd be helpful to mention it. But that should be reserved to very strong players (e.g. GMs or 2600+) and cannot easily be judged by amateurs (I assume we agree here).
My point is it does not statistically "correlate" with "player skill" once we extend the selection of players to a larger selection like "all GMs". Then, according to the suggested metric, all players who never won any international tournament would be equally bad which is not true.
In particular such a metric is useless when it comes to detecting cheaters among GMs or among all players. Therefore it can hardly serve as an example of how complex a good estimation of player skill is and therefore how difficult it is to find cheaters with "inhuman" skill patterns.
Regarding the quoted summary:
I can partly see the point. One could even argue the opposite: the analysis would be most suspicious if any stated idea was exactly one of the top engine lines for a large number of ideas. Humans and engines do not think alike.
However, in order to play such a game vs the world's best player in a classical time control it is reasonable to expect a super gm level understanding of positional ideas and core variations. If his post game analysis seemed to display a strong contrast to such expectation it'd be helpful to mention it. But that should be reserved to very strong players (e.g. GMs or 2600+) and cannot easily be judged by amateurs (I assume we agree here).