Latest Updates:
Page Index Toggle Pages: [1] 2 3 
Topic Tools
Very Hot Topic (More than 25 Replies) What does being an 'openings expert' really mean? (Read 10355 times)
an ordinary chessplayer
God Member
*****
Offline


I used to be not bad.

Posts: 935
Location: Columbus, OH (USA)
Joined: 01/02/15
Re: What does being an 'openings expert' really mean?
Reply #38 - 05/17/18 at 19:13:51
Post Tools
@ReneDescartes - I think he quoted you not because it was your idea, but because you articulated it clearly.

Nickajack wrote on 03/31/18 at 04:40:32:
Another problem for some of us in this discussion is that 'opening expertise' is an ill-defined concept.

Actually the definition is what the thread is about. Your hippo example is good. Anti-theory can also be theory.
  
Back to top
 
IP Logged
 
ReneDescartes
God Member
*****
Offline


Qu'est-ce donc que je
suis? Une chose qui pense.

Posts: 1157
Joined: 05/17/10
Gender: Male
Re: What does being an 'openings expert' really mean?
Reply #37 - 05/17/18 at 18:43:01
Post Tools
I was just explaining what I took brabo to mean, which had been understood backwards.

The possibility of self-selection bias? That's not a flaw in the comparison. What occurred is still a piece of evidence. A randomized controlled study hasn't been done, so we don't have as much or as good evidence as we'd like. But such studies, too, almost always have some possible confounding variables. We don't know too much, but least brabo is arguing from experience.
  
Back to top
 
IP Logged
 
Nickajack
Junior Member
**
Offline


Yabba Dabba Doo!

Posts: 52
Location: USA
Joined: 02/21/17
Gender: Male
Re: What does being an 'openings expert' really mean?
Reply #36 - 03/31/18 at 04:40:32
Post Tools
ReneDescartes wrote on 03/07/18 at 21:35:55:
. Fischerrandom results would then not show an extremely highly correlation with normal ratings. That the two types of results did turn out to be highly correlated is thus a piece of evidence against opening knowledge's having a big role in determining normal rating.


The flaw in this comparison is that we're only comparing the ratings of those who chose to play FRC in the first place. Players disliking FRC would not have competed and would have no such FRC ratings to compare. We would have to require all players to compete in FRC before a fair or exact comparison could be made.

Another problem for some of us in this discussion is that 'opening expertise' is an ill-defined concept. To most it probably means being booked-up, knowing long variations in fashionable openings, or a great variety of lines, etc. But, what if it should mean just having the ability to carry out the first phase of game with skill, regardless of the extent of one's actual memorized knowledge? I suppose a better word to describe such skill would be 'understanding', rather than 'knowledge'.

For instance, what would we call a player who excels at playing a weird opening like the Hippo, or anything else along those lines, with great success? On one hand he is avoiding the opponent's main book prep, while he's employing his strong understanding of opening principles to outplay the opposition on his own turf, so to speak, where he has a lot more familiarity with the possible set-ups. Wouldn't such a player also be deserving of the 'opening expert' classification?

[Such a player also would be likely to play FRC quite well, at least no worse when compared to his regular chess.]





  

Dubious, therefore playable Undecided
Back to top
 
IP Logged
 
kylemeister
God Member
*****
Offline



Posts: 4655
Location: USA
Joined: 10/24/05
Re: What does being an 'openings expert' really mean?
Reply #35 - 03/10/18 at 02:38:59
Post Tools
Hmm, I guess you're probably referring to that RHM book from the '70s.  Now you've gone and made me curious as to the 10th move in question.

(Incidentally, the German ChessBase site just had a piece on Geller, on what would have been his 93rd birthday.  Includes a pic from a game Geller-Tal [battle of smokers].
https://de.chessbase.com/post/zum-geburtstag-von-efim-geller)
  
Back to top
 
IP Logged
 
an ordinary chessplayer
God Member
*****
Offline


I used to be not bad.

Posts: 935
Location: Columbus, OH (USA)
Joined: 01/02/15
Re: What does being an 'openings expert' really mean?
Reply #34 - 03/10/18 at 00:35:28
Post Tools
TopNotch wrote on 03/09/18 at 17:03:31:
Actually to the above could be added a third group:

Opening Memorytician ...

This third group is the one we used to call "booked-up fish". Actually at one time it was me. I remember the turning-point game very well. Me at 1600 Elo was playing the white side of a Najdorf against a master. He made an inaccuracy at move 10. At move 16 it was +/- ("black does not have enough compensation for the Exchange"), but sadly it was also the end of my book. After that, except for one forced recapture, every one of my "own" moves was a mistake, and at move 22 I resigned because I was losing a rook, when I would be a piece down.

But there is more! In the post-mortem he admitted he didn't know where he went wrong. I showed him the correct 10th move, he said "Oh! Because..." and before I could get another word out of my mouth he played the theoretical variation and gave the assessment "good for black", which is the same thing Geller wrote in the book. Yes, it's true, I was not only booked-up, I was "overbooked". I played a second-rate line for white because I figured my opponent would know the best line and would not know the sideline. And at least in that calculation I was correct.

But there is more! Because even though I knew Geller's assessment, when I had looked at it at home I had had no idea how to continue for black (or white for that matter). I remember actually trying to analyze it for white and giving up because I didn't know what black should be trying to do. (This was before computers, obviously.) So basically I would have been helpless at any point I was out of book, whereas the master was able to re-invent the book on his own.

This one game really drove home to me that I was wasting my time studying sharp openings. Within a few months I revamped my repertoire to stuff like 1.b3 for white, 1.e4 Nc6 and 1.d4 e6 2.c4 Bb4+ for black. Basically I forced myself to learn the middlegame by playing non-theoretical openings where memorization would not help. I played a lot of random junk up until about 2300, at which point I had to knuckle down once again and get a decent repertoire. But in the meantime I was able to devote a lot of time to the middlegame and endgame. And the constant experimentation in the opening allowed me to eventually find the best openings for me, which are nothing like the razor-sharp lines I was playing as a junior.
  
Back to top
 
IP Logged
 
TopNotch
God Member
*****
Offline


I only look 1 move ahead,
but its always the best

Posts: 2057
Joined: 01/04/03
Gender: Male
Re: What does being an 'openings expert' really mean?
Reply #33 - 03/09/18 at 17:03:31
Post Tools
From my Human over the board perspective, Openings Expert is best divided into two groups, as follows:

Opening Practitioner: This type of expert knows his own opening repertoire extremely well and has a deep understanding of the typical positions arising from them. An example of such a player in the UK is GM Mark Hebden.

Opening Theortician: This type of expert has a wider knowledge of more Openings and Opening lines, but often lacks the depth of understanding of the various nuances arising from them that  accumaltes over time from sticking to a well worked out and well tested repertoire.   

Actually to the above could be added a third group:

Opening Memorytician: This class of expert is usually found at club level, he has memorised reams of long theoretical lines but his understanding of the Opening these lines spring from is often superfical. The memorytician can reproduce many lines but if questioned struggles to explain the plans and ideas behind the moves. This kind of expert often complains that they can't understand why they play better against masters and struggle against 'weaker' players. The reason is simple, weaker players unwittingly depart from established theory very quickly and without a firm strategic and tactical grasp of the opening one is playing it is more difficult to punish such deviations. On the other hand, when facing a master the memorytician gets to mindlessly regurgitate his memorised book lines, confident the master will follow established theory and perhaps not be familiar with the latest wrinkle. On second thought, can a memorytician really be be classified as an opening expert at all? I leave that question up to the readers.   

Ideally if we improve or calculation and evaluation skills and apply that to opening study, rather than jumping from one opening to another after every painful loss, we should see marked overall improvement in our play.

Bye 


  

The man who tries to do something and fails is infinitely better than he who tries to do nothing and succeeds - Lloyd Jones Smiley
Back to top
 
IP Logged
 
brabo
God Member
*****
Offline


Welcome chessfriend

Posts: 951
Joined: 02/02/07
Re: What does being an 'openings expert' really mean?
Reply #32 - 03/08/18 at 07:37:09
Post Tools
TalJechin wrote on 03/08/18 at 00:08:39:
Maybe he only gains 50 Elo points compared to others - I'm not sure how to compare that in a fair way. The example with engines limps a bit, since engines don't really use their books to create their own novelties - maybe Alpha Zero did in a way by playing itself on a supercomputer, and the general opinion seems to be that it's more than 50 points stronger than Stockfish...

Don't drag the match Alpha Zero against Stockfish in this discussion! There are too many loose ends as it was purely an advertising campaign from Google.

To measure the impact of opening study you need to collect data and make some statistical analysis. I did for my article. It makes little sense to discuss this topic any further if you don't approach it more scientifically.
  
Back to top
WWW  
IP Logged
 
brabo
God Member
*****
Offline


Welcome chessfriend

Posts: 951
Joined: 02/02/07
Re: What does being an 'openings expert' really mean?
Reply #31 - 03/08/18 at 07:25:33
Post Tools
an ordinary chessplayer wrote on 03/07/18 at 23:02:29:
What doesn't make sense to me is the idea that since it is "only" 50 Elo, it is not worth acquiring. Getting += as white is quite a pleasant feeling! I want it, and I am willing to do the homework to get it. The same goes for net 50 Elo in the other phases of the game.

I remember a supergrandmaster (Anand ?) once stating that he feels sorry for all the players working hard at chess and never becoming good at it. A Dutch FM once told me that he found my opening-study ridiculous. At 2250 you are a patzer and at 2300 you are still a patzer.
I know many people think this way.
Still I never cared about those opinions. I play and study chess for myself. I feel good and my blog is a clear proof of how much I love the game.
So I am the last person that will tell anybody to do this or that. No for me the only point in this discussion is what to expect from opening-study in terms of rating gain. I believe that is rather limited surely if we compare with other methods to gain rating.
  
Back to top
WWW  
IP Logged
 
TalJechin
God Member
*****
Offline


There is no secret ingredient.

Posts: 2892
Location: Malmö
Joined: 08/12/04
Gender: Male
Re: What does being an 'openings expert' really mean?
Reply #30 - 03/08/18 at 00:08:39
Post Tools
I guess we're talking about different things by now. To me an opening expert is someone who has specialised deeply into opening theory, keeping up with current development in a lot of different openings and analysing one's own novelties and so on - which is quite rare.

While opening knowledge is very common, most 1800 players knew 20 moves in the Ruy Lopez ML already in the 1980s, before the chess book flood. But here in Malmö, except for 2-3 GMs, there's only one player I'd say is an opening expert (and it certainly isn't me, btw). We've played in the same team a few times over the years, so I've seen him win some very quick games - and also some quick draws when the opponents were up to the challenge.

Maybe he only gains 50 Elo points compared to others - I'm not sure how to compare that in a fair way. The example with engines limps a bit, since engines don't really use their books to create their own novelties - maybe Alpha Zero did in a way by playing itself on a supercomputer, and the general opinion seems to be that it's more than 50 points stronger than Stockfish...



brabo wrote on 03/07/18 at 21:44:31:
A+B+C = X
B+C = Y
X = Y
What is A?
A = 0

Replace A with opening knowledge, B with middlegame skills, C with endgame skills, X = standard chess rating, Y = Fischer random rating

It is a very simplified mathematical model but easy to understand.


For one thing, since the middlegames are quite peculiar in FR you could just as well say that C=Y and A+B=0
As someone said, chess is 99% tactics (though that's not mentioned in your simple formula) and the higher your rating the better you calculate...
  
Back to top
 
IP Logged
 
an ordinary chessplayer
God Member
*****
Offline


I used to be not bad.

Posts: 935
Location: Columbus, OH (USA)
Joined: 01/02/15
Re: What does being an 'openings expert' really mean?
Reply #29 - 03/07/18 at 23:02:29
Post Tools
brabo wrote on 03/07/18 at 21:44:31:
A+B+C = X
...
Replace A with opening knowledge, B with middlegame skills, C with endgame skills, X = standard chess rating, ...

@brabo - I could go on at great length as to why your so-called model does not in fact model the data, but since I can't do that without being insulting, I will skip it. Maybe your model was tongue-in-cheek, if so then it was worth a chuckle.

@everybody - To the extent openings expertise is equivalent to memorization, then I agree Fischer Random does remove it from the equation (where the equation looks nothing like brabo's). If openings expertise has other components, then the contribution in Fischer Random is not necessarily zero.

Despite my constant fencing in this thread, I don't really dispute Brabo's point that "high expertise" in the openings is probably worth less than 50 net Elo compared to normal ability, as long as we agree that the normal baseline is far greater than zero. Ceteris paribus, 50 net Elo would raise the winning expectancy in a game from 0.50 to 0.57, roughly the difference between = and +=. That makes total sense to me, and it feels right when I compare it to my own results when I am "out of book" versus "in book". What doesn't make sense to me is the idea that since it is "only" 50 Elo, it is not worth acquiring. Getting += as white is quite a pleasant feeling! I want it, and I am willing to do the homework to get it. The same goes for net 50 Elo in the other phases of the game.
  
Back to top
 
IP Logged
 
brabo
God Member
*****
Offline


Welcome chessfriend

Posts: 951
Joined: 02/02/07
Re: What does being an 'openings expert' really mean?
Reply #28 - 03/07/18 at 21:44:31
Post Tools
A+B+C = X
B+C = Y
X = Y
What is A?
A = 0

Replace A with opening knowledge, B with middlegame skills, C with endgame skills, X = standard chess rating, Y = Fischer random rating

It is a very simplified mathematical model but easy to understand.
  
Back to top
WWW  
IP Logged
 
ReneDescartes
God Member
*****
Offline


Qu'est-ce donc que je
suis? Une chose qui pense.

Posts: 1157
Joined: 05/17/10
Gender: Male
Re: What does being an 'openings expert' really mean?
Reply #27 - 03/07/18 at 21:35:55
Post Tools
I think brabo meant that if opening knowledge had a big effect on normal rating, then we would see in fischerrandom that players with lots of normal-chess opening knowledge were much weaker without this advantage, while those relying on other skills in normal chess were not weaker. Fischerrandom results would then not show an extremely highly correlation with normal ratings. That the two types of results did turn out to be highly correlated is thus a piece of evidence against opening knowledge's having a big role in determining normal rating.


Edit: ah I see that dfan beat me to it.
« Last Edit: 03/08/18 at 11:52:27 by ReneDescartes »  
Back to top
 
IP Logged
 
dfan
God Member
*****
Offline


"When you see a bad move,
look for a better one"

Posts: 732
Location: Boston
Joined: 10/04/05
Re: What does being an 'openings expert' really mean?
Reply #26 - 03/07/18 at 21:25:24
Post Tools
TalJechin wrote on 03/07/18 at 20:19:23:
"if opening knowledge plays a serious role then we should see this in Fischer Random"

Isn't that a bit like saying that reading a text in Thai would prove if it's good to be able to read English? Fischer Random removes the opening theory, so how could it say anything about the worth of knowing opening theory?

brabo's point, as mentioned in an earlier comment, is that the more important opening theory is, the less Fischer Random ratings should correlate with regular ratings, but he observed that the ratings have a high correlation.
  
Back to top
 
IP Logged
 
TalJechin
God Member
*****
Offline


There is no secret ingredient.

Posts: 2892
Location: Malmö
Joined: 08/12/04
Gender: Male
Re: What does being an 'openings expert' really mean?
Reply #25 - 03/07/18 at 20:19:23
Post Tools
"if opening knowledge plays a serious role then we should see this in Fischer Random"

Isn't that a bit like saying that reading a text in Thai would prove if it's good to be able to read English? Fischer Random removes the opening theory, so how could it say anything about the worth of knowing opening theory? It may prove that the rating system works though, but a high rating doesn't mean that you're an opening expert.

One idea could be to compare ten games from two players of roughly equal strength and note how they score their wins. A theoretically prepared player could be expected to gain an edge in the opening and keep it or enlarge it until it's a win. Someone who plays less theoretically would presumably have almost no advantage in the opening and outplay the opponent later somehow.

Any players that fit the bill and both played in for example the same world team championship (that way you could choose the ten games with highest average rating from both)? Maybe Marcus Ragger vs Granda Zuniga - or some others? - those were the first to come to my mind.

  
Back to top
 
IP Logged
 
brabo
God Member
*****
Offline


Welcome chessfriend

Posts: 951
Joined: 02/02/07
Re: What does being an 'openings expert' really mean?
Reply #24 - 03/07/18 at 17:43:33
Post Tools
an ordinary chessplayer wrote on 03/07/18 at 16:02:27:
Again, this may not prove what you think. It's possible that openings knowledge correlates rather well with middlegame and endgame ability.

True but then you assume that people of the same strength have more or less the same opening knowledge. Reality is different. We do see big differences even when players have exactly the same rating. Therefore if opening knowledge plays a serious role then we should see this in Fischer Random. However it is not.
  
Back to top
WWW  
IP Logged
 
Page Index Toggle Pages: [1] 2 3 
Topic Tools
Bookmarks: del.icio.us Digg Facebook Google Google+ Linked in reddit StumbleUpon Twitter Yahoo