Friday, April 27, 2018

Fashion part 2

My previous article proved that there exists an enormous variety of openings. It is today completely impossible even as super-professional to know all existing deviations in your repertoire by heart. Nevertheless we shouldn't overestimate the value of openings especially at amateur-level. Many of my opponents don't play critical lines at all and just want to skip the opening to get a playable position in which the player with the strongest technique instead of memory can win.

I mean an ambitious player should rather focus the study of openings only at the critical lines. The many less dangerous systems can be checked too but in most cases it is sufficient to develop your pieces and get an acceptable middle-game. Besides even if you concentrate at the remaining important openings then still a lot of study remains as we can see by the testimonies of many professionals.

I believe it is not so easy for an amateur to make the distinction between important and less critical openings. How should I recognize which openings I need to study or which ones can I ignore? A coach can surely help here but not everybody has access to such luxury. In any case it is wrong to believe that openings played by higher rated players will automatically be more profound. As proof I checked my 100 most recently played games which I split in ratinggroups. For each of the groups I defined the average depth of the opening (linked to the deviation compared with earlier played games). I start with the games in which I had white.
For black we don't see any important difference.
So it seems there exists no link between somebodies rating and the opening-knowledge. No that is a wrong assumption. A higher rated player will have a better knowledge about openings. The confusion is created by the mix between opening-knowledge and opening-novelties. 99% of my opponents are followers. I mean that they will rarely use original ideas which are worked out at home in their openings. This isn't a surprise as the rating-groups consist almost exclusively of amateurs spending little or no time at doing individual research of openings.

The real pioneers and leaders of openings are of course our top-players and we should not forget correspondence-players. If we want to know which openings are critical then we should first check their games. This is something I do already for some years see e.g. how I described the preparation of my games in the article using databases or check my article the expert part 2 in which I stated that I focus at the games of +2600 rated players.

However keeping up to date the critical openings is something easier said than done as today we have more than 200 + 2600 rated active players. So every day there are novelties popping up which a professional needs to check. Last the Dutch top-grandmaster Anish Giri twittered that "If more than 2 twics are missing on your laptop then something went wrong in the routines." In my article how much money do you spend at chess I also wrote that I download at least twice per year the free twics but I don't have/ spend the time to filter all relevant +2600 games for my repertoire to study the novelties.

I guess for most amateurs it is the same. You can't expect amateurs working every day at their repertoire. Besides it doesn't make much sense. Much more interesting for us is to use summaries made by a professional which explains all the new critical opening-lines. This allows us to get up to date very quickly by a minimum of effort. Openingbooks and dvds are our first address to check. The disadvantage of those media is that they are very quickly outdated and can't manage to keep track of the latest trends. To really follow the fashion, chess-magazines need to be consulted preferably with the accent solely at openings. I recommend 2 subscriptions: chesspublishing.com and chessbase magazine.

Of course you will see quite some overlap between both magazines. Although different authors are working for the magazines, the same trends are noticed. That was last the case for the Armenian Winawer-line of the French defense Both published beginning of January a summary of the most recent developments in that system see chesspublishing January 2018 and chessbase magazine 182. Strangely I encountered the line already before newyear in my game against the Belgian expert Nathan De Strycker played  in the 56nd Christmas-tournament of Deurne.
I was lucky that my opponent couldn't remember well the analysis of this line as otherwise the game could've gone differently. As I don't have a subscription, I decided to make my own extensive analysis about the opening. I was surprised that this hypermodern-system is very playable for black see below the summary. I made my analysis using the modern monte carlo-mechansim so by playing many quick computer-games see my article computers achieve autonomy.
So now we finally know who creates new trends and how we can easily detect them. The next question is how interesting this is for the amateur. Well this isn't an easy question as I am not acquainted with the content of the magazines. I guess averagely once per year something could be useful for my standard games. That is not much but also largely depends on my own choices. I almost don't vary my repertoire as I use the scientific approach to choose openings. Also I play few games : I wrote 38 with a standard timecontrol last year in my article surprises part 2.

Briefly almost 100 euro per year for a magazine is a lot of money for me. For professionals the situation is very different. Last the Amercian grandmaster Alexander Lenderman grumbled with the famous Russian proverb "скупой платит дважды" after his painful defeat in 22 moves a week ago against the Amercian top-grandmaster Fabiano Caruana. The opening was treated in the most recent edition of newinchess yearbook 126 (this magazine is released only 4 times per year contrary to the monthly editions of chesspublishing and bi-monthly editions of chessbase) but Alexander missed it while Fabiano didn't.
My Russian father-in-law uses the proverb also regularly. A cheap person pays twice. Losing an extra half point in the US-championship can very well cost many times more than 100 euro see US-prizemoney and I don't consider yet other interests like qualification for the olympiad, the title,...

Let us go back to the mortals and it is doubtful to follow fashion in chess. There will always be players (mainly youngsters) being booked up by the latest novelties. So you do risk sometimes to play against a fashionista but don't panic as openings very rarely decide a game between amateurs.

Brabo

Tuesday, April 17, 2018

Which openings do my opponents play?

The answer which I gave at a reaction upon my previous article (see Dutch version of this blog) wasn't complete.  I wrote that the number of openings even with a very narrow repertoire is gigantic. But gigantic is something intangible. Can't we define a more concrete number? Well I'll give it a try with this article.

However before we start the number-crunching, we should first agree about the definition of what is an opening. I consulted multiple sources and discovered there exists no consensus about this. The summary by ECO (Encyclopedia of chess openings) is the only standard existing today but the system only categorizes the openings in 5 main-categories and 500 sub-categories. So ECO classifies openings but doesn't tell us what exactly is an opening or how many moves counts an opening.

In fact it doesn't really matter for this article as we are only interested in how far somebody can be booked up. In other words how many moves can a player memorize of an opening. Of course this depends of the player and the opening itself. I know quite some lines beyond move 20 see e.g. mistakes and copycats. Beginners often don't know more than just a couple of moves. World-class-players on the other hand sometimes know lines as deep as move 40. Still a high rating is not a guarantee of knowing much about an opening see surprises part 1. In short we don't know what players know or don't know and that is only intensified by many players consciously hiding their opening-knowledge see secret.

So I am obliged to arbitrarily draw a line for the opening. As reference I use the default settings to create an openingbook in Chessbase. On my blog I showed countless examples of my openingbook see green movesto analyze using a computer part 3studying openings part 2using databases, ... but nowhere I explained which settings I used to create that book.
Default (standard) 20 moves are used with a deviation linked to ECO. This means we use averagely 20 moves for an opening-line. More moves are used for lines in openings defined by ECO as popular while less moves are used for lines in openings defined by ECO as secondary.

So 20 moves will be used in my research as the base to define an unique opening-line. That finally allows us to check the openings played by my opponents. My personal database of standard games counts today +800. Let us see how often the same 20 first moves are played in those games or complete games if the game lasted less than 20 moves. An hour of scrolling through the database was sufficient to extract the answer out of it. The result was stunning. Only twice I got exactly the same 20 moves of an earlier game on the board. Besides I am pretty sure that these 2 unique cases can be fully linked to very deliberate choices of my opponents. In one game the Dutch IM Edwin Van Haastert copied my lost game played a couple of weeks earlier against the Belgian IM Thibaut Maenhout. It was a full scale battle with many mistakes but the final one was made by my opponent when he missed a devilish trick.
The second unique situation appeared in a game of the club-championship of Deurne. I already once won in 2009 against the 1700 player Pascal Francois. In 2011 Pascal repeated the opening as the opening is theoretically healthy. I agreed as there exists an important difference between the evaluation of the engine and the practical chances in standard play.
When we met a third time in 2013, Pacal had learned from the losses. You can't select an opening just by looking at the evaluation of the engine so Pascal deviated much earlier with an interesting alternative.

I want to return to the original question of which openings my opponents play. In the meanwhile we have the answer. Every game a new opening is played except some very rare cases. So we can't predict which openings will appear in the future. Therefore it makes little sense to study the openings of the opponents.

You could even state that it makes no sense to study openings at all at our modest playing-level. That was insinuated in a reaction of the Unknown One see article of 2012. However this is a bridge too far for me. In many openings it is an undeniable advantage to know a number of moves. 20 moves can be a good arbitrary line to define an opening but it tells very little how useful studying openings can be see my recent article the (non-)sense of blitz.

As each opening has its own very specific characteristics, I can't assign one number to the number of moves which one should know to get an opening-edge. Besides this also depends on the knowledge of the opponent as only crucial additional information of the opening will lead to an advantage. Therefore the second part of this article will be covering the effect of studying openings using the scientific approach which I apply in my games. How fast does the opening-knowledge expand when only playing a very narrow repertoire? I have more than 20 years experience with this method, so I can definitely show some remarkable statistics about it. In all those years I played exclusively the same openings and only made adjustments to the repertoire when a line was broken. Below you can see the evolution of my opening-knowledge in the standard games where I was playing white.
There is little difference in the evolution of my opening-knowledge for the black games I played.
In about +400 white and + 400 black games we see an average deviation at move 8-9 compared to earlier played games. In less than 200 of the +800 games I deviated from an earlier game so where I introduced something new which I learned from earlier made mistakes. It is remarkable that we see barely any progress of the deviation during the last 15 years although I kept more or less the same level of activity. So a couple of hundred games is not enough to create depth in somebodies repertoire. I assume my strategy could work for an extrapolated number of games. We saw this in the the project Alpha zero for which 44 million training-games were played. Naturally no human will ever be able to play so many games.

Despite averagely we see a very early deviation in the games, still in a substantial % of games the deviation from earlier played games happens later. I made a graphic about this to illustrate the % of played games linked to the move-number where the deviation happened. Below you see first the graphic of my white-games.
Later deviations occur less frequently in my black repertoire. This has to do with the Dutch opening which I play. White has a large variety of interesting lines in this opening which permits much more early deviations (I mentioned this already in my article a Dutch gambit part 2).
From the graphics we can deduct that in 27% of my white-games there is a duplication of the first 10 moves from earlier played games. For the black games this is only 16%. For the first 15 opening-moves we see that the share of white-games already shrunk to 3% while for the black-games to 2%.

I conclude this long article. The percentages are small but not negligible. Studying the openings of your own games will bring some dividends later. We all have different ambitions and priorities so there exists no rule about which amount of study is reasonable or not. Anyway if you study only for gaining some rating-points then I fear you will get disappointed in the long run.

Brabo

Monday, April 2, 2018

Surprises part 2

Last year I played 38 standard games of a slow time-control. I won 23 of them, made 10 draws and lost only 5. That seems a fantastic result but most of the games were played against weaker opponents. Besides only 11 were rated by fide so in the end I only gained a couple of ratingpoints. Many years already I can only play some small local tournaments to maintain more or less my playing-level.

Losing 5 games out of 38 is not much but each loss is one too many for an ambitious player. On the other hand losses are the best lessons to improve. That was part of the critics I received at chesspub because of my article to analyze using a computer part 2. Technically I make some high quality analysis but it is not clear how I learn something from it. Each game is unique (except in rare cases) so in each game new mistakes are made.

Of course something will be learnt from analyzing carefully games but often more can be achieved by looking to the total picture. Is there a common denominator in the mistakes? In my 5 most recent losses I noticed 1 important aspect. In each of the games I got very early into problems. In 2 games I even didn't survive the opening. 1 was already covered in my article evolution. The other was my game against Dries Janssen of which I already showed the final part in my previous artice.

No doubt my too optimistic play was a major reason for the defeat. Both kings stayed in the center but my king appeared to be much more vulnerable. However even more important to me is the fact that my opponent had seen the critical position already in his preparation. He was aware of a similar mistake made in that position see the game Haroon Azizi - Anneli Damau played in 2003 which also was countered by the same refutation (see moves 9,10 and 11). So aside from the technical mistake we should definitely focus at the disturbing difference of foreknowledge between both players. I am not yet even taking into account the rating-difference.

It is not a lack of study from my side. Earlier in the same year I had lost another game in the same opening. Because of that I had made an in-depth study of all the current theory. Of 5.h3 there exists only 1 mastergame in the mega-database but still I hadn't forgotten to check that possibility. Unfortunately during the game I wasn't able to remember my notes from that particular line. There were hundreds of them and 6 months later you just forgot most of it. I expect nobody can fully remember such analysis including absolute worldclass-players. Besides my analysis of 5.h3 were very superficial as the line can't be considered critical for the evaluation of the opening.

No my mistake was of course being too predictable for my opponent. I stick too much to the scientific approach of playing chess so my opponents can easily prepare a very dangerous surprise. I did some research of my games to illustrate this more clear. I selected all my standard-games which I played against somebody that I had played already once before with the same color. For this type of games I could assume  my opponents knew in advance my personality and adapted their opening-strategy to maximize their winning-chances. After filtering, 146 of my 829 standard games remained. It proves once again what I already stated in my article  matches that our chess-world is small. From each of the 146 games I wrote down which person deviated first from our previous mutual game(s) and at which move.
If we summarize then we see immediately a clear distinction between myself and my opponents. In 128 of the 146 games my opponent deviated first or in other words they tried to surprise me. Only in 18 games I was the one innovating from an earlier played mutual game. We also see this difference in the move where the game deviates from earlier play. My opponents averagely deviate at move 4 while I only do averagely at move 10. There is of course a connection between number of games and the move when the deviation happens. The next question is if my lack of flexibility costs rating points. Well I was surprised to find out that this isn't really the case. Below you see my TPR in the same rating-categories of my previous article.
Every medal has 2 sides. My lack of surprises is compensated by more experience. I also expect many players rather deviate to avoid a preparation instead of having made themselves a very elaborated preparation. It is again my negative remark of players being rather lazy than tired. Only for the highest rating-category there are doubts. The TPR doesn't decrease dramatically but I don't have a good stomach-feeling about my games. Especially Flemish top-players knowing me for decades will try to profit by using their preparation to counter my very narrow repertoire. So in that sense I do fully agree with the reaction at my article password of the Belgian IM Steven Geirnaert.

My article at the end of last year killer novelties indirectly proofs this with an example from my game against the Belgian FM Matthias De Wachter. Mixing openings is a necessity to improve which can also be deducted from  the almost 5 year old article the list of strength. On the other hand I do want to warn the reader that changing openings is not the holy grail either. In another loss this season the cause was definitely partly related by choosing a line which I hadn't studied earlier. I didn't know the mainline and my deviation was refuted by my opponent. So you can't play any unknown opening successfully to avoid being too predictable.

Brabo

Thursday, March 22, 2018

Comebacks part 3

In 2 years time my son Hugo managed to eliminate his original material-handicap. We started with 23 points (1 pawn = 1 point) as odds (see strange material balances part 2). Today only a (heavy) time-handicap (18 minutes against 1 minute and 15 seconds) remains. I find this surprising as I expected to still give bigger odds because I am almost 1000 points stronger. My son Hugo has only moderate technical skills so this situation is very different compared to the handicap-match against the Dutch expert Jaap Amesz which I mentioned in my article swindles.

I warned in my article sacrificing that often we sacrifice too optimistically. Today I want to refine my view by adding that even a small material-advantage in most cases will be decisive. Except beginners still dropping pieces, we very rarely see players recovering from a material-disadvantage. This sounds contradictory to what I wrote in my article comebacks. In that article I talked about 11 out of 100 of my games a deficit of 3 points was caught up. In 7 games even a comeback of more than 6 points happened with 2 extreme cases of 29,67 and 32,06 points.

The reason is that a computer-evaluation very often strongly deviates from the material-position on the board. Their evaluation is the final result of the critical line after both sides made the best moves. This final-position can deviate a lot from the actual position in terms of material if we are dealing with tactics. Sometimes those variations are extremely complex, making the engine-evaluation losing its connection to the practical chances on the board see annotations. It is another example of what I already described in my article to analyze using a computer part 2. An engine helps you to define very quickly your errors but then you still need to add the right interpretation to those mistakes. Unsurprisingly this often goes wrong, creating frustrations and aversion from the engines.

The difference between an engine-evaluation and a material-balance is clearly shown when looking at the comebacks. To illustrate this properly I again used the same 100 games of my initial article about comebacks. However I didn't just count the pieces on the board to make my investigation. Only when there was no clear compensation, I took the material-difference into account. Gambits or theoretical drawn positions can't be considered as comebacks. Below we see first the summary of my opponents.









Next is the summary for myself.










Only 5 pawns were recovered by my opponents in a total of 36 comebacks. I made up 13 points spread over 38 comebacks. So in most cases a pawn down meant a loss. Beside the exceptions can be almost exclusively linked to special cases. Or the position is tactically very sharp so a higher probability exists to recover from a small material disadvantage see my article the einstellung effect. Or the material is already very reduced so a higher probability exists to reach a theoretical drawn-endgame see e.g. my article practical endgames.

We can also assume from above tables that rating barely has any effect at the material-balance. I experienced this end of last year in a very painful defeat. I was obliged to win the last round of the maneblussers-tournament to get the tournament-victory and with the 1900 rated player Dries Janssen as opponent my chances looked bright in advance. However in an obscure opening-line I totally misplayed my position and very early lost a piece without any compensation.
Despite white surely didn't play the most quickest win, never the result was put in question. The handicap was far too big to ever hope to comeback. Besides my opponent wasn't a beginner either. Resignation immediately after losing the piece was a valid choice but something I couldn't cope with at that time. Also don't forget that it was 18 years ago that I lost another standard-game against a -2000 rated player see my article White chooses in the opening a drawing-line.

So you better don't get behind or your name must be Magnus Carlsen. Anish Giri didn't pass the opportunity to write a funny tweet about it: The world champion Magnus Carlsen is now officially a full piece stronger than the rest of the mankind. In other words Magnus can win against anybody even with a heavy handicap of a full piece.
Exceptions proof the rule. Miracles happen sometimes at chess so you do wonder if maybe it makes sense to continue always till mate. The American grandmaster Grigory Serper showed some other miracles in his article Why you should never resign? but leaves it up to the reader to decide when to resign or not at all. Personally I think it is dubious and mainly a waste of time to continue 100 lost games till mate just to steal 1 extra half point.

Brabo

Monday, March 5, 2018

e4 e5 openingbook

25 years I already open my games exclusively with e4. Also as long I answer 1.e4 exclusively with e5. In other words I have quite some experience with the position 1.e4 e5. It is therefore no surprise that some players asked for my opinion of the recently published repertoire-book Playing 1e4 e5 a classical_repertoire by Nikolaos Ntirlis. However to be able to judge a book, you first need to get the  chance to read it. That is a problem as I don't buy any books about openings since 20 years. Fortunately my team-mate Daniel Sadkowski was so gentle to borrow me his book for a week which allowed me to check the content.
I know the author Nikolaos since 2009 when we were discussing furiously some analysis at chesspub forum lasting several months. Despite our initial disagreements things went afterwards much smoother between us. We both realized that we only were trying to search the truth of a position so no reason to feel emotionally bad about any critical analysis.

In 2009 Nikolaos was still unknown. Today he has built a solid reputation as opening-specialist. Playing 1.e4 e5 is already his 3rd opening-book and in the meanwhile his 4th about 1.d4 d5 is released see qualitychess. Besides any of those book was welcomed very positively. That is an extraordinary performance especially if you know that Nikolaos doesn't possess a fide-rating. By hard working, a lot of research and knowing how to maximally use the engines a lot can be achieved. By the way he also tests his analysis in correspondence chess with some decent success. This is enough introduction of the book. Time to evaluate the content.

Well I will be short. I warmly recommend this book to anybody interested playing this repertoire. The quality of the analysis is excellent and the light writing-style makes it very easy to read and digest. However for myself there was not much new. That is also to expect if you have 25 years of experience with the repertoire. This book would've been very useful for myself 15 years ago on my road to the FM-title. Meanwhile I made in most cases exactly the same conclusions as Nikolaos which is of course a compliment for the book.

Still I wouldn't write this article if I could not add some remarks to the book. So what follows next should be considered as an annex to the book and not an attack. The first thing I want to discuss is the chapter about the Belgrado-gambit. I noted Nikolaos claims a small advantage for black so more or less refuting the gambit. The review from the Australian grandmaster David Smerdon warns the readers that many refutations of so called black advantages are well hidden in the book. However this was not the case for this gambit. I have studied it several times in the past as happened last year with one of my students at the bjk. Never I was able to refute the gambit so I became curious if Nikolaos really has found something close to a refutation. Unfortunately after doing some research I had to conclude that Nikolaos had missed an important idea. I really would've loved to see that annoying gambit disappearing from practical standard play.
Of course this is just a detail as the gambit anyway remains a very small side-system in practice. Totally different is the Spanish Breyer-opening which is used as the terminal of the repertoire-book. The keystone of any Spanish repertoire is always a soft spot. Somehow you must make a choice of how you will fight against the critical mainline. Now about the Breyer-opening you can write a complete book. On the other hand I was a bit puzzled to see only 1 real chapter about this big opening. Besides I also wonder if this is really the best choice to make for such kind of book. I guess the target-audience is between 1800-2200 elo and the Breyer seems a risky choice for them.I demonstrated recently how risky it can be in the last round of Leuven. My opponent FM Arno Sterck also Belgian youth-champion - 18 was already lost just after the opening.

Nikolaos recommends a different line of the Breyer-opening. He very often refers in his analysis to the correspondence-games of the Russian Senior International Correspondence-master Igor Telepnev. However also from pure theoretical perspective I have my doubts about the selected system. There are some lines which are nowadays rather shaky.
Naturally I admit that a 1800-2200 player will rarely encounter such highly sophisticated analysis. On the other hand we should not ignore that maybe the Breyer-opening is too complex for the amateur. But which opening is better? I play the Spanish Chigorin but I can't recommend that although I obtain very acceptable results. The Berlin and the Marshall-gambit are surely theoretical more stable but these openings won't seduce the average player. So I also don't know what could be the perfect cornerstone against the Spanish. I need more experience with different systems to get a better view of each qualities. Only by experimenting you can find out which systems are enjoyable. Anyway don't get discouraged of this article and give the Breyeropening a chance.

Brabo

Wednesday, February 28, 2018

To analyze using a computer part 3

December was a dark month for Stockfish. First Stockfih was beaten by Alpha Zero see chess.com. Next Stockfish didn't manage to qualify for the TCEC-super-final of season 10 the first since years. If you would only look at those facts then many would decide to stop working with Stockfish.

However if we are investigating more closely the data then we see a very different picture. The beating by Alpha Zero was eventually only a difference of 64-36 which corresponds to about a gap of 100 elo see the fide-handbook. Besides there was a lot of critics on how Stockfish was operated. In my previous article I already explained how easily you can get a difference of a couple of hundred ratingpoints by just manipulating some parameters. This is not even taking into account the missing opening-book and the very limited hash-table which definitely also impacted the playing-strength of the engine. In short a match where Stockfish would get better conditions, could well reverse the final result but this will never be allowed of course by deepmind of Google.

Also the early elimination of Stockfish after stage 2 in season 10 was no disgrace at all. Contrary to previous years after stage 2 immediately the super-final was played. Consequently a lot depended on how the best engines performed against the weaker ones. Stockfish was the only engine not losing a game but finally ended a half point from the leaders. So the 2 finalists (Komodo en Houdini) never proofed their superiority. Personally I believe Stockfish is the strongest engine today available. After the final it won the TCEC-rapid and the newest release 9  (available since beginning of this month) leads at ccrl with 39 elo.

Naturally Chessbase doesn't report anything about the success of Stockfish on their website. The engine destroys their market. Even when a new top-engine like Houdini 6 is introduced then mainly negative comments are received. You really need to be stupid to pay 100 euro while a stronger engine is available for free. An open forum creates extra visitors but can also cause damage to your business. Personally I think such comments are inappropriate. First you get more than just an engine for the 100 euro. Also you can't expect that everybody wants to work for free. There is nothing wrong with trying to make a living from creating new interesting things.

Still if we only concentrate at the engine then we can wonder how necessary it is for ourselves to get an extra engine beside Stockfish. Do other engines have an added value for us? Well honestly I doubt it for 99,9% of the players. The quality of Stockfish's analysis suffice for any player till at least 2600 fide and maybe even higher. Only for some theoreticians like correspondence-players and the world-top in otb it becomes doubtful to only rely at Stockfish. I call it doubtful as today it is really not clear if an extra engine will still bring something extra. To support this statement, I made during the 2 weeks of the last Christmas holidays a special research-project.

In the 6th round of the last Open Leuven again like last summer at Gent see evolution, I didn't stand a chance against the Belgian IM Stefan Docx. Again I was surprised in the opening and was trailing the whole game. However this time the problem of the opening was more serious than last time. Despite many hours of analysis I was not able to repair the system at home. In the end I had to admit the opening was not fully correct so I should look for something else instead. However I am rather reluctant today to learn something completely new from scratch. I play few games so the work should be proportional. Stefan recommended the classical Dutch as that is the closest related to my repertoire. Just recently our current world-champion Magnus demonstrated the viability of this opening in a secret online blitz-game.
Nevertheless Magnus can probably play anything especially at blitz and still win. Even in that Magnus' game black's position was several moves pretty dubious. So I wasn't convinced yet to take up the classical Dutch in my repertoire. I needed to know more about the quality of the opening. Of course the first thing you need to check are recent articles/ books about this opening. The e-book The Killer Dutch, published in 2015 can probably considered today as the current best up to date theoretical work about this opening. The author is the English grandmaster Simon Williams. He is the biggest expert in the opening. Unfortunately from theoretical point of view this book comes short. Simon doesn't write for theoreticians but he explains the opening from a pure practical use in tournament-play. That is understandable as the practical player is his main reading-audience. So some critical lines are a little too easy categorized as harmless.

If no good theoretical references exist then the only thing which remains is to start your own research. However that is easier said than done. Even an opening like the classical Dutch contains today a myriad of variations. Below you see a screenshot of my current personal opening-book only built with games of the Megadatase in which at least one color has + 2300 fide.
That is +1700 games and we still need to add the correspondence-games and the engine-games which also could influence the evaluation of the different lines. In my article studying openings part 2 I already explained that 100 games often take about a week to digest. So I realized in advance that I had to change my working-methods to process +1000 games within an acceptable time-frame. The first adaptation was to prioritize the lines played the most often in practice instead of the lines recommended by the engine. Especially at a very early stage of the opening which is here the case, we often see that the engine plays inferior moves compared to a strong opening-book (see also the earlier reference to the match between Alpha Zero and Stockfish). It is necessary to analyze side-lines to support the main-lines but analyzing side-lines to detect the main-lines is mainly a waste of time.

A second important time-improvement without loss of quality was expected by not checking everything anymore by a 2nd engine (see my old article to analyze using a computer part 1). It is very time-consuming to switch between engines even if 2 computers are used. That is the link to my introduction in which I announced a special research-project. Instead of checking everything twice, I only did for some positions which I considered critical for the evaluation of the opening. This means only the positions where best play of both sides could/ would still give an edge to white by the first engine.

In the end only 18 positions remained to be checked by a second engine. The result was stunning. Only for 3 positions there was a conflict due to the smallest possible difference of only 1 hundredth of a pawn. I discussed this absurd phenomenon in my article annotations. I like to use strict boundaries to achieve a very objective method of analyzing but in some exceptional cases this can create some weird evaluations. In other words the only conclusion of the project is that the extra analysis with the 2nd engine was not adding anything substantial.

It again demonstrates how drastic the engines have evolved in the last years. 10 years ago it was rare that 2 top-engines were so often agreeing. You could easily find positions in which one engine would tell you that white is winning while the other one would state black is winning. Extra analysis were necessary to find who was right or wrong. In some cases the truth was somewhere in middle. Today not only we see top-engines prefer the same move but also sometimes show the same main-line. In fact it is not so surprising as engines are getting closer to perfection. Besides Stockfich applies an open source strategy. Everybody is allowed to see their code and learn from it. Of course other top-engines copy stuff which again diminishes the differences between them. Before the old top-engines were more closed.

So this means we can skip the 2nd engine forever. No, that conclusion we can't make yet. It is not because we don't see differences for the classical Dutch that there exist none between the engines. The qualities of an engine are not depending on just one opening. It is also the most important reason why tests of engines happen with a wide range of openings. Recently I detected a serious difference of evaluation in a position popping up from a Spanish Breyer-opening. See below screenshot in which we see both engines calculating in parallel.


While Stockfish claims a big advantage, Komodo states it is approximately equal. So for some special positions some extra analysis remains necessary to know which engine to trust. Anyway the number of special position quickly diminishes.

Theoreticians will still have to use a second engine. For players not interested in maximum quality, it makes no sense anymore to buy a 2nd engine. I will use also less a 2nd engine in the future. Only for detecting small differences (0,3 pawn) I still see an added value. Today a 2nd engine has become redundant in positions with an evaluation higher than +2 or lower than -2.

Finally maybe some reader wonders what I concluded about the classical Dutch. A very concise summary of the weeks of analysis can be found below.

Brabo

Addendum 21 March 2018
At http://www.chesspub.com/cgi-bin/chess/YaBB.pl?num=1369191586/75 an important improvement was shown for black by the German FM Stefan Buecker in the classical Dutch. The novelty 13...a5 instead of the played 13...Bb7 revitalizes this line.

Tuesday, February 20, 2018

To analyze using a computer part 2

More than 2 decades ago I started to play chess. At that time a lot of scepticism existed about engines. Many avoided them as those things were considered not to play real chess and surely won't be able to teach you anything. This feeling was very widespread as even till a couple of years ago some older players still had never worked with any engine despite playing chess sometimes already for decades.

Such thoughts seem unimaginable for today's young generation used to work with many different chess-programs. Few will still deny that a computer can be very useful to learn chess. Besides we see everyday new programs popping up extending the current applications. Obviously this changed world has created the need for guidance. However as was already valid about 6 years ago see part 1, today still very few or no good literature can be found about this subject. Therefore I expect we will see an increasing demand for special info-sessions about how and which chess-software to use. Recently I gave presentations for LSV and Mechelen. The +60 years old Belgian FM Johan Goormachtigh has offered 5 courses in KGSRL (Gent) which will also include how to work with Chessbase. I guess his content will be limited to the basics considering the target-audience and his rather sporadic use of the tools.

So every ambitious player will today submit his games to the critical evaluations of an engine. Nowadays probably the DroidFish Chess app is the most used engine to get a quick verdict of a game. Almost 100% of the young players possess a smartphone so after their games they will use it to quickly get an assessment of the moves. Unfortunately most of them won't look at the game anymore at home.  2 months ago there was a discussion at chesspub about how useful such homework could still be. Some poster claimed that I am just losing my time by analyzing deeper my games and it would be better to spend that time to other chess-activities like reading chess-books, solving exercises,...

He is definitely not alone with this view. In the past more than one international master told me they were spending much less time at analyzing their played games. So maybe indeed there are better methods to improve at chess. On the other hand it is not a hard proof. Maybe those international masters would've been today (much) stronger if they would have spent more time at analyzing their games. Another argument is that the current top-engines are so strong that it makes no sense to  give an engine a lot of time to analyze the moves. I made a quick experiment to test this statement by running a full analysis using the Fritz 15 interface with below hyper-fast configuration and using the engine Komodo 11.
This means Komodo gets only 1 second per move. Next I use a threshold of 30 so if my move deviates 0,3 pawns from the best move then the engine will comment. Finally for the opening I use as reference-database a correspondence database. For the test I use my game against the Dutch IM Xander Wemmers which was already covered in my article secret as that game contains a number of mistakes. The amazing output is produced in just a couple of minutes by the machine.

I and Xander belong to the 1% best players but still at the rate of 1 second per move Komodo 11 could discover all the important mistakes while only running on a very moderate portable. It just shows how much stronger our best engines are today. Even with some very serious handicaps the devices still play much better than us. I made a small study to illustrate how much better this exactly is.
  • Estimated fide-rating of the current top-engine at the rate of 1 minute per move: 3200 (CCRL and SSDF talk about 3400 elo but I think the actual fide-elo can be easily a couple of hundreds lower.)
  • 1 year old top-engine: -52 elo (see my article rise of the machines part 2)
  • Extra ply : 66 elo (see citeseerx.ist.psu.edu)
  • Extra engine running in parallel: - 1 ply (some tests on my laptop)
  • Extra line running in parallel: - 1 ply (some tests on my laptop)
  • Halving the time: - 1 ply (see wikispaces.com/Depth)
Example: 3 year old engine, 2 engines running in parallel, 15 seconds per move, 3 lines running in parallel
  • Start-base : 3200 elo
  • 3 year old engine: - 156 elo
  • 2 engines running in parallel: - 66 elo
  • 15 seconds per move: - 132 elo
  • 3 lines running in parallel:  -132 elo
Analysis are happening at the strength of 2714 elo.

So even if you use an outdated engine in a crippled way then the output is still at the level of a super-grandmaster. Shouldn't that be sufficient for us non professionals? Well I think we should be careful. It does make sense to strive for the highest quality. An opening-analysis is something you want to reuse later so to avoid any rework it is better to do it properly. Besides even killer-novelties can occur on amateur-level.

Also we should admit that the output of an engine is often very difficult to interpret correctly (see above dump). It is one thing to know where the mistakes are made but to understand and correct them is sometimes very hard. Often a lot of extra analyzing is necessary to get a full diagnosis. Check my article the butterfly-effect in which I demonstrated how I discovered by making some deep analysis why a small change of the position creates a big change of the evaluation. A technique which I often use for this kind of work is to let the engine play against itself.

Finally I also believe that working daily with the top-engines will improve your own understanding of chess. You need a tool to detect the small positional errors as it is not enough to just correct the big errors to become a master. Also just looking continuously to strong moves will be very good for your own development. Any trainer will recommend their students to watch and study grandmastergames and engines play even better than any grandmaster. The same comment I make about the lomonosov tablebases. No human is capable to copy the accuracy of the tablebases. Still I do experience that just consulting regularly the tablebases has improved my intuition in many endgames. I can much better estimate which endgames have or do not have serious winning chances.

When I started to analyze my games in 1990 using the assistance of my very first table-computer Mephisto Europa A (playing-strength 1700 elo) I was obliged to spend a lot of time just to achieve a minimum level of accuracy. This need doesn't exist anymore today but I learned in the meanwhile there are other reasons to still analyze deeply your own games. In any case it is much worse for your own development to analyze too little than to analyze too much.

Brabo