Olympiad Quiz – Way too early prediction

I took the freedom of guessing on some of the final results of the Quiz. They may change, but at the moment, Gollum is in the lead with a projected 11 correct answers. It will take a bit of time to get confirmed which Iranians that made norms and other details. And some games are still in play, meaning that the final results may change!

The excel sheet is here.

51 thoughts on “Olympiad Quiz – Way too early prediction”

  1. Obviously this a very fluid situation. The Eljanov-Beliavsky game could decide the winners of the Open Section and the Russia-China match is very unclear in the Women’s section.

  2. I think you misscalculated John, Andrew and Colin’s expected result .

    In the fide handbook you can look at the expected result (1.49 here: https://www.fide.com/component/handbook/?id=174&view=article).

    * Andrew Greet is at 2455 against a 2467.9 rivals (-12.9 elo difference), so his expected result from 10 games is 4.8 points (48%).

    * John Shaw is at 2454 against a 2393.4 rivals (60.6 elo difference), so his expected result from 9 games is 5.265 points (58.5%).

    * Colin McNab is at 2434 against a 2288.7 rivals (145.3 elo difference), so his expected result from 9 games is 6.255 points (69.5%).

    Hence the combined expected performance would have been to score 16.32 points. Between the three of them they got 16 points, hence they performed lower than expected.

    It is close, so it is entirely possible that I messed up somewhere, but if the calculus hold, then The answer to Q9 should be ‘No’.

  3. Common mistake would be to go by Elo +/-. For calculating Elo changes, the rating difference is capped at +/-400. But I haven’t checked if that is relevant here at all.

  4. Even so, Tobias’ comment on Elo changes and performance thing definitely matters in this case, since they played against a few 400- players and beat them.

  5. I think only Andreew met his expected result. The gaps between Colin and John’s performance and their expected look way more than Andreew’s plus.

  6. Of course Q9’s answer is no.How could they achieve their expected score? John and Colin finished with – rating and none of them could win against better player. Come on

  7. Outperform means to do better than expected. The expected result is calculated via the table in the link in my other post. Chessresult may cap at 400 elo points because elo is calculated that way, but the expected performance is not based on win/loss of elo.

  8. @Gollum
    We will definitely decide on whether or not they made their expected score on what the final change in rating will be. With opponent’s rated 1600 in this event, I would NEVER use those ratings in an average.

    Andrew won 8,7, John and Colin together lost 6. So yes, they made their expected score overall.

    When you decide on a performance score for a title and for rating change, you do cap the rating of the opponents. As we have not specifically said that we will not do as is done everywhere else that matters on handling the ratings on performance scores, by simple legal principles, we will do as is done elsewhere.

    I have updated she sheet. We will announce the winner when John is back, but everyone can see the outcome for themselves. The latest changes means we have a different winner than we had when I put in preliminary results yesterday.

  9. I am sorry but I have difficulty accepting the last situation of the “expected result”
    If you had wanted to ask about elo change, you could have easily said “Will John, Colin and Andrew combined finish plus according to elo at the end?”. Many have been confused with the real question 9 and then they decided to consider the performances each individual will do. In this respect, we can simply say that Andrew’s good performance is not enough to compensate Colin and John’s relatively bad ones. So my objection is that there was no simple ELO word.

  10. I am disappointed (mainly because I will not win), but as the participation is free, there is nothing to complain about, except that hope to have more luck next time, and thank QC for organizing all this.

    I must confess that I never took into consideration the elo gain, nor the exact formula from which to derive the answer to Q9 when I answered it, and it only makes a difference because it is so close.

    Maybe the proper way to calculate it would be to calculate the expected result in each individual game and add them all up (so a 800 elo difference is 1 for expected result, and a 400 elo point is 0.99, which is almost the same), and in that case the results may be the same as the elo calculation, but I’m not going to do it because it is a lot of work 🙂

  11. I think Gollum hits the point : it is a free contest so there can not be any complaint. However, if this particular point gives just two different winners can QC just split the prize between them?

  12. We will use the normal accepted ways of deciding what is the correct answer. What people wish for and accept is not going to be a factor, nor is who will win, which obviously will be decided on how we interpret the results and will lead to disappointment for some.

    Please understand that we cannot declare everyone as winners.

    I will debate it with the others at the first given opportunity, but here are a few of my own observations:

    1. I would not accept “clever” claims. See below.
    2. I would go by standard methods of rating performance calculation. This excludes 100% scores against 1900s as being 2750+, for example.
    3. To say that Andrew, John and Colin performed below expected score and at the same time won rating does not make sense to me.
    4. Only the first few questions were on the women’s Olympiad, so we have not and will not check if any of the women from Iran scored norms. Especially as the question refers to the “Iranian team” and not teams; again avoiding cleverness.

    The final winner will be declared shortly after everyone is back.

  13. @Gollum
    Q9 Will John, Colin and Andrew combined achieve their expected score?

    Expected score is a standard term used for calculation rating. Andrew won more rating than John and Colin lost. I really see this as non-controversial. The question is not: will two of them win rating and it is not, will they combined make a performance rating higher than their average rating.

    The question is about “expected score”. Not performance rating. The rating system operates with a maximum gap of 400 points in rating calculations, and I presume so will we.

  14. I do not mind the least that people argue their case here and actually find it quite reasonable. Who would not want to win? As long as everyone understands that we have no favourites and will decide on the fair interpretation of the rules without looking at the results.

    As I will be the only one in the office who have looked at the spreadsheet, I will not vote, but simply lay forward the possible interpretations. John, Colin and Andrew will vote on which interpretation that makes sense to them.

    None of our friends are among possible winners. If we wanted to give free books to our friends, we would not need to stage a competition. It is meant to be a bit of fun, but we will certainly learn from this experience and formulate the rules clearer for next time.

  15. The ELO rating is calculated per every game separately and then it is accumulated (the performance not, calculating ELO change based on average rating may give a bit different result). Still in both cases if the gap between the players is too big it has to be calculated as smaller gap, otherwise one could loose rating just by playing someone much weaker 🙂 One way or the other if we talk about expected score as performance or as total of ratings gains and losses I think they achieved that score anyway.

  16. The question is, “Will John, Colin and Andrew combined achieve their expected score?” and the judges are John, Colin and Andrew! Clearly I blundered on that question.

    Thanks for running this contest, it was good fun and increased my interest in the olympiad.

  17. Great competition but feel for Gollum. Won’t get his precious books…
    If it makes him feel any better my dismal score dropped similarly to a poorer than random score of 4. I bet on the Scotland team doing poorly as you’ve all been editing too much and not playing enough. delighted you proved me wrong! Thanks again guys.

  18. Btw. I saw that no one managed to score a golden 0 points either :-). Usually that would be my trademark!

    The real important thing is that I managed to score better than both John and Colin, while beating Sam Shankland on tiebreak!

  19. Hi QC team, thanks for the great (and free) competition – made me keep my fingers crossed for Scotland:) ps: What would 20 precious free books have turned Gollum into?!

  20. @Gollum
    There has been debates over Pearlescent or Metallic blue in the flag. I will post lots of photos in due time. First I want it to be finished. I guess it will be another three weeks at least. It needs quite a few layers of paint.

  21. I am not going to go through them all. If someone sees that their own score is incorrect, please tell me. The formula should generally work, I think.

  22. About one in 4/5 did not send in the solutions in the required format. I spent 30 minutes reformatting the last 150 or so once the deadline was up. Must have missed the extra spaces.

  23. On the Olympiad I thought the Shirov-Rapport game was crazy!!
    Yifan Hou losing to a lowly ranked politician was a surprise.

    Now the event is over I’m sure I speak for others when I say I’m looking forward the the excepts to be put up! Hopefully it will be this week

  24. On the spreadsheet, incidently the two guys who invented the spreadsheet made no money from it, I scored 5 points. So it’s probable 10 of my answers include either an errant space character or are otherwise badly formated. Please confirm my recalculated score is 15/15 and conceed the prize books are all mine.

    Colin’s draw with Shankland got a nice write-up in chessbase. There was no mention of the near-draw on board two of that match.

  25. @Alex Colovic

    The main blog notes strongly imply that you felt that the player in question was cheating.
    ” On board 4 GM Odeev (2401) … He was obliterated…. You can check the game yourselves with an engine if you wish – white’s play was on a very high level and even though he missed one or two stronger moves it proved enough to kill the GM….What happened was that white’s player was searched and an iPhone and an iPad were found on him. How on Earth could a player sneak through not one, but two electronic devices after all the apparently (but only apparently) strict security? ”

    Any reasonable reader infers from this that you are accusing the player of cheating there.

    It is helpful that you backtracked on this in the comments, after the intervention of the captain, for which you can be commended. This shows grace. Ideally, the comments should not have been written in the first place. It would be helpful if you were to offer an apology to the player in question. It may also be helpful to delete the section of the post naming the individual, or the whole post.

    Future employers of the young fellow will search online before employing, as happens when employing any new staff, and such comments are unhelpful, to say the least.

    I spoke to the Japan players on the weekend in Tokyo at a tournament. Naturally, there was some discussion about this issue. As a higher board put it to me: “A 2100 player beats a GM. Then a…

  26. @Simon Bibby

    Simon, your comment seems to have been ended prematurely by the blog software (there is a limit on comment length). You can complete your thoughts in another comment. Just mentioning this in case anyone thought a QC human was censoring you.

  27. …phone is found on him. That looks terrible.”
    Indeed, yes, but entirely innocent. I was told that the player rushed to the bus (nearly missed it) and had not emptied his pockets. Disorganised, but not cheating.

    This is discussed on ECForum, where the esteemed chess columnist Leonard Barden notably concludes that the game shows nothing of cheating. He’s right, of course. The evidence is there – it is the game score. Black, a low-ranking GM (around 2400), just self-destructs. Not a good game really.

    There is something on which I can certainly agree with Alex Colovic, from all I have heard. It is unfortunate that the front gate screening was sub-optimal, and failed to find the phone on the way in. Most unfortunately, for the player and the Japan team.

    Secondly, and most importantly, it is poor that the ‘anti-cheating’ approach put in place had a single result – a false positive. Someone busted for cheating, who plainly did not. It surprises me that this has not been discussed. Is this not considered a serious problem? Or does it not matter, as the player is ‘just’ an amateur?

    (sorry software, sorry John S, thanks)

  28. To my opinion every player who wants to play in an event like the chess olympiad has to show at least a little bit of professionality.
    Everybody knows of the cheating issues and the anti-cheating rules, be the measures taken good or bad.
    I cannot feel any sympathy for someone being defaulted for bringing a mobile phone along – it’s his own fault.
    “late for the bus”? “forgot to empty his pockets”? “found it in my conflakes”?
    More excuses?

  29. The circumstances were presented to me as *what happened*, by a team mate. An annoyed team mate who was waiting on the bus. Stupidity is often mundane. It is what it is. Yes, stupid. And I am sure the player now sees that. Dealing with university age youngsters every day as I do (I teach at uni), sadly being disorganised is not so uncommon. Agreed, little sympathy there.

    But, to be clear, not ‘cheating’.

    ‘conflakes’? Is that a breakfast eaten by conmen? Before being caught no doubt, and doing porridge.

  30. @Simon Bibby: I fully agree. Alex Colovic did complain about the new anti cheating measures – mainly the possibility of checks during the game. And rightly so. On the other hand we all have to work towards an atmosphere with less paranoia and prejudge, as otherwise the hardliners will take over with more escapist measures against the nature of competitive chess. This shouldn’t mean to ignore the cheating problem, but to find appropriate measures – also with the help of players.

    It might not be rational (and I see that there is no other way than to declare the game lost for the japanese player), but I still feel at least some sympathy for anyone who loses outside the board, because few individuals have cheated in the past (and yes, also because they themselves are disorganized).

  31. ‘conflakes’ – I’m too stupid for my keyboard.

    I have a teammate who is always some minutes late for the match, or if it’s a match in another town, smoe minutes for the appointed meeting.
    Every time I swear I’m gonna kill him, but he’s been lucky so far.

  32. Thomas :
    I have a teammate who is always some minutes late for the match, or if it’s a match in another town, smoe minutes for the appointed meeting.
    Every time I swear I’m gonna kill him, but he’s been lucky so far.

    Just tell him a different time for the appointed meeting than anybody else knows – 5 minutes earlier or so. Might work at least once 😉

  33. @Simon Bibby
    I agree that “any reasonable reader infers” but that just shows how far we’ve gone in the fear of cheating, we look for it everywhere even if it isn’t there. If you can, try reading the passage without knowledge of cheating, just the plain words – it says the guy played a great game and killed a GM. The “engine” is there so that a reader can check the quality of the game himself. What I’m trying to say is that cheating has got so deep into our (sub)conscious that it’s impossible not to think about it. Of course, once I discovered the facts, thanks to the comments by their captan, GM Stojanovic, I immediately offered apology, even though I meant no harm nor accusation. I am the last person to wrongfully accuse anybody, for each of my most famous cases of being cheated on (http://www.alexcolovic.com/2015/04/my-experience-with-cheating-part-one.html and http://www.alexcolovic.com/2015/05/my-experience-with-cheating-part-two.html) there have been dozens where I didn’t dare do it because I wasn’t sure. In these two however I was dead sure!

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top