We have done a thread about this, but I am too lazy to look for it.

There are three books by Genna Sosonko which are excellent. I also enjoyed "Linares, Linares!" by ten Geuzendam a lot.

Thanks. This is on the right track. Preferably something longer and published in a reasonably sized magazine or e-magazine.

I think we should have team malkovich games, its basically a 2 vs 2 game where partners alternate moves and each player has to post their thoughts, of course they cant read what their partner posted. I dunno it sounds fun, best part is that the players dont have to have similar skills, just pair the best and the worse player together and let the epicness begin

Interesting idea ... it could also work with players looking at their partner's comments as well, particularly if there is either a veyr large (or perhaps small) gap in playing strength.
Would be happy to be involved in this if others wanted to get a game started.

I'd be happy to team up with hiphop and take on all comers. I play a more solid style so should be fun

What's wrong with being a little crazy?? :P

Nothing. In fact i play best when it gets crazy. But i want the 2nd move to rule out BDG shenanigans

sounds fun to me especially as my current malkovich opponent seems to have gone walkabouts.

if this happens the stronger player will be the ones making the move, plus the idea is that the players dont have a clue of what long term plans his partner has.

The main idea of the game is that you dont know why your partner made the move, you have to figure it out yourself. Openings are going to be really intresting

Anyway how strong are you guys? Im going to need Anand to team up with me.

I'd volunteer to be the weaker half of a team for sure. That sounds awesome.

Am I the only one, when first coming across a notated excellent-move comment (e.g. 27...Bc5!!), to get really excited and perceive the move as being "yelled?"

Local public library hosted its free chess tournament, 15 people showed up (new record). I thought I'd have to work but got off early and made it.

Game 1: Played black in opposite-sides castling Sicilian. Busted through first at the cost of a knight for two pawns, attack never did become decisive, lost.

Game 2: Played black against older gentleman who hung a knight, a queen and three pawns in the first 10 moves.

Game 3: Played white, crushed a Fried Liver win for style points

I've seen god knows how many chess games and I still read it this way.

Thanks a lot. I'm going to be doing this now. I always mentally think of 0-0 as "castles" and 0-0-0 as "triple castles".

Candidates and Naka-Pono are getting the attention now, but there has been a v strong tournament in Lublin with excellent games http://www.szachy.lublin.pl/en/games.html, Shirov even wheeled out the KG in round 4 vs Alekseev and won.
Old school was victorious in the last round of the Bahia Feliz tournament : Vallejo-Granda 0-1 and Nyzhnyk-Mecking 0-1. Vallejo still won the tournament and Nyzhnyk finished second though

Thought about referencing the Vallejo-Granda game on here. The finishing of that game was really really nice. It seemed sound too but incredibly complex and I didn't get around to Houdini checking it.

Someone made an interesting point on the chess.com forums (I'm as shocked as you are) that had be doing some thinkin' and considerin'.

Take your average beginning chess player. They decide they want to learn to play, they come on the internet and get the standard advice: tactics training, opening principles, basic endgames. They follow it for a couple of months and are 1000-1200 strength.

At that point, should they enter their first rated tournament if that's what they are interested in? There's obviously no right or wrong answer for everyone, but it brought up some interesting thoughts.

I was in that position last November, and I entered the Thanksgiving Open in St. Louis, scoring 2.5/6 in the U1400 section for a 1089 performance. I found the experience to be inspiring, spawning the chess-obsessed forum whore you see before you.

A strong player on chess.com argued that a player at that level should wait another few months. Otherwise, they may get stuck with a provisional rating that trails their ability by quite a bit in a few months' time, and it can take awhile to balance that out if they aren't able to play tournaments very regularly.

I have to admit, I do find myself in that situation, but I still think I wouldn't trade that experience last November for the chance to start over as a better player.

If they're interested in it then why not? If they hate the experience then they can always just play online. If they love it then good for them they've found another fun way to play.

Im playing a correspondence game on chess.com, am I allowed to look at opening databases or would that be cheating?

What's the idea, to maximize your rating? I guess if being 1300 instead of 1200 is going to get you a place on an Olympiad team or an invitation to Linares, you could consider "managing" your rating like that. If you're a male over the age of 7, though, you should just play and let the rating sort itself out. You're going to be complaining you're underrated anyway, so you may as well do it when there's some truth to it.

sexist ban.

This is funny, but true, despite the fact that if one is underrated in any significant amount (say, by 150+ points) one can gain a huge number of rating points every tournament one plays. (Assuming that one plays in tournaments with people of similar or mixed strength, at least.)

I suspect it has to do with people believing that their "true" rating is what would come from them playing their absolute best chess every move of every game. Which is silly when a game between two "A" players is probably 30% to see one of them drop a piece to a simple combination. (I would love to see a real statistic along those lines, if anyone knows of someone who has done that kind of research.)

I tried once to do some research along those lines, but I couldn't figure out how to objectively define the criteria. What exactly qualifies as "dropping a piece to a simple combination"? For the most part you know it when you see it, but to avoid bias you need an objective way to define it (and you need to automate the search process if you want a meaningful sample size). I could try to do some analysis if we defined the criteria clear enough. Maybe search for blunders which alter the computer evaluation by 2.00 or more? Searching a random sample of games in which both players are rated between 1950 and 2050?

I'd say a swing of the computer evaluation by 2 or more when the game is currently evaluated between -2.5 and +2.5 would probably be a good start. It might be necessary to refine the criteria a little more based on what kind of games that finds (and misses).

I'd look for players between 1850 and 1950 (Class "A" players) -- I'd guess the "simple tactical error" rate starts to drop off pretty quickly between 2000 and 2100, but it would be really interesting to see that as well. (That is, to do this study of the "simple tactical error rate" by rating band.)

If you start to do this, definitely make a new thread on the subject, as I'd be very interested in following along with the details, and I'd be happy to review some games from the hit/no-hit population to refine the selection criteria to make the study results more robust.

(Would you be coding the test suite to do this sort of analysis, or does something like this exist off-the-shelf? Are there good libraries in C++ that let one interface with UCI engines like this? All of a sudden, this seems like something that might be possible to do, and I am intrigued.)

Good point on limiting it to points where the evaluation is within an appropriate range of 0. I now remember that one of the problems I ran into last time I tried this was all those "blunders" where the evaluation shifts from +25 to +21 that aren't actually blunders at all.

As for how to do it, I hadn't really planned to get that fancy, I was just gonna do it manually (that is run the blunder-check function manually, and record the results afterward.) Now that you mention it, though, you probably could code a UCI interface to run specific types of tests like this relatively easily... this I shall have to consider further*

*That is to say, AFTER I graduate. I have a 50-ish page paper due in 10 days, that is required for graduation, that I haven't really started per se... so no chess studies in the next couple weeks...