| page 1 of 2; games 1-25 of 28
|1. Kramnik vs Naiditsch
||0-1||57||2015||Dortmund Sparkassen||D41 Queen's Gambit Declined, Semi-Tarrasch|
|2. Nisipeanu vs W So
||1-0||28||2015||Dortmund Sparkassen||B23 Sicilian, Closed|
|3. I Nepomniachtchi vs Caruana
||½-½||46||2015||Dortmund Sparkassen||A07 King's Indian Attack|
|4. G Meier vs Yifan Hou
||½-½||59||2015||Dortmund Sparkassen||E06 Catalan, Closed, 5.Nf3|
|5. Caruana vs W So
||0-1||69||2015||Dortmund Sparkassen||B90 Sicilian, Najdorf|
|6. Yifan Hou vs Kramnik
||0-1||28||2015||Dortmund Sparkassen||C65 Ruy Lopez, Berlin Defense|
|7. I Nepomniachtchi vs G Meier
||½-½||25||2015||Dortmund Sparkassen||C11 French|
|8. Naiditsch vs Nisipeanu
||0-1||49||2015||Dortmund Sparkassen||B12 Caro-Kann Defense|
|9. G Meier vs Caruana
||0-1||51||2015||Dortmund Sparkassen||A04 Reti Opening|
|10. W So vs Naiditsch
||0-1||36||2015||Dortmund Sparkassen||D37 Queen's Gambit Declined|
|11. Kramnik vs I Nepomniachtchi
||1-0||55||2015||Dortmund Sparkassen||A04 Reti Opening|
|12. Nisipeanu vs Yifan Hou
||½-½||33||2015||Dortmund Sparkassen||E16 Queen's Indian|
|13. Caruana vs Naiditsch
||1-0||41||2015||Dortmund Sparkassen||E04 Catalan, Open, 5.Nf3|
|14. G Meier vs Kramnik
||0-1||54||2015||Dortmund Sparkassen||C67 Ruy Lopez|
|15. Yifan Hou vs W So
||½-½||40||2015||Dortmund Sparkassen||B18 Caro-Kann, Classical|
|16. I Nepomniachtchi vs Nisipeanu
||½-½||75||2015||Dortmund Sparkassen||B12 Caro-Kann Defense|
|17. W So vs I Nepomniachtchi
||1-0||49||2015||Dortmund Sparkassen||D70 Neo-Grunfeld Defense|
|18. Kramnik vs Caruana
||0-1||38||2015||Dortmund Sparkassen||D78 Neo-Grunfeld, 6.O-O c6|
|19. Naiditsch vs Yifan Hou
|| ||½-½||64||2015||Dortmund Sparkassen||E48 Nimzo-Indian, 4.e3 O-O 5.Bd3 d5|
|20. Nisipeanu vs G Meier
||½-½||42||2015||Dortmund Sparkassen||E42 Nimzo-Indian, 4.e3 c5, 5.Ne2 (Rubinstein)|
|21. I Nepomniachtchi vs Naiditsch
||1-0||68||2015||Dortmund Sparkassen||D36 Queen's Gambit Declined, Exchange, Positional line, 6.Qc2|
|22. G Meier vs W So
||½-½||43||2015||Dortmund Sparkassen||D12 Queen's Gambit Declined Slav|
|23. Kramnik vs Nisipeanu
||½-½||83||2015||Dortmund Sparkassen||A13 English|
|24. Caruana vs Yifan Hou
||1-0||39||2015||Dortmund Sparkassen||D31 Queen's Gambit Declined|
|25. Nisipeanu vs Caruana
||0-1||30||2015||Dortmund Sparkassen||C52 Evans Gambit|
| page 1 of 2; games 1-25 of 28
TIP: You can make the above ads go away by registering a free account!
< Earlier Kibitzing · PAGE 22 OF 22 ·
|Jul-07-15|| ||choosea: > is 0.5 + x/800. ???
>> "Assume that ... "????
>> "The chess Elo rating system also uses K-factor or K-coefficient. The possible values of K-coefficient in FIDE are 10, 20, and 40. The second most commonly used K-factor value combination is 16, 24, and 32 in some other systems..."
>> "..old ones 10, 15, and 30 were in force from 1 November 2011 until 1 July 2014."
>> "Note. Before 1 November 2011 the values of K-factor were 10, 15, 25."
> "..possible values of K..." ?????????????????????????????????????????????
>> "...player’s expected result for the game (from 0.92 to 0.08 in the FIDE system, though many National Federations run it from 1.00 to 0.00)"
> "because FIDE has been using the rule of 400 points in rating difference since 1 July 2009.
The rule of 400 points goes, "A difference in rating of more than 400 points shall be counted for rating purposes as though it were a difference of 400 points". "
>>"...as though it wer..." ???????????????????????????????????????????????????
>> "If the rating difference is 0, each player has the winning probability 0.50. If it is 100, the stronger player has the winning probability 0.64 while the weaker 0.36."
> "With a 100-point difference, it's 5/8, or around 60% only." ????????????????????????????
> " As for FIDE, it is a different picture now... " ????????????????????????????????????????
|Jul-07-15|| ||ChemMac: <choosea> Yes; I know all that **nonsense**. For CALCULATING ratings, sure; have fun with the numbers, for those who like to. All I was talking about is the simplest way of predicting likely tournament scores, without needing a calculator or computer. |
If I want to enjoy complexity, it's better to look at scientific puzzles.
|Jul-07-15|| ||choosea: >> "Calculating Performance Rating
Performance Rating = Opponents' Average + Performance Change
Performance Change is based on the Performance Ratio. If a player scored 9 in 9 games, his or her Performance Ratio is 1.00 and Performance Change +800. If he scores 4.5 in 9 games, it is correspondingly 0.50 and Performance Change will be 0. The needed value is taken from the Table of Performance Change.
Please notice that your Performance Rating does not depend on your own rating but does depend on your Opponents' Average and "how you performed" (Performance Change). Performance Rating is very important for getting the Grandmaster and International Master norms."
|Jul-08-15|| ||Sokrates: Great accomplishment by Caruana. He appears as the most serious opponent to Carlsen once again (after a relatively weak period). Also great were the two runner ups, So and Nisipeanu. Kramnik once again had to suffer three defeats, but I think he played some good games too.|
|Jul-08-15|| ||Gypsy: <Sokrates: ... Kramnik once again had to suffer three defeats, but I think he played some good games too. >
Hou Yifan vs Kramnik, 2015|
|Jul-08-15|| ||AylerKupp: <<jith1207> Who would have thought that <ChemMac> with that complex theorem avatar would give such a simplistic explanation to <Wavy>'s math problem.>|
The mark of a truly knowledgeable person is to make any explanation simple, no matter how complex the subject. That's why my explanations are always so verbose and complicated.
|Jul-08-15|| ||AylerKupp: <ChemMac> A minor correction: According to the FIDE Percentage Expectancy table converting rating differences to P(Win), it is a rating difference of 800 points, not 400, that results in a P(Win)=1. I do agree with your comment about meaningless but, being obsessed with the illusion of precision, that's why I wanted to know the official FIDE meaningless calculation for TPR. But TPRs (which were my original question) are not used to estimate expected results; they can only be calculated after the tournament is over to see how much better or worse the player performed relative to his rating and the rating of his opponents.|
Calculation of the expected results can be done the way you indicated <provided that there is not a large difference between the players' ratings>. The reason for this is that the FIDE Percentage Expectancy table is derived from the cumulative probability distribution based on the normal distribution, and this cumulative probability distribution is an S-shaped curve which is pretty much linear as long as the rating differences are relatively small, say 150 points, but becomes increasingly non-linear as the rating differences go beyond that range, You can find the FIDE Percentage Expectancy Table in sections 8.1a and 8.1b of https://www.fide.com/fide/handbook.....
If you are interested in the subject of ratings (and I can't imagine why you would be), here are links to 3 articles that you might find interesting given all your spare time: http://en.chessbase.com/post/sonas-... (he apparently never wrote a Part 2), http://en.chessbase.com/post/the-el..., and http://www.randalolson.com/2014/05/.... But be forewarned, reading these articles may have an adverse effect to your apparently rapidly deteriorating IQ. ;-)
|Jul-08-15|| ||AylerKupp: <choosea> It looks like you found "Chess Elo Rating in Simple (Terms?)", http://www.chesselo.com/index.html. However, the "simple" description seems like something I would write.|
And it is unfortunate that, since "Performance Rating is very important for getting the Grandmaster and International Master norms, that the FIDE method for calculating is not quite correct. Their Performance Expectancy Table (see my post to <ChemMac> above> is based on a normal distribution and, while this is appropriate for a larger number of games (30 or more according to both Dr. Elo and FIDE itself; see https://www.fide.com/fide/handbook....., section 8.56), I believe that it is not appropriate for the relatively small number of games played in a single tournament.
For these smaller number of games a Student or t-distribution is more appropriate, and this distribution is different depending on the number of games in the tournament. But a cumulative probability distribution derived from a t-distribution is almost identical to the cumulative probability distribution derived from the normal distribution used to calculate TPRs if the rating difference between the player and his opponents is relatively small, and only becomes relatively large as the rating difference increases. So it probably makes no significant difference in round-robin tournaments when the players' ratings are similar, but it might make a difference in Swiss tournaments such as the upcoming World Cup 2015 where there may be a significant difference between the players' ratings.
|Jul-08-15|| ||Marmot PFL: <Conclusion: Caruana and Nisipeanu played well above their ratings, and everyone else without exception, allowing for the underrating of those two players, performed exactly as expected. So much for any perceived aging reduction of Kramnik! HE hasn't changed; other players have become stronger!>|
As Kramnik's rating drops his expected performance will also drop. So he can continue to meet his expectations even with lower scores.
Whether his rating decline is attributable mainly to age is debatable. Ratings for younger players like Radjabov and Aronian have also declined, while older players like Toaplov and Anand are still in top form.
Kramnik said as players age it's harder for them to concentrate, due to lowered testosterone. Maybe so, but motivation and overall health must also be factors.
|Jul-08-15|| ||Gypsy: <Marmot ...
As Kramnik's rating drops his expected performance will also drop. So he can continue to meet his expectations even with lower scores.
Whether his rating decline is attributable mainly to age is debatable. ...>
Kramnik's OTB skill is as good as ever; his opening prep is not. (Only, my impression, of course.)
|Jul-08-15|| ||Absentee: <Marmot PFL: As Kramnik's rating drops his expected performance will also drop. So he can continue to meet his expectations even with lower scores.>|
For his rating to drop he must first score below his expected performance.
In this case the egg of lower performance comes before the chicken of lower rating.
|Jul-08-15|| ||Marmot PFL: < Gypsy, Absentee> Good points, this wasn't Kramnik's worst Dortmund (he was 2nd to last a few years ago for instance) and has bounced back before. Needs better opening prep to get a few fast wins (or fast draws) and not have to grind out long games with younger players.|
|Jul-08-15|| ||ChemMac: <AylerKupp> and others:|
Yes; a 400-point difference should mean, if I recall, a 97% win probability, not 100%. *I* say: meaningless, for what was intended: a simple - or simplistic - quick way of estimating the likely scores in a tournament like Dortmund, with a spread of not much more than 100 rating points, or the forthcoming Sinquefield Cup, which I could do in my head. We now know that Caruana, for one, is capable of considerably surpassing such expectations, because he has done so twice in the past two years.
That Sinquefield Cup!! It will be, I think, pretty exciting, with perhaps less drawn games than usual in super-GM tournaments. Unfortunately I'll be in London when it starts, away from the Internet.
|Jul-09-15|| ||AylerKupp: <ChemMac> Close enough, a 400-point rating differential results in a P(Win) ~ 0.92, you need to get to a rating differential of ~ 538 in order to get a P(Win) ~ 97%. But to get a player's expected score for a tournament you just need to figure out the difference between the player's rating and the average score of his opponents, look up the P(Win) from the table, multiply the number of games to be played by the P(Win), and round it to the nearest 0.5. By now I have a spreadsheet template where I enter the players' names and their rating and it calculates all those most likely expected scores plus, after the tournament is over, each player's TPR and rating gain or loss.|
But surely they have Internet cafes or their equivalent in London. Will you have an International Internet-capable smartphone? You can Google "Free WiFi London" and get a map of lots of spots, depending on where you will be. Even if you only use it once a day, that will be enough to keep up with what's happening. All eyes will be on Carlsen to see how he rebounds from Norway 2015 and, as you mentioned, Caruana seems to be in good form.
I've never been to London and hope to be able to spend some time there in the future. I have some friends who are avid theater-goers and they go to London at least once per year.
|Jul-10-15|| ||choosea: > results in a P(Win)=1
> "Odds are a numerical expression, always consisting of a pair of numbers, used in both gambling and statistics. In statistics, odds for reflect the likelihood.."
> ???????? : 1
> "... The sum of the probabilities of all elementary events will always = 1. "
> "..we have decided to exclude...from the population for the purpose of selecting a sample. Thus probability ... is zero." ????????????????????????????????????????
>> "Known Probability: .. each .. has a probability ... Both these probabilities are known,though they are not equal."
> "..represents the probability that ..will win. ..represents the probability he will not win."
> "Probability occurs always"
> "..an impossible event and a zero probability event is different.."
> "This is an ambiguous phrase which would be interpreted in several different ways"
> "I have heard it said that anything passed 1 in 10^50 is considered zero probability. "
> "A sure event is the one that contains the whole sample space. "
> "There will be cases in which it is not clear if it is a sure event."
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
|Jul-10-15|| ||AylerKupp: <choosea> The only thing I can comment on is the P(Win)=1 statement. Technically this can't be true because, although it is probably infinitessimaly small, there is a non-zero probability that the player with a rating advantage of more than 800 points will have a fatal heart attack, be struck by a meteor, etc. So it is probably better to speak of P(Win)>0.99 ot whatever. However, I am not going to lose much sleep over it.|
|Jul-10-15|| ||Marmot PFL: <Technically this can't be true because, although it is probably infinitessimaly small, there is a non-zero probability that the player with a rating advantage of more than 800 points will have a fatal heart attack, be struck by a meteor, etc. So it is probably better to speak of P(Win) 0.99 ot whatever. >|
I don't know how common they actually are. I suspect they might be more common, because of improving young players whose ratings haven't caught up to their ability.
|Jul-10-15|| ||latvalatvian: Re: the iron bound laws of Steinitz: it's a metaphor.|
|Jul-10-15|| ||plang: <it's a metaphor.> |
it's a metaphor what?
|Jul-27-15|| ||AzingaBonzer: <toyosaki: There's no elo inflation.>|
Therefore, Wesley So at his peak was stronger than Fischer at his peak, right?
|Jul-27-15|| ||AylerKupp: <AzingaBonzer> Hard to say. At 21 Wesley So has probably not yet reached his peak. Do you have any basis for saying that Wesley So at his peak (whenever that happens) will never be stronger than Fischer at his peak?|
|Jul-28-15|| ||AzingaBonzer: <AylerKupp :Hard to say. At 21 Wesley So has probably not yet reached his peak.> Sorry, I was unclear. By "peak", I meant So's peak rating so <far>: 2788. For comparison, Fischer's rating was 2785. It's absurd to think that a 21-year-old, no matter how talented, is stronger than Fischer at his peak, who effortlessly tore through Soviet grandmasters back in the days when the Soviet Union practically dominated chess. And yet that is exactly what you are forced to conclude if you accept that there has been no rating inflation during the intervening years. (By the way, I understand that this is not your position, but it is quite obvious from <toyosaki's> comment that it is one that he holds, and my original comment was addressed towards him.)|
<Do you have any basis for saying that Wesley So at his peak (whenever that happens) will never be stronger than Fischer at his peak?>
Certainly. Fischer was No. 1 in the world. So far, the highest Wesley So has been is No. 7, and I doubt he'll reach Carlsen's level (for comparison, Carlsen was No. 1 when he was 19, a full two years younger than So). The only counter-argument to this is a smaller player pool back in Fischer's day, so No. 1 meant less (to be No. 1 out of 100,000 players means you're at the 99.999th percentile, but to be No. 1 out of 1,000,000 players means you're at the 99.9999th percentile - ten times rarer), but this is a fairly weak argument in my view.
After all, to be No. 1, it matters far less how big of a pool of players you're competing with, but how big of a pool of <immediate> competitors you have, i.e. the players strong enough to be playing Fischer on a regular basis. And as I mentioned above, this was the era of Soviet domination, of Spassky and Petrosian et al. It's unclear to me that this particular pool of players was any smaller than the pool that So has to compete with today. The only way to make the no-inflation argument work is to suppose an increase, not in quantity, but in <quality>, and moreover, such an enormous jump in strength that the No. 7 today is stronger than who many consider to be the either the best or second-best player of all time. I mean, <really>? Computer preparation can only take you so far. Until a way is found to objectively analyze chess strength, of course, such a judgment is doomed to subjectivity, but let me put it this way:
Suppose a foolproof way to quantify chess strength is found at the end of Wesley So's career. I tell you that I'm about to analyze So's strength and compare it to Fischer's. Before I do, however, I offer you a 1:1 bet of US $1,000 (inflation-adjusted) that So will not be stronger than Fischer. Would you accept such a bet? If your answer is "no", (and I greatly suspect that even if it is not, a large number of people's answers would be), then you are implicitly revealing that you consider the odds of So being stronger than Fischer to be less than 1:1, i.e. he has less than 50% probability of surpassing Fischer. If I asked a bunch of people this question (ignoring Filipino chess fans' answers for reasons that should be obvious), my guess is that more people would answer "no" than "yes".
And yet his peak rating so far is 2788--3 points higher than Fischer's. And as you mentioned, it is unlikely for him to have even reached his true peak yet. Even if you think that So at his peak might (with emphasis on the "might") be stronger than Fischer at his peak, how likely is it that he's stronger <now>? Rating inflation is practically the only reasonable hypothesis.
|Jul-28-15|| ||Gypsy: <The only counter-argument to this is a smaller player pool back in Fischer's day ... > |
I am not even that sure about the smaller pool claim: Back in Fischer's time, practically everyone in USSR (and about every other guy in their satellite countries) played chess.
|Feb-29-16|| ||Lambda: I tried estimating the size of the player pool throughout history once upon a time by counting the number of players within 100 points of the world number 10, or some similar methodology. From the 70s onwards, the results were pretty constant. So the world number X today is probably of similar intrinsic strength to the world number X in the 70s, whatever "intrinsic strength" means.|
(The big jump was in the 1930s.)
|Feb-29-16|| ||perfidious: <AK: The mark of a truly knowledgeable person is to make any explanation simple, no matter how complex the subject. That's why my explanations are always so verbose and complicated.>|
< Earlier Kibitzing · PAGE 22 OF 22 ·
Spot an error? Please suggest your correction and help us eliminate database mistakes!
NOTE: You need to pick a username and password to post a reply.
Getting your account takes less than a minute, totally anonymous,
and 100% free--plus, it
entitles you to features otherwise unavailable.
Pick your username now and join the chessgames community!
If you already have an account, you should
Please observe our posting guidelines:
- No obscene, racist, sexist, or profane language.
- No spamming, advertising, or duplicating posts.
- No personal attacks against other members.
- Nothing in violation of United States law.
- No posting personal information of members.
See something that violates our rules? Blow the whistle and inform an administrator.
NOTE: Keep all discussion on the topic of this page.
This forum is for this specific tournament and nothing else. If you want to discuss chess in general, or
this site, you might try the Kibitzer's Café.
posted by Chessgames members do not necessarily represent the views of Chessgames.com, its employees, or sponsors.|
your profile |
Premium Membership |
Kibitzer's Café |
Biographer's Bistro |
new kibitzing |
Tournament Index |
Player Directory |
Notable Games |
World Chess Championships |
Opening Explorer |
Guess the Move |
Game Collections |
ChessBookie Game |
Chessgames Challenge |
privacy notice |
Copyright 2001-2018, Chessgames Services LLC