^

that's actually a pretty good ice bucket challenge.

Hou Yifan is doing very well in the Grand Prix. Her most recent game was pretty cool.

http://mobile.nytimes.com/2014/09/01..._r=2&referrer=

Interesting idea, doubt it will take off though.

A friend asked me this question and I'm not sure of the answer.

For Carlsen to have a 1% chance of losing a game (not factoring in draw %, just strictly win %), what would his opponent's FIDE rating be?

I also (highly) doubt that it will take off.

The majority of the competition must be pretty tough, would be pretty sweet to play in though.

Above 1200 with 15|10 time control for the first time ever. I've been noticing tactics in my last several games somehow. I'm not calling this a breakthrough until i can stay above 1200 but it's starting to feel like I've been making progress after seemingly an eternity of not improving one bit. Chess is HARD. Heh.

^How old you are when you start playing/first learn chess has a huge impact on how high your rating can go imo. How old were you when you first learnt chess?

If you want to improve, start an improvement scratching post.

I've known how to move the pieces since I was fairly young, but I never played the game until the last two years. And in terms of actually trying to get better only since March or so. Progress is slow but I'm about 200 points stronger than I was then. I wouldn't say I take the game seriously, it's just a new hobby. I do practice tactics a few times a week and watch a lot of chess on YouTube (mostly this, I love watching others play).


My only goal is to get to like 1400-1500 some day and to have fun playing and following chess. If I get beyond that, great.

Yeah perfectly good goal to reach. you can always improve with practice and study. it's just that you improve alot slower and with more effort than when you're young.

I'm still fairly young at 19. not the optimal age but young enough that my learning should be pretty effective. Only I haven't been playing chess much at all for the past few months, mostly I'm putting all my concentration into poker since that's a game that actually makes me money.

The Dmitry Komarov plugin is great if you play on lichess. (The sound is lagging in this video, but it plays right in time with the moves.)



If you take a while to make a move, it adds color commentary, which can be hilariously irrelevant to the game (e.g., talking about Nakamura's cream or whatever).

It's hard to say because rating differences are based on the expected amount of points, not the full win / draw / loss ratio. Earning 0.99 points per game implies being rated 677 points higher than the opponent as per the FIDE handbook, title req-ts, paragraph 1.49.

The theoretical formula for the expected score (in points per game) is 1 / [1 + 10^((Ra-Rb)/400)], where '^' denotes raising to a power, Ra is the player's rating and Rb is the opponent's rating. The FIDE formula for calculation of performance ratings is based on this one (slightly rounded).

If you wish to know when Magnus will have 1% of losses, you should figure out how often he'll draw to apply the above formula, which is of course a hard task, as he'll draw weaker opponents less often than his average 44% (but how often?), and, to make things worse, it vastly depends on the opponent's style, not only the rating.

He said "not factoring in draw %" to preclude all of that complicated ****, I believe.

Yes, and I explained why his question (as opposed to, say, 'vs whom will Magnus earn 0.85 points per game?') leads to even more complications because the Elo / FIDE formula is too crude (based on points, not separate win and draw counts)

Once again, I believe there's no definitive answer to his question, except that a player rated 700+ points lower is almost guaranteed to score <0.01 points per game and hence automatically win less than 1% of games, but this estimate is too nitty because draws happen. A very aggro 2250 CM might as well win 2% of games with Carlsen (while losing all the rest), while an extremely Karpovian 2600 GM might win less than 1% but draw 40%.

Also, the time control matters a lot (there are way fewer draws in bullet than classical chess ldo, and Magnus is more likely to simply blunder a random mate in bullet).

Speaking of time controls, this has reminded me of Kingscrusher's scalps, esp the defeat of Sam Shankland, who was 2581 OTB at the time (while KC was 2141).

It's not very convincing, as Sam's ICC 5-minute rating was only 139 points higher than KC's, but it illustrates well how easily a GM can blunder the queen (at 5:00) out of nothing even with over 3 minutes on the clock.

An easy opening tactic I missed (and implemented the idea when it was too late and losing ).

[hidethisplease]qxd5+ 1-0[/hidethisplease]