Reflections on the ‘database’ generation
For competitive chess players of nearly every level of skill, pre-game opening preparation has become an indispensible part of our tournament routine. The opening phase of the game has been turned into a relatively stress-free exercise that can be leisurely ‘played’ at home or in his hotel room immediately before setting off for the tournament hall.
(NOTE: This article was originally published in 2008, but I periodically re-publish it because of its timeless nature.)
Furthermore, given the extensive and up to date databases available, this work can be completed with a high degree of predictability for any given adversary.
There is no doubt that easy and cheap access to chess information is changing the way we compete in today’s tournaments. Once upon a time we took it for granted that over the board (OTB) chess could be divided into three distinct phases: the opening, the middlegame and the endgame.
Today the OTB player finds that it is necessary to play only two phases (the middlegame and the endgame). Some might be inclined to look at this as a natural extension of an increasingly scientific and mathematical attitude towards the game, and they might be right.
There is also an inevitable psychological dimension to this situation, as the modern player’s confidence in his basic skill and ability is becoming increasingly associated with (not to say becoming dependent upon) his success in doing his ‘homework’. It can even be painful for him to imagine how his results might suffer should he suddenly be deprived of the opportunity to review his every opponents’ games, results and favourite opening schemes before actually having to show up at the board to play.
Some of the modern chess player’s most precious qualities -especially his self-reliance, his belief in his ability to succeed and his necessary faith in the basic principles of the game- are finding themselves having to be re-thought. Changes in the warrior’s arsenal necessarily require changes to his priorities, to his tactics and to his basic approach to tournament chess. He must adapt if he wants to remain a successful competitor.
Our ancient and noble game has gradually succumbed (over hundreds of years) to the same influences that have made progress possible in virtually every other aspect of our daily life: innovation, technology and information. While I personally have witnessed several profound changes in the way I compete in tournaments in just the past 20 years, past generations of chessplayers were not oblivious to their own changing circumstances either.
I quote an interesting perspective given by the great Emanuel Lasker from his 1925 classic ”Lasker’s Manual”:
Emanual Lasker held the world title for 28 years
‘For hundreds of years, Chess players had started their games in a happy go lucky fashion. After a few such chance moves, complications arose and in these complications skill and sagacity were displayed; they considered that the start of the game, compared to the importance of the hand to hand fight which ensued afterwards, was insignificant. Then one day some genius, now unknown to us, began to pay attention to the different ways of opening the game…from that day the problem of Openings becomes the point upon which attention has been centred and remains so, one may say , even to the present day (1925)
To visualise the beginning of this evolution we may surmise that at an ancient date, when players of original talent, whom today we would call ‘natural’ players, predominated over all others, some unknown genius , with a penchant for collecting information, made notes of the beginnings of good games, compiled them, classified them, and exhibited his work to a few friends. As a natural consequence, some of the industrious and intelligent learners would, in the first dozen moves, overcome superior players of that day, by employing the tactical manoeuvres gleaned from the manuscript of their compiler friend. One can imagine the surprise of spectators and the wrath of the defeated masters as they observed newcomers, without natural talent, waging a strong fight purely with the aid of a book of compiled information.
Their wrath evaporated of course, but the cause of it endured. Since those days we have continued to have compilers of ‘variations’ , players who fight according to the book, and those with natural talent who, however, can no longer climb to the summit.
Lasker: a man of vision
There is justification for the compiler. But can a player hope to become a master merely by studying a compilation?” (page 39, Lasker’s Manual)
Lasker raised several insightful and interesting points that are equally relevant today:
1) the Opening has increasingly become the centre of our attention and work in chess almost to the exclusion of everything else
2) industrious and intelligent learners can and do successfjully compete with stronger and more talented players because they ‘know’ the Openings better
3) chessplayers are becoming ”compilers of variations”
With reference to point 3, I prefer to use the term ‘database generation’ to describe today’s ‘compilers of variations’. Today the use of chess databases encompasses the function of compilation only too efficiently and accurately. It seems remarkable, but almost every young player today who starts to compete in chess tournaments learns to be comfortable with the use of databases even before he learns endgame fundamentals!
Change in the chess world over successive generations of players has been gradual, but relentless. What Lasker wrote back then was of course particularly true for 1925, but nevertheless many of his observations are still applicable in today’s world. Information is and always will be valuable and useful to intelligent and industrious individuals. The difference is that had databases existed in 1925, they would not have exceeded the compilation of more than 20,000 games, whereas today (2008) they would have exceeded 3,000,000 games!
More has changed than just the number of games compiled, however. If Lasker were alive today, rather than ask a question that might have had social significance in 1925 ( ”can a player hope to become a master merely by studying a compilation?”) , he probably would have looked at the situation from a different perspective and he might instead have asked ”What happens to our noble game when the masters themselves become expert compilers?”
It is curious that Lasker focused primarily on the issue of the effectiveness of this compiled information– not for the very best players (elite players)–but instead for those players who were likely less strong and less talented. I think that this insight shows us part of his genius. Probably Lasker felt that the elite players did not need to build a big reliance on compiled information, and that instead with their natural talent and skill they could figure out everything pretty much at the board. And therefore Lasker felt that this compiled information would be more attractive to players of less skill and talent, looking for a way to compete against stronger and more talented players.
Back in 1925 the number of masters probably numbered less than several hundreds and there simply was not such a pressing need for masters to become worried about amateurs playing their openings strongly. But today there are tens of thousands of masters and almost a thousand grandmasters. And there have been qualitative changes that have accompanied the gradual process of change. Today every master uses an up-to-date database, with state of the art computing engines (such as Fritz or Rybka) to assist him in working out the tactics of any variation of the opening.
The internet allows them to get the latest theoretical wrinkles that were played yesterday anywhere in the world. They can be so well prepared, that they can play the opening just as well as Kasparov himself or any other world champion. (Maybe even better!)! In fact, every master can expect (if he uses database preparation) to play the opening at a much higher level than he will play the middlegame and endgame! He can exceed his skill level in the opening phase by hundreds of points and will likely get out of the opening with a much better position than a player of similar level could have dreamed about 50 years ago when databases were unheard of! And all this without increasing one’s chess understanding. Technology is wonderful , isn’t it!?
When a grandmaster ‘Y’ sits down to play master ‘X’, he wants to play ‘X’ and not Garry Kasparov! Not that he would not mind playing such a great player as Kasparov, but there is a very simple reason for this: while GM ‘Y’ might gain rating points for drawing with Kasparov himself, he would surely lose rating points for drawing master ‘X’! The grandmaster wants to compete with the player sitting in front of him. And as in any fair competition, he has reasonable expectations that his superiority and talent over his opponent will be a decisive factor in the outcome of a one on one meet.
No one is left unaffected by technology
This , therefore, is the typical problem that the average grandmaster faces when he plays in mixed tournaments today. And this situation is complicated further by a characteristic of human nature: as we get older we form habits and get into routines. We can become fixed ‘targets’. The experienced and well known grandmaster will have thousands of games of his included in the database of his opponent which will indicate exactly what his opening preferences are with both colours.
His opponent will likely have nowhere near as many, and might even have an insufficient number of games for his grandmaster opponent to prepare correctly. Furthermore, while the grandmaster might have experimented with numerous openings throughout his long playing career, the database will also indicate what this grandmaster plays today, facilitating very much the opening preparation of his less experienced opponent by allowing his to limit his work to exactly the variation that his grandmaster prefers.
The inexperienced and less strong player (because he is computer savvy) will likely achieve the exact same position on the board in the tournament hall as he has in his hotel room. Not only this, he will have had the opportunity, at his leisure, to become quite familiar and comfortable with this position, having being able to access other grandmaster games from the same position, and also aided with the use of some playing program engine to help him understand the tactics of the position without ever having to do any independent work!
The information highway is forcing us to re-think how we play chess
I am talking from my own experience, of course, but also from simple observation of the tournament practice and results of my colleagues. I have seen young, unknown and relatively gifted players come up and quickly achieve results that have nothing to do with their real strengths. They very often have the extra advantage that stronger players don’t want to draw with them, and therefore have ‘draw odds’ — the grandmasters are more likely to take extra risks trying to win even positions.
These same young players , however, often soon become victims themselves when they start to have hundreds of their own games in the databases. Their opponents can see what they like to play, what they don’t like to play, and prepare accordingly. The stronger players are less likely to avoid drawing with them also. As a logical result, it is common for us to see young talented players come up, get great results and a high elo, only to see both the results and elo drop over the following years. However, the damage has already been done: the stronger and more experienced grandmasters have become rating point donors to players of less skill, experience and talent!
It is important for me to emphasize that this negative aspect of the effect of databases in tournament competition only has significance in mixed tournaments, where there are players of a variety of different strengths. The so-called ‘super tournament’ is left untouched, since the players are of similar rating, they don’t need to fear the draw and the preparation of each player is relatively simplified because his opponent is also experienced and well known (especially to his database!).
But for the rest of the chessplayers, there is a dilemma. How to effectively combat the high levels of opening preparation in today’s tournaments? Is there an effective way? Or are we players supposed to be resigned to the inevitable progress of our ever changing and seemingly less human world? Some people have been extreme and have suggested changing the rules of the game by shuffling the pieces before the start of each game. Others have suggested not letting anyone know the identity of their next round opponent until the round actually begins (eliminating pre game preparation all together).
Both suggestions are drastic, the shuffle chess idea in particular not to my liking just on the ‘principle’ that I like chess just as it is!
I have another idea. Rather than look at the changes (databases, playing programs, etc) as necessarily having an entirely negative influence , I prefer to consider them as challenges. Challenges to me to become a better competitor. I love challenges, and they force chessplayers to improve their methods and modenize their ideas to fit the spirit of the times that they live in.
I believe that the serious chessplayer must find a way to reduce the effectiveness of his opponent’s pre-game preparation. To do this, I feel it is necessary for the serious player to become less predictable with his choices of opening . If his opponent can no longer have, say for the sake of convenience, a 90% chance of getting the position that he has in his hotel room and instead only has a 25% of getting the position that he wants, then that is already big progress! This means that the opponent will have spent less time getting comfortable in his hotel room.
But in order to do this effectively, the chessplayer must increase the number of openings that he is willing to play. That is, the modern tournament competitor must start to play 2 or 3 times the number of different openings that he currently finds in his opening repertoire. You must become a moving target! Yes, a lot of work, you say?. And you are right, but there is a bright side: you don’t need to know each opening perfectly, since your opponent will also be in the same situation! The game outcome will become more influenced by natural skill and talent!
Lasker wrote ”What is immobile must suffer violence. The light winged bird will easily escape the huge dragon, but the firmly rooted big tree must remain where it is and may have to give up its leaves, fruit, perhaps even its life.”
In any case, my suggestion is a practical solution to a real problem. I see no reason why it can not succeed. I recall what the great Misha Tal wrote of his pre-game preparation for the great Bent Larsen in the 1979 Montreal tournament:
How can you prepare against a Larsen?
”’Preparing for a game with Larsen is a matter which is either too complicated, or too simple. The Dane’s repertoire contains practically all opening systems, and one’s chances of guessing the variation are no better than in a lottery. Therefore at home it was decided to begin the game with the advance of the king’s pawn. At that the preparation came to an end…”
Tal realized that sometimes preparation was impossible
I think that the database generation can be beaten over the board.
Related Topic
Hello Kevin,
Thank you for this great article!
Your insights together with Lasker’s postulates has given me hope and encouraged me to return…
Cheers! Love your website!
Bong