Strategic Test Suite (STS): STS(v7.0) Offer.007

A continuation of the discussion started here:
jomega's Blog • Strategic Test Suite (STS): The EPD file's best and alternate best moves. • lichess.org

Another interesting position for which Stockfish 14 failed to find the best move in the 3 second (depth 24) test. However, Stockfish had the correct idea; just delaying by a move. The position is about whether White should offer a Queen Exchange.

The Human Perspective

After 1.Qb4, if Black exchanges Queens, then Black is left with his weak queenside pawns (a6 and c7), a cramped position because of the White c6 and e5 pawns, and the possibility that White will exchange his Knight for the Bishop where the Bishop stands, which currently would result in Black having an isolated weak e-pawn. And, of course, White is a pawn up on the board. Black has several possibilities after the Queen exchange, including trying to get rid of his weak a-pawn, moving the Bishop (say to c4) so it cannot be exchanged where it is, and beginning the centralization of his King with ...Kf8. Also, ...Nd5 and ...g5 should be considered in the position after the exchange. So Black is not without resources. If Black does not exchange Queens, say plays ...Qe8, then White can play 2.f4 to strengthen his e-pawn, and redeploy the Queen depending on how Black plays.

Stockfish's Perspective

Stockfish actually 'plans' to play Qb4 in several of the variations that it came up with that don't begin with that move; so they are basically the same idea. For example, 1.f4 delays Qb4 by a move, and since, as I show in the study, Black cannot stop Qb4 with ...Nd5, White can afford the delay.

Trying to annotate the variations from this position deeply, would be very messy. The issue is that there are so many moves for both sides that are different ideas, and Stockfish gives many of these a very close evaluation score. And this situation continues throughout the look-ahead. Kotov in Think Like a Grandmaster called such a situation a bushy variation tree. Doing that sort of analysis is outside the scope of this post.

EPD Changes Needed

The EPD should be changed to Qb4=10, f4=10, Ra1=9. I find the moves Qc3, and Bd3 entirely against the spirit of the intent of the position, and they are not superior moves either. I'd take those two out as alternatives.

Links

- The Strategic Test Suite (STS) home page.
https://sites.google.com/site/strategictestsuite/
- The STS-rating code.
https://github.com/fsmosca/STS-Rating