Polugaevsky Variation Miniature. (B96)

Why is 14. Rd1 a book move instead of 14. 0-0-0 to protect the king and put the
rook on the same square?

I was wondering that myself. All I could come up with was that Qg5+ could potentially be a future concern, and maybe white can't spare the tempo for Kb1. But that seems a small price to pay for the king protection in most cases.

It's a good question and I'm not sure what the answer is but let me speculate.

1. A machine evaluation of the position after both Rd1 and 0-0-0 comes out about the same.

2. The evidence of various databases shows a huge difference in the scores achieved by each move: around 60% for a White win after Rd1 and only 25% for a White win after 0-0-0 (a much smaller sample though).

3. The above two notes leads me to believe that there's some strategic difference in the two positions, that is outside of the machine's evaluation or maybe "over the horizon". Maybe the White King is in fact safer on the e1 square, it looks pretty good there. Black will have difficulties getting at him with the pawn on e6 half-closing the file in a "good way" whilst placing the King on c1 would put it on a half open file in a "bad way". The Black Queen-side Rook is bound to move to a7 followed by a travel along the second rank (maybe c7), coupled with the pawn on b5 this might produce some extra chances for Black that don't exist with the King on e1.

I realise that may just be a load of flannel but I'm struggling. Perhaps, Reb will return and give us a "Master's eye view".

Ivanchuk recently stated that the Polugaevsky has been refuted by engines.