Trust but verify

This was my third year of playing correspondence chess for Surrey in the county championship. My first attempt pitted me against an engine abstainer and ended spectacularly, but last year's games were tedious draws and not worth blogging. This time I was up to board 5 for Surrey's first team, and it was all a lot more exciting.

I'd never heard of my opponent, Ralph Mansson, in OTB circles, but he is related to a 189 and appeared to have a 100% score before this mini-match, so I was expecting a test. Since we followed modern repertoire books for a long time in both games, he seems to know his published theory - but that's not always enough, especially when you're playing someone who shares an office with a chess bookshop!

My game with black was quite an easy draw, since I just followed some analysis I made last summer. The pawn sacrifice worked well, and White was unable to consolidate his position:

But the game as white was much more satisfying, not to mention complicated. My opponent faithfully followed a new repertoire book right through to the end (on move 29!) and was rewarded with a lost position:

Annotating correspondence games is a challenging thing - you need to transform the mass of unintelligible variations you worked on at the time into clear, instructive prose. That's not easy when the games are so tactical in nature, but hopefully I managed to show that even this sort of chess has a strong logical basis.

I'm more confident that I showed the perils of trusting published analysis. Some of it is very good (despite the knock above, readers will know how highly I rate Negi's ongoing saga) but nobody is perfect and mistakes will creep in. I haven't read enough of the Solozhenkin book to know if it was just a one-off error, but the shady assessment, and the ease with which it's refuted, make me very suspicious. Either way, his book loses by force! My opponent should have checked this before going into it - the decisive error is only two moves from the end of the published line, giving him plenty of time to sidestep it. The rest was pretty, but in no way difficult for a good engine.

Anyway, the event will carry on for a while yet, but Surrey are on 6½/11 at the moment and can hope to finish in a high position. I'm still happy just playing these two games a year, so now it's back to over-the-board chess for a while. I'm writing this post from Bristol, where I am playing a rapidplay on Sunday; before then, the grading list is at its end and that can only mean a spreadsheet splurge.