<chessgames.com> <The only thing that has changed is that the spirit of the request is being written into our plans for another feature, the Sandbox feature. You see, the Sandbox will not only allow people to upload specific games but also allow the upload of specific variations. This in turn can lead to a tree-structure discussion forum, if the person who uploaded the variation is willing to organize it as such. I know it might seem like "so far it's all been nothing but talk", but this is very good talk. We are big proponents of killing multiple birds with one stone, and the variation feature of the Sandbox will be satisfying many different demands simultaneously.>

The team could use a short break. Give us time to prepare & be ready for the next encounter. The discussions we have been having recently are good. Let's get everything that we think should be corrected on the table now.

Quarterly winners for 2008 were: February 1 won by <<< rinus >>> , May 1 won by < firebrandx > , August 1 won by < g.mueller > and November 1 won by < kwid >.

Anyone can do it. No twisted logic required.

Move forums.

I'm not normally stickler for terminology and a terminology complaint is actually not my main issue with the "sacrifice" way of talking; I think it is wrongheaded about chess engines in a more important way. Because it is easier to evaluate material than positional factors, historical chess engines tended to notoriously overweight material in their centipawn evals compared to other factors. And for that reason, it was common where one disagreed with an engine to suppose the engine might be overweighting material compared to positional factors in the position. But modern engines like Rybka3 use methodologies like regression to optimize their material and positional weightings over millions of positions, and their positional heuristics are much better than they used to be. So the assumption that the disagreement about the position probably stems from an overweighting of material factors no longer has much if any validity. In summary, the "sacrifice" way of talking is misleading, even taken as an analogy.

"Offer" centipawns is better? I simply mean to choose a RV-move scoring 10 centipawns worse than another, which I've noticed the team almost never does. IMO out of too high respect for high-ply infinite analysis.

Move forums don't work as well as the AT in this respect because people would have to first figure out which forum to go to, and then they'd have to read through posts just to see if a line has already been analysed. I don't think I'd need to explain why this method is much less efficient than the AT.

:(

My observation is that there is often disagreement about what constitutes a compelling argument for an alternative. In particular, a significant minority of the team feels that a compelling argument can be advanced of the form "I tried a bunch of continuations to high scoring Move1 and they all led to draws therefore I advocate the alternative lower scoring Move2 which I haven't yet played around with as much, but is very unlikely to lead to worse than a draw." The majority of the team doesn't consider that form argument to be truly compelling, but some team members apparently do. So that disagreement often underlies some heated debates in game after game. This happened many times in our two draws, but it also happend many times in the games we later won due to opponent errors in even positions. IMO, it's worth taking into consideration whether we would have been given those opportunities if we had made "desperation" moves prior to their occurrence.

Best would be all moves to 100%, which is of course impossible.

Then what move would we play?

Also, when the moves are close in score but lead to different sort of games, it may be possible to just argue thematically about why one or the other sort of game is better.

I don't think Anand or Kramnik would do better against us than an expert CC player like Umansky, if anything I think they'd do worse. Anand and Kramnik aren't used to the slower time controls, and they rely more on psychology rather than the truth of the position.

The strategic ideas behind proposed moves should be expressed by chess terms. (Although I don't believe that will happen.)

<isemeria> Ply is term weak players like me can communicate and contribute with. Because my engine is as good as your engine.

Anand and Kramnik have a extremely deep understanding of chess, and if allowed to use engines during a match? voila:).

:0)

There should be more discussion about strategic ideas, and those are especially important in the opening phase and in the search for additional good moves (by the team and by the opponent) later in the game. But it is one of the key strengths of the world team that it compares the merits of different lines objectively. Talking about the comparative scores given by an engine at equal ply depth to a pair of positions is one good objective way to make the comparison, as it is free from the emotions, biases, politics of the particular game and, to the extent that the engine is strong, fairly accurate.

I think it would be hard to direct the entire teams time by percentages even if you had a formula in mind. But as an exercise one could review past World Team games and try to assess the best method for finding and demonstrating actual improvements. Ideally the method should be systematic and not based on the actual moves played later in those games. At the same time one could try to figure out which methods would lead to worse moves than what the team played.

An absolute solution proven beyond any shadow of a doubt is like a dream come true. How about looking for inferior moves at deeper depth or hope for a "LUCKY FIND" from one of our members?