If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Policy / Politique
The fee for tournament organizers advertising on ChessTalk is $20/event or $100/yearly unlimited for the year.
Les frais d'inscription des organisateurs de tournoi sur ChessTalk sont de 20 $/événement ou de 100 $/année illimitée.
You can etransfer to Henry Lam at chesstalkforum at gmail dot com
Transfér à Henry Lam à chesstalkforum@gmail.com
Dark Knight / Le Chevalier Noir
General Guidelines
---- Nous avons besoin d'un traduction français!
Some Basics
1. Under Board "Frequently Asked Questions" (FAQs) there are 3 sections dealing with General Forum Usage, User Profile Features, and Reading and Posting Messages. These deal with everything from Avatars to Your Notifications. Most general technical questions are covered there. Here is a link to the FAQs. https://forum.chesstalk.com/help
2. Consider using the SEARCH button if you are looking for information. You may find your question has already been answered in a previous thread.
3. If you've looked for an answer to a question, and not found one, then you should consider asking your question in a new thread. For example, there have already been questions and discussion regarding: how to do chess diagrams (FENs); crosstables that line up properly; and the numerous little “glitches” that every new site will have.
4. Read pinned or sticky threads, like this one, if they look important. This applies especially to newcomers.
5. Read the thread you're posting in before you post. There are a variety of ways to look at a thread. These are covered under “Display Modes”.
6. Thread titles: please provide some details in your thread title. This is useful for a number of reasons. It helps ChessTalk members to quickly skim the threads. It prevents duplication of threads. And so on.
7. Unnecessary thread proliferation (e.g., deliberately creating a new thread that duplicates existing discussion) is discouraged. Look to see if a thread on your topic may have already been started and, if so, consider adding your contribution to the pre-existing thread. However, starting new threads to explore side-issues that are not relevant to the original subject is strongly encouraged. A single thread on the Canadian Open, with hundreds of posts on multiple sub-topics, is no better than a dozen threads on the Open covering only a few topics. Use your good judgment when starting a new thread.
8. If and/or when sub-forums are created, please make sure to create threads in the proper place.
Debate
9. Give an opinion and back it up with a reason. Throwaway comments such as "Game X pwnz because my friend and I think so!" could be considered pointless at best, and inflammatory at worst.
10. Try to give your own opinions, not simply those copied and pasted from reviews or opinions of your friends.
Unacceptable behavior and warnings
11. In registering here at ChessTalk please note that the same or similar rules apply here as applied at the previous Boardhost message board. In particular, the following content is not permitted to appear in any messages:
* Racism
* Hatred
* Harassment
* Adult content
* Obscene material
* Nudity or pornography
* Material that infringes intellectual property or other proprietary rights of any party
* Material the posting of which is tortious or violates a contractual or fiduciary obligation you or we owe to another party
* Piracy, hacking, viruses, worms, or warez
* Spam
* Any illegal content
* unapproved Commercial banner advertisements or revenue-generating links
* Any link to or any images from a site containing any material outlined in these restrictions
* Any material deemed offensive or inappropriate by the Board staff
12. Users are welcome to challenge other points of view and opinions, but should do so respectfully. Personal attacks on others will not be tolerated. Posts and threads with unacceptable content can be closed or deleted altogether. Furthermore, a range of sanctions are possible - from a simple warning to a temporary or even a permanent banning from ChessTalk.
Helping to Moderate
13. 'Report' links (an exclamation mark inside a triangle) can be found in many places throughout the board. These links allow users to alert the board staff to anything which is offensive, objectionable or illegal. Please consider using this feature if the need arises.
Advice for free
14. You should exercise the same caution with Private Messages as you would with any public posting.
15. Have fun!
(Thanks to Nigel Hanrahan for writing these up!)
Reasons to believe there's FIDE rating "inflation"?
Re: Reasons to believe there's FIDE rating "inflation"?
This dichotomy between Jeff Sonas and other statisticians like Macieja is well presented here Impressions from FIDE Rating Conference. Please look especially at figure 6, as well as the statement "Nowadays [2010] there are almost 112,000 players with FIDE Elo ratings compared to 33,384 in year 2000.". I stand firmly in the Macieja camp.
This dichotomy between Jeff Sonas and other statisticians like Macieja is well presented here Impressions from FIDE Rating Conference. Please look especially at figure 6, as well as the statement "Nowadays [2010] there are almost 112,000 players with FIDE Elo ratings compared to 33,384 in year 2000.". I stand firmly in the Macieja camp.
That's perfectly reasonable and I tend to agree with this view. I would be in the Macieja camp as well.
However, I don't think that the current increase in FIDE ratings is due to players being better. Correlation does not imply causation etc.
Nigel Short crossed the 2700 barrier for the first time when he was about 40 and he himself admitted that his play was nowhere near his 1993 level.
Re: Reasons to believe there's FIDE rating "inflation"?
Raw output from my program, first from last year's Gibraltar Open:
ShortGib2012: InvExp with s = 0.0783174, c = 0.538969; Unit weights
------------------------------------------------------------------------
IPR: 2787 from 0.049306, 2-sigma range [2609,2965], 2.8-sigma [2538,3037]
IPR if 291 positions faced were test suite: 2791, st. dev. 89.10
AdjIPR: 2787 via 0.049067/0.049067 = 1.000004: 2609--2965; 2.8s: 2538--3036
Adj. AE/turn: 0.049306 stdev. 0.006112, index 0.000358
ShortWch1993: InvExp with s = 0.098871, c = 0.544656; Unit weights
------------------------------------------------------------------------
IPR: 2623 from 0.060561, 2-sigma range [2477,2770], 2.8-sigma [2418,2828]
IPR if 594 positions faced were test suite: 2614, st. dev. 73.22
AdjIPR: 2623 via 0.061162/0.061161 = 1.000019: 2477--2769; 2.8s: 2418--2828
Adj. AE/turn: 0.060562 stdev. 0.005023, index 0.000206
I don't ask players what they feel about the level of their play. I just crunch the numbers. If their numbers dignify their rating, they deserve the rating. :D
Re: Reasons to believe there's FIDE rating "inflation"?
In order to answer this question, would it be possible to input all of the world champions' games into Houdini, find out which players it likes the most, the least, and in what order, and to what extent, then compare their FIDE ratings to Houdini's appraisals? If it is found that Houdini rates Fischer or Caplablanca higher than Anand, then would this demonstrate that inflation is a fact? Or, the opposite, if it found that Anand was better than Fischer to a greater extent than their respective FIDE ratings indicated, would this suggest that deflation has actually taken place?
In order to answer this question, would it be possible to input all of the world champions' games into Houdini, find out which players it likes the most, the least, and in what order, and to what extent, then compare their FIDE ratings to Houdini's appraisals?
This was done with Crafty to depth 12, and later with Rybka 3 to reported depth 10 (which is said to be "really" depth 13 or 14, as Rybka groups the bottom 4 nodes of its search) by Matej Guid and Ivan Bratko as described here. I have analyzed every world championship match and more with Rybka 3 to reported depth 13 ("really" 16 or 17), with preliminary results reported in my "Compendium" paper. I am in the process of converting to Houdini 3 depth 17 and Stockfish 2.3.1 depth 19, which my results show to be no weaker than Rybka 3 depth 13.
In my opinion the Guid-Bratko work is OK for "likes the most" and "in what order", but is insufficient to put a number on "quality". The extra data-specific features in my work needed to achieve the quality-rating goal and put it on the Elo scale are:
1. Stronger engine: upping depth by 3 ply gives over 200 Elo increase in this range (though it takes about (2.5)^3 = 15 times as long per move).
2. Analyzing all reasonable moves to the same depth, which involves Multi-PV mode in the 20-50 range, not Single-PV mode or Multi-PV under 5. This takes 20-30x as long per game!
3. Accounting for an observed logarithmic scaling law: a difference of (say) a quarter-pawn means a lot more when the game is even as when one side is a Pawn or more ahead.
Anyway, regarding the WC's and Houdini 3, you shall have your wish by summer I expect! Quicker if I can find volunteers with available desktop-CPU cores (I don't want to torture laptops with 24/7 scripted analysis) to help with the data collection :)---Houdini and Stockfish are well enough behaved to avoid the level of manual intervention that made this a headache with Rybka 3.0 and 4.0.
Re: Reasons to believe there's FIDE rating "inflation"?
Fwiw very strong players in the 19th and early 20th century on occasions announced checkmates in well over 10 moves, even in ostensibly complex positions, I noticed in an old book I used to own. Not sure today's best players could always do the same, but finding very deep mates is not normally a vital chess 'skill' to have though afaik. Then again, the old master games I noticed may have been played without a clock.
Anything that can go wrong will go wrong. Murphy's law, by Edward A. Murphy Jr., USAF, Aerospace Engineer
I wasn't referring to your post, and I agree that the rating system can't measure "the quality of one's play". Nor should it, in my opinion.
I definitely agree. However, the existence and availability of cheap chess engines opens a lot of possibilities that ELO had no way of addressing.
One thing would be for the makers of chess programs (and databases) to provide a way of calculating the "objective" strength of play over one or a sample of games. That might give your average player the tools to do some useful research in that area.
Maybe Chessbase 2013 will have that. What do you think? :-)
Raw output from my program, first from last year's Gibraltar Open:
I don't ask players what they feel about the level of their play. I just crunch the numbers. If their numbers dignify their rating, they deserve the rating. :D
Ken, maybe you can get the makers of commercial programs and chess databases to buy your algorithm and put it in their products. If it makes you rich, so much the better.
In the meantime I hope you'll continue with this approach.
there are mechanisms that put points into the system, and players getting better with time is NOT one of these.
agree. Changing player skill level doesn't add or subtract points from the system; at most, it moves them from one player to another as their relative skill levels are reflected in their OTB results.
All the basic concepts that I mention are explained there.
The article does make the same claim you have made that a rating floor causes points to be added to the system, but it does not explain how or why this would happen. So it's no more convincing than your earlier claims that the rating floor adds points.
Here is an explanation of one way it could happen:
In principle, a player at a rating floor could perpetually lose, and while that player's rating would never go down, his opponents would be credited with wins against a player 350-400 rating points lower. In such cases, games lost by a player at the rating floor would not be zero-sum, and there would be more rating points in the system after the game than before. (on average, less than 1 pt per game)
three questions:
1) is the mechanism I've described above a significant source of inflation?
2) is there some other way a rating floor would add points to the system? [added-see below]
3) does FIDE even have a rating floor? As has been previously noted, FIDE used to publish only ratings above 2200, but not publishing a low rating is not the same as not calculating a low rating.
agree. Changing player skill level doesn't add or subtract points from the system; at most, it moves them from one player to another as their relative skill levels are reflected in their OTB results.
Exactly, but it seems that some people around here think that ratings magically go up when stronger players enter the pool.
The article does make the same claim you have made that a rating floor causes points to be added to the system, but it does not explain how or why this would happen. So it's no more convincing than your earlier claims that the rating floor adds points.
As we wrote previously, it's obviously a complex process and multiple things can happen. But the floor makes it so that a player (say John Doe) with a true strength just below 2200 may gain points 'off the radar' (i.e. against sub 2200 competition) and then lose these points 'on the radar', thereby pouring some points in the system.
-John Doe goes on a good streak against sub 2200, gets his rating to 2250 (i.e. above his 'normal' level)
-John gets on the list, loses against 2200-2300 opposition (as expected)
-The 2200-2300 guys now have these 'easy points' in hand and may be expected to lose them as well, most probably against higher rated opponents.
-John goes back 'off the radar' and may get his rating up again.
Basically, every time a John Doe goes on a good streak, it provides for easy points to the 2300ish crowd.
Not saying it's the only way to add points in the system, but it does seem plausible.
But the floor makes it so that a player (say John Doe) with a true strength just below 2200 may gain points 'off the radar' (i.e. against sub 2200 competition) and then lose these points 'on the radar', thereby pouring some points in the system.
-John Doe goes on a good streak against sub 2200, gets his rating to 2250 (i.e. above his 'normal' level)...
I don't see how this could happen.... but maybe I don't understand what you mean by "rating floor".
By "rating floor" I think of a rating below which FIDE will either not allow your rating to drop, or below which it will treat you as unrated.
Assuming, for the sake of argument, that FIDE has a rating floor of 2200, how would a player below 2200 gain points by beating other players rated even further below this floor? If it really is a rating floor then all those games would be unrated by FIDE. It's not that a player could gain points "off the radar", there are no points below the floor to be gained.
Tweet
Comment