Backgammon Online

Divide and you will unmask!
Strange name, real matters
[ Sign up | Log in | Guest ] (beta)

Have you ever played Backgammon online? It is a lot of fun. So come play Backgammon online with us!

Sign up now to play backgammon online for FREE!
backgammon online

dc_montana 42 ( +1 | -1 )
Chess Survey Hi, everyone! I joined Gameknot recently and enjoy playing here. Good games all around. Anyway, I'm doing some research into game intelligence and would appreciate if you took my short survey on economy in chess. Here's the link:


I'm open to feedback about it. Thanks in advance.
dc_montana 3 ( +1 | -1 )
I've got 7 respondents so far. Only another 43 to go. :)
loreta 2 ( +1 | -1 )
I I simply did not understand it :-(
tim_b 21 ( +1 | -1 )
Nor me :( Are the white pieces more economical in position A of the first pair because they are all instrumental to the checkmate, whereas they are less so in position B because some pieces have no part in the checkmate?
More: Chess
greyrabbit 16 ( +1 | -1 )
Thank you Loreta! You've no idea how much better that makes me feel. I don't intend ANY disrespect, but I thought it was just me being a chess dummy again!

dc_montana 12 ( +1 | -1 )
"Economy here refers to White's overall use of its pieces, their power and their contribution to the checkmate"

tim_b, yes.
dc_montana 5 ( +1 | -1 )
loreta and greyrabbit,

What is it exactly you don't understand?
More: Chess
far1ey 7 ( +1 | -1 )
So white uses all his pieces to the best of the pieces abilities? Correct?
dc_montana 3 ( +1 | -1 )

Yes, pretty much.
doctor_knight 36 ( +1 | -1 )
I think maybe we are trying to hard to come up with a mathematical criteria to take the survey. It's not a test! I think dc_montana is just trying to get a survey on general opinions of chessplayers. Answer how you feel. There are many ways you could look at economy; it seems that the point is to find how most chess players feel about economy.
sf115 9 ( +1 | -1 )
I've taken the survey and I am interested to see what the results are. can you publish the results soon please?
ccmcacollister 68 ( +1 | -1 )
so It sounds like he is talking "Economy" as defined in Chess problems, like far!ey says. Rather than using the least pieces available to you, or having expended the least pieces.
Instead, that each or the most pieces on the board have some contribution to the theme!? EG mate. etc.
Thus while a fools mate would be very effecient in terms of using few pieces, it is very uneconomical in Problemistic terms since only the Queen delivering mate, and the pieces that surround the mated king which trap it onto its square are actually participating in the mate theme .... yah?
dc_montana 55 ( +1 | -1 )

Well said. I have actually developed a computational evaluation function for economy in chess and this survey is intended to determine if there's a positive correlation between the computer's evaluation of the positions and human perception of economy in chess. The evaluation function, if it works, could be used to automatically identify economical checkmates in large chess game (or problem) databases for human appreciation.


I now have 11 respondents. I need 39 more and will post the results as soon as my analysis is complete.


Yes, you are right.
dropcut 77 ( +1 | -1 )
I took the number of direct and indirect pieces involved in the mate and divided that with the total number of pieces (both only on the winning side), then I multiplied that with 10. To reach an even more accurate result I'd have to know the position that arose one move back.

The problem is that you're defining a term that I don't think you've defined well. There wasn't a good enough explanation of economical. The better the explanation the closer the human and the computer generated formula will become. The guess the participant makes is based upon a loose definition. So what's the point? You'll get a value when you compare human vs formula that corresponds with the accuracy of the term you've described vs the formula.

dc_montana 60 ( +1 | -1 )

You've described the exact reason why I chose *not* to give a very precise definition of economy. The survey then would be pointless.

The idea is that chess players should have some innate conception about what economy in chess means. Problem composers even more. It's not a completely alien term to them. Based on their understanding and consensus about economy, I can validate my evaluation function if there turns out to be a positive correlation between the two. How else would I really know if the evaluation works unless I test it against players?
dropcut 114 ( +1 | -1 )
"Economy here refers to White's overall use of its pieces, their power and their contribution to the checkmate."

In a given position, since we have no clue on previous moves, we can only evaluate the specific position. Either a piece contributes to the mate (blocking check, checking, blocking escape squares or backing up blocking pieces) or it doesn't. The pieces not contributing are of no value. It's either 1 or 0 points for a piece, because if you'd remove a contributing piece, there would be no checkmate. And we were asked about the economy of the pieces in the checkmate.

I fail to realize how increasing the blurryness of the term from down to earth to hovering above ground would accomplish anything. You admit that if the term were more precise, then there would be no point in the survey. Maybe I'm missing something, but isn't that like showing four fingers to someone, then walking away untill they almost can't see the hand and then put up three fingers and ask how many fingers you're holding up now? Maybe it's better to just post the formula with an exact definition of "economy" so we all can discuss and possibly adjust it?
dc_montana 91 ( +1 | -1 )

The 'formula' was developed after long and careful study of chess literature. Discussing it here would not accomplish much, not to mention invalidate the survey. There are also some copyright issues. I have already tested it enough in other more 'objective' ways (e.g. comparing tournament game and composition checkmates) to convince *me* at least, that it works. However, surveys such as this constitute supporting evidence.

As a respondent, I wouldn't be too concerned about what *exactly* is meant by economy beyond the list of things I suggested in the survey that *should* be considered. Evaluate the positions based on the understanding of economy that *you* have. If your understanding is 'incorrect' in any way (i.e. based on a poor interpretation of the term), rest assured that there are controls within the survey itself for me to determine that and exclude it from the list of valid respondents.
rallyvincent 74 ( +1 | -1 )
I have to admit... ...that I've problems interpreting "economy" by the factors you described. I would go for "achieving the mate as fast as possible with as less complications as necessary".

E.g. : I gave the 2-rooks back rank mate a 10; but not because it involves the only white pieces besides the King, but because two rooks can drive the King to the mate with a small number of moves. If white would have two Knights on the second rank, I still would have given a 10.

From the ten pairs, I only thought of four positions to be "economical"; all the other positions involved way too many pieces that have to be placed in way too many moves to IMO be called "economical".

So I'm not sure if you can use my answers, but you can sort'em out easily, since I gave my user name.

Rally V.
doctor_knight 96 ( +1 | -1 )
hey you probably shouldn't post your evaluation of the problems until the survey is complete.

Anyway. Stop worrying so much about what he means by economy. I think chessplayers should be smart enough to understand that the whole point of the survey is to compare the intuitive feeling of economy of humans vs a mathematical procedure. Just answer what position you personally think utilizes all of the pieces the best. The point is to compare a formula to how humans think. This isn't an exam that will knock off 100 of your rating points. Just use your intuitive sense (this is what is being compared right?) and answer what you personally feel uses all of the pieces most effectively. A human shouldn't need a formula to decide what he/she thinks is the most economical. This is about what people think and not necessarily about what is correct.

Remember, it's a survey, not an exam. And we are not computers either are we?
ccmcacollister 23 ( +1 | -1 )
already ... I am dissatisfied with my answers; having approached it using problemistic economy, but only considering the pieces of the mating side and not whether those from the mated side contributed to the economy. So doc, if mine seem far-out too, you'll know why :)
dropcut 43 ( +1 | -1 )
. I got very interested in how this would be evaluated, especially when the term "economy" isn't something that is clear cut defined, even by the chess society. So a conclusion, thanks to dc's overall study, that helps defining economy would clearly be something we all gain from. I hope that you'll link us the results some time in the future.

Well, that was my final piece of feedback. Good luck.
dc_montana 15 ( +1 | -1 )
27 respondents and counting. :) Only 23 more to go.

I will post the results here as soon as the survey and my analysis is complete. It should be quite interesting.
greyrabbit 38 ( +1 | -1 )
I think I understand now... Its a matter of judgement rather than analysis. Although I'd perhaps disagree with the economy definition, to me the more economical would be to use the least effort and pieces to achieve the goal.

So whilst I wish you success DC, I think my judgement is sufficiently naive as to essentially be useless to you. I would be interested in seeing your results, especially if an explanation is given....

dc_montana 17 ( +1 | -1 )

It is indeed human judgement of economy that I'm interested in. I mentioned in the survey some of the things that *should* be considered. *How* exactly you go about that is up to you. :)
tim_b 4 ( +1 | -1 )
OK I've done it, so expect some blunders, lol!
greyrabbit 9 ( +1 | -1 )
Me too! But as I mentioned before with my judgement.....tried me best anyhow!

dc_montana 15 ( +1 | -1 )
Thanks tim_b and greyrabbit. That brings the number of respondents up to 32. I need another 18 more. Then I can analyze the results and show them to you all.
dc_montana 12 ( +1 | -1 )
Hmmm... no new respondents in 3 days. I still need 18 more. Anyone who hasn't taken it already, please do. Thanks.
ionadowman 8 ( +1 | -1 )
dc_montana... ... You might need to start a new thread to hook more people in...
dc_montana 12 ( +1 | -1 )
Ion I'm not sure if I should do that. I still need 17 more respondents, though. I'll give it a little more time.
budders 11 ( +1 | -1 )
Just noticed this thread and took the survey. Interesting stuff! Looking forward to the results.
ccmcacollister 14 ( +1 | -1 )
dc_m It would seem to solve things if perhaps one Team would step-up to finish it off ...
How about it Teams!? Anyone game for the Honor!!?
dc_montana 3 ( +1 | -1 )
12 more respondents required. :)
brulla 5 ( +1 | -1 )
I took part... ...too. It was fun. Wonder how you will evaluate and assess the results.
ccmcacollister 14 ( +1 | -1 )
one thing the survey may reveal... chess players become more scarce as the summer approaches! If it proves to be anything like my experience with tournament attendances. :)
dc_montana 108 ( +1 | -1 )
The Results Well, I got only 40 respondents and had to make do. About 5 from those had to be disqualified because they failed the 'control' position (i.e. the first one) which indicates they had a poor understanding of economy to begin with. Here's what I found.

There was full agreement (more than 50% votes) between your evaluations of the positions and the computer's assessment except for pairs 6, 7 and 8. The computer chose the following positions as being more economical. 1:a, 2:b, 3:b, 4:a, 5:b, 6:b, 7:b, 8:a, 9:a, 10:b.

As such, there is a weak but positive correlation between the actual scores of those positions (by the computer) and the average of the scores you all entered (at least those who did).

Overall, I think players tend to have a rather superficial approach to economy in that they might just look at one criterion (e.g. number of pieces on the board) but upon deeper analysis, would probably agree with the computer if they took into account things like the mobility of the pieces and how much of that contributes to the checkmate.

I would like to thank everyone for participating. Feel free to comment.