CrowdChess, which we covered last year, has initiated a match against chess Grand Master Gawain Jones, one of the top ranked players in England. CrowdChess members have twelve hours per turn to debate and vote on each potential move, at which point the winning move will be executed.
The game will be worth watching to see how the “wisdom of the crowds” fares against the GM, but I fear that the voting system used is too democratic. In the current setup, each vote is weighted equally, independent of experience or rating. The concept of having a “human super computer” capable of beating any challenger is compelling, but we’ll never see anything like that if the voting pool is watered down by people who are clueless.
Crowdsourced chess matches against well-known figures are nothing new. In 1999, MSN Gaming Zone held “Kasparov versus The World”, a four month long chess match that Kasparov called “the greatest game in the history of chess.” The winner? Kasparov, after 62 moves.
I don’t see how this can be that useful. Chess is not a game of this move right now, its a game of planning your strategy moves in advance. So how exactly would they win, if the moves are decided by amateurs who are looking at the current picture?
Clueless, or malicious.
@Andrew
Members can make their case for a certain strategy on the site’s forums, so it’s possible to coordinate moves.
@ Andrew, my thoughts exactly. Even if all the players were experienced, how would that help them beat 1 person who is better than any 1 of them? All their strategies would be inferior to his. Of course, I’ll stop talking if they actually beat him…
I’d be curious what (if any) collaboration and or discussion is allowed for the chesscrowd mob. For it to be useful, these needs to be a lockout to prevent the opponent seeing any of the thinking behind the move.
The crowd is obviously going to get toasted! Chess masters are good because of their ability to think ahead strategically. I don’t see any facility that allows you to easily plot out the multiple trajectories that your proposed move could result in and play them back for other people to easily evaluate and critique. The software needs to support the human creation of branches that a particular move could result in, which demonstrate why that is a good move. They also need to get rid of that stupid “Yes” button that encourages every idiot on planet earth to vote for a move without thinking.
I think this would be improved by allowing people to think in tandem with their computers, using distributed chess software such as ChessBrain (there appear to be others as well). I see no reason to not have software automatically suggest a slew of outcomes and allow humans to choose between them. If the grandmaster was also allowed to use such software, what you’ve essentially done is raised the bar as high as you possibly can, and that’s where it gets interesting! Humans+computer vs. Human+computer.
As other posters mentioned, the grandmaster should obviously not be allowed to see the groupthink, as that will make it even easier to subvert it.
@Martin English
They can collaborate on the site.
My problem is that the people who spend their time collaborating will have the same pull as the guy who logs on drunk, votes for a move, and passes out. That’s what I like to do when I drink, anyway.
I would agree with Brian. I’ve watched Grandmasters play and their focus and ability to think multiple moves ahead with such confidence is what makes them formidable.
With only high school chess team experience, I feel I’d have little to bring to the group.
Stan from CrowdChess here. While it’s certainly a compelling idea to weight votes based on experience and rating it also takes away from other, newer players who might be even better than the “experienced” ones thus putting them at an unfair disadvantage at least from the start.
However top players do have certain privileges, for instance if two moves or more have the same number of votes supporting them than a move will be made based on which move has the most number of top players supporting it (top players are players with the highest percentage of move proposal to move voted on and made ratio).
But never say never, we will definitely keep this idea in mind!!
If you decide to vote for a move for this game on CrowdChess, I suggest the move posted by ‘alterego’. After a lengthy analysis I have determined that his move has been extremely well thought out.
Also just to set one thing clear, the GM is not allowed to enter in to his opposition’s play room per our agreement, so don’t worry, he will not spy!!
If you have at least some belief in the wisdom of crowds, this game probably won’t be that easy for the GM, even if he is better than the crowd. The idea is that each voter has some accurate information and some inaccurate information (most would probably have more bad than good), but all of the errors effect the resulting answer in different ways and tend to cancel each other out more consistently, the larger the group. Check out “The Wisdom of Crowds” by James Surowiecki.
The biggest problem may be that during the discussions many of the voters may follow the advice of a better player and ignore his own reasoning. This may distort the population and cancel out some of the positive benefits of having the crowd decide. Suroweicki calls this the “cascade” effect.
Awesome experiment, either way.
Wow, sounds like fun. I’m curious how all those votes will affect the endgame.
Back in the days of the USSR, World Champ Karpov played the readers of some Soviet commie youth magazine and it took him 50 moves or so to beat them. He said they played good chess.
So I think this will work.
This site is just a gimmick. There is chess software you can install on your PC that will easily beat Gawain Jones or any other GM.
Computers are better than humans at chess, so it is pointless anyway.
And for any chess player it is obvious that the crowd is going to suck.
curious experiment, although I’d be surprised if the wisdom of the crowd prevails when it comes to a strategic, multi-faceted think 4+ moves out game like chess. The crowd might successfully avoid poor moves, but I’m skeptical if they can conjure up a coherent strategy.
I’ve gotten hooked on the game with the daily puzzles and correspondence games on chess.com. Nice to see though that there are multiple sites to play.
And i love the tension that the “wisdom of the techcrunch comments section” (since the consensus here is that the GM will win) is at odds with the wisdom of the crowdchess. Only one crowd will be right/win.
anyone know how to view the previous moves?
A piece down without compensation, issues with development, all within nine moves against a flexible Black structure that already has a built in kingside pawn storm, I’m surprised “Resign” isn’t an option already.
Ha, greate line: “…if the voting pool is watered down by people who are clueless”.
Sad thing is, it applies to electing governments too, where the voting pool is soooo watered down by clueless people… and to which the politicians have to cater.
More votes to people who can demonstrate knowledge, anyone?
I suspect most watching will be pulling out their old Microsoft chess games and duplicating the moves to see what is the best option. Then it becomes a game of persuasion to see whose opinion is heard by the masses, looking forward to it!!
The ex-band Phish did something similar during their 1995 tour. They had a large chess board that would go from behind the stage to the Greenpeace table during the show. Phish would make their move at the beginning of the show and the phans would make their move during the set break.
They worked in 2 games before the end of the tour with the band winning the first game and the audience winning by concession during the new years eve show in NY.