
Here is the critical first game of the Candidates, Vancouver 1971:
Nobody – not Byrne, and not Taimanov – thought that 27.h3 was a bad move, until in 2004 Kasparov suggested 27.Bb5. And then Lakdawala in 2015 suggested 27.Ba6! Aagaard improved Lakdawala's variation and then Karolyi and Timman thought White was winning.
But is this really correct? What do the readers think?