Neoclassical Chess: Main article

Neoclassical Chess: a Bright Future for the Game

Download archive PDF

Gabriel F. Bobadilla

1      Summary

We introduce Neoclassical Chess, an evolution of the game of chess, and as we argue, essentially the only one possible that solves the problem of the excessive influence of the computer in the opening phase and, at the same time preserves the integrity and history of the game.

All rules are identical to the rules of chess. The only difference is that the game starts on the position after the first three full moves of a game taken at random, just before the start of play, from relevant recent chess master practice. Choosing the starting position in this manner and continuing as in classical chess constitutes the closest game to classical chess that eliminates the above mentioned problem, and reconciles the needed variability of the initial position with the preservation of the heritage and tradition of the classical game.

As different from chess960 (also called Fischer-Random chess), the history and tradition of classical chess is fully preserved, and furthermore, contemporaneous opening preferences and their evolution are naturally integrated into the new game. As a result of this slight modification to the way the game of chess is started, a huge impact is made on the study of the opening, with a potential substantial cascade of benefits for the chess world: substantial increase in the “over-the-board” (OTB) part of the game, which we believe is the core conceptual issue at stake, increased competitiveness at elite level and attendant decrease in the number of early draws, increased delight of chess amateurs and spectators, spread of the knowledge of the full history of the game to all players, a more varied and inexhaustible game for the dedicated amateurs, and predictable increased overall economic value of “investing” in a professional chess career.

Finally, we argue that the switch to neoclassical chess may well enhance the well-being of professional and elite players. As I will argue, while neoclassical chess is a valid alternative for the competitive chess of the future, both Neoclassical and classical chess can coexist harmoniously and reinforce each other, resulting in a significant enhancement for the chess world.


2      the current state of the game and world of chess

The enormous influence of powerful personal computers and readily available chess software on chess has brought changes so that many recognize that is now “a different game”. For instance, Vladimir Kramnik states:

“I don’t know whether computers are improving the style of play, I know they are changing it. Chess has become a different game, one could say that computers have changed the world of chess.”

All phases of the game have been affected, resulting on widespread access to material and a much improved availability of training resources and an improved general level of play. While on the middle-game and endgame study the general view is that this influence has been positive (even if some have expressed worries about the unfavourable impact on the creative aspects of chess), on the opening phase it has been generally considered at best a mixed blessing, or downright worrying, as we will show later from comments and opinions expressed by outspoken top players. Few are as expressive as that of Viswanathan Anand in an interview in 2013:

In a sense, computers have killed the opening phase. (…) So if anything can be done, it is to rebalance the game. That can only happen by concentrating on the middle and end game

The influence of the powerful personal computers has exacerbated the problem that had arisen with the profound study of the opening since the 70’s. This well-known problem that has been salient for several decades now has many ramifications, among them:

  1. excessive emphasis of preparation on the opening over the other two phases of the game
  2. excessive role of rote memorisation of opening variants, in particular computer-prepared lines

However, I believe that the core issue at the heart of many of the symptoms of this “computer malaise” is that real OTB (“over the board”) play begins later and later in the game at top-level chess (and generally at all strong levels), sometimes at more than 20 full moves. This is the real issue that affects the game and that overrides any complacent view of the state of chess. My own view is that the decline in OTB ratio (fraction of the game that is “out-of-book”, i.e. thought out or “discovered” by the players during actual playing time) is an unambiguously negative development. This view finds support in both the modern perspective of chess as an organized sport (see the history of chess in [Eales]), and also in diverse references on the satisfaction of the spectators and the players themselves.

2.1    crisis ? what crisis ?

Many leading players have not been shy about expounding their opinions about either the state of affairs of the opening phase or the effect of the computer compounding the problem. We have quoted Anand above, but among many others: Ljubomir Ljubojevic, Aleksandr Grischuk, Shakhriyah Mamedyarov, Hikaru Nakamura have expressed a variety of concerns. In some cases, this predicament may lead them to suggest chess960 (or Fischer-Random chess, which we will cover below) as a possible solution for the future. Indeed, a less severe state of matters than is the case today was already a motivation for the development of chess960.

It is remarkable, in comparison with other sports or disciplines, that many among the elite practitioners recognize such an unfavourable situation in the way the sport is conducted. The general tone seems to be that of resignation however: seemingly, there is little that can be done about it.

It has been mentioned that one of the bright spots of the state of affairs of the game is the successful extension of its introduction in children education, its expansion as a competitive sport for children and teenagers, and the robust, unabated amateur interest. The spread of children and teenage competitions has been an undoubted success of international and national chess federations.

So some may think that given that the problem we are describing affects only top-level chess or at most its professional levels, which is a small part in quantitative terms (less than a few percent) of the whole chess-playing community, it is not so serious. I am doubtful about this optimistic views. The success of chess is partly based on an emulation effect and on the quantity and quality of the following it provokes on chess spectators. If the top of the game loses lustre, this will be felt in time on the whole chess community and the future of the game itself. Besides, strong amateurs also encounter the problem of the opening phase in a less extreme version than top chess players.

Hence we are led to the title of the present work: it would seem to imply that the future of classical chess is anything but bright. This may seem to disregard the quantitative success in the spread of the sport, the number of players and events etc. However, I believe that concerns about the future of the game is what lies beneath the concerns by top players referred to above.

2.2    opening preparation


Opening preparation, to the extent that it becomes a pure rote memorisation exercise, is the least pleasant part for most players with few exceptions. It is possible that many players do find pleasurable or at least interesting many other aspects of opening preparation. The key question however is: would you, as a player, spend the time in the rote memorising part if you didn’t have to ? A similar part of preparation is the review of recent games in the opening lines of interest to the player (see [Kramnik]) and not even that is seen as particularly pleasant but as a needed highly “technical” routine. I am not stating that this applies to all time spent preparing openings. Also, one should not confuse competitive opening preparation with opening research, where time is spent with a serious purpose (often in long-term preparation, book writing).

Opening preparation has always been important, but now with the availability of databases it is simply essential in competitive chess, not only in top-level chess, and devours a substantial part of the time of preparation for specific tournaments and matches. In fact, the amount of time spent can easily override the time devoted to the pleasures of the game themselves. All in all, pre-game and pre-tournament preparation (i.e. just before specific games and events) has a share of the total preparation at top and professional levels even larger than it would otherwise appear.

Of course, top-level, ambitious players have the only possible success-oriented attitude: they do what they have to do. Discipline, a good work ethic, the tension of a confident competitive attitude and the help of cognitive dissonance to smooth out the inconvenience all make it possible. It is well possible that some players have ended up “liking” this process, a successful strategy of adapting because of necessity, in particular if they are endowed with an excellent memory and feel at an advantage. This does not change the issue at all. There are no reasonable bounds to the psychological self-conditioning that humans are capable of in the pursuit of success if they are gifted with enough ability, ambition and discipline.

One wonders how many of recent early chess retirements of luminaries of the game and abandonments of promising youths are the result of a burn-out specifically induced by the extremes of opening preparation in today’s competitive chess.

Has computer opening preparation turned the noble game into an opening exam? Perhaps so, at least to an important extent. What is the joy of a game that has been turned into an examination?

2.3    The state of the chess profession

This was examined by John Nunn in “Current State of the Chess World” in [Nunn]. Nothing much has changed in 10 years, only for the worse: tournament prizes have decreased in real terms, partly as a result of the economic crises. But also it seems to be a long-term trend, due to the difficulty of attracting sponsorship and making World Championship matches as exciting as in the Kasparov-Karpov era. The dominance of the best computers over humans, the high percentage of draws at top play, and the effect of excessive “by the book” play, are all plausible negatives for attracting interest in top competitions.

The largest issue, I believe, is that in chess the risk of a comparatively short career has come to be accepted, so that the “value” of the prospective investment in a chess career has decreased. Computer opening preparation facilitates artificial upsets by weaker opponents, and that is a disincentive to long-term investing in a chess career for talented players. It is also a pity that this fact blunts the potential of chess as a lifelong rewarding hobby, as the experience and broad understanding of strong senior players counts less and less as a countervailing power versus the ability to memorise the opening.

2.4    the state of the game of chess

The introduction of the computer has had as a side-effect a much higher predictability of the opening. This in turn has given rise to increased evasive maneuvering. Excessive avoidance of prepared variations happens not only at top-level games, but surprisingly also between master and strong club players. The knowledge of other players’ repertoire through databases encourage opening diversification, but this runs into the very high maintenance costs of extending the opening repertoire.

It is often the case that some lifelong chess careers are spent playing a few openings. The professional wonders: when will I have a few free months to prepare other openings? This degree of specialization may or may be not a concern for many. For the author, the concern is that in their development, players may become trapped in their initial opening choices and not achieve their full potential (a “path-dependence” phenomenon). Experienced and well-rounded elite players, given enough career time and dedication, will end up finding “their” style (and even evolving it) and their optimal repertoire, but even strong professional players may experience this problem given the difficulty of obtaining “free” time to learn new openings under the pressures of career and rating progression.

Hence this issue results on an increased narrowness in style and chess knowledge, and encourages artificial upsets (surprises), i.e. lesser players winning over better ones, and the possibility of sustaining a rating above one’s level of ability with a very controlled preparation and narrow style. This was always the case to some extent in chess, but the new computer opening issue has carried it to extremes. If one contemplates the history of chess (for example [Eales]), the progression from elegant past-time to sport already brought some initial subtle change: was it fair to study too much in preparation for this then-gentlemanly activity, instead of trusting one’s mental powers to improvise over the board ? Then the transition to organized sport brought with it, as in any other case, the acceptance of intense preparation. Is not the most recent transition towards even deeper levels of studiousness, and to the extent that is based on rote opening memorisation, a final, excessive stage which has come to negate, at least partly, the purpose of the game ?

2.5    An economist’s perspective

Here I add a perspective founded on the economic concept of utility, which does not limit itself to monetary value but it is much broader: in this perspective, the enjoyment of the players and the delight of the chess public are as good as money, while unnecessary boredom, unpleasantness and stress count as a negative. No imposition of values or preferences is made.

It is interesting that with this perspective, if you witness a rapid-play tournament by adults or (i.e. assuming non-addictive behavior, where play is enjoyed with measure and self-control) there is no waste in this activity. People enjoy their time, thus reflecting (“revealing”) their preferences.

By contrast, if you see professionals waking up early in the morning for an unpleasant but needed memory refresher of some opening lines before the tournament’s game, the economist may well think: what a waste. Both players dislike it and there is no purpose to it. They would be better off sleeping or taking a walk. Cannot they just allow the opening books to be open, up to a certain move, besides the board when playing?

So the paradox in this case is that the “serious” activity (not the lighter one of rapid play) is the wasteful one.

Seen from that perspective, the state of top-level chess is a landscape of waste and lost opportunity. Time that could be better or more pleasurably spent elsewhere is devoted to computer-prepared opening memorisation. As you read this, an army of hundreds of talented individuals engage in overusing a skill (rote memory) and dilapidating their time unpleasantly on it, when it is not a pleasurable or efficient human skill and its value has been dramatically eroded in the rest of professional adult life. Great players devote their life to the game and produce OTB ratios in many top-level games below 50% of the game, and sometimes much lower, when OTB is what the spectators of the game want to see and, ultimately, sponsors and the general public will pay for.

Note that the chess world itself is not that impressed with memory recitations. People just prefer OTB play. The players themselves do feel a major uplifting for finding great moves OTB. Of course they think it is fine to win on a preparation, and when the stakes are high, winning (or not losing) is what counts. However, a computer-prepared win is clearly less satisfying for all involved: winner, loser and spectators, be it on a world championship match or a club game. Also, it negates the sports and artistic dimension of chess, inevitably based on lived performance, which is what makes it exciting to be a spectator.

2.6    The Chilling Moment

Adjournments have been eliminated purely as a result of the availability of very strong computer engines to analyse any position.

Then one day I wondered: is not the initial classical position of chess just another adjourned position?. The Big One. Caissa left Black’s move, a pass, on an open envelope. Now it is White’s turn to come to the board and make its move, in full knowledge of Black’s one, with the benefit of supercomputers to help White in analysing to death all possible beginnings of the game after “resumption”.

True, the initial position it’s a much more “open” position and less compromised than any of the usual ones that occurred in adjournments of the past, but as differently from those, we have had gigantic human brainpower thinking and conferring about it for decades and supercomputers running on it, too.

If the initial position is just another adjourned position, how should chess be prevented from going the way of adjournments?

2.7    fischer-random (Chess960) and other variants

2.7.1     Introduction

The only proposed modification of chess originated by similar concerns as those I have described that has gained some acceptance is Fischer-Random chess (now called chess960), which we will abbreviate as “FR”. While some think that FR’s time will come someday, it is now apparent that the fact that FR basically does away with the history of chess and its games and all known opening theory has been an insurmountable issue for most in the chess community. The argument that, after all, the initial classical position is 1 of the 960 possible in FR, so that somehow the history of the game as we know it will be part of the future of chess in FR is totally disingenuous. One position in 960, or the full history of classical chess being consigned to be 0.1% of the future of chess is consigning the history of the game to irrelevance. See the cogent critique in [Shiller] on why FR is not the future of chess, and also in [Mindsports]. Of course, if one likes FR for itself (in this case, the idea of shuffling the pieces in the first row), it is in abstract as fully respectable as any other game, and can claim to be one of the most, if not the most, similar to chess known nowadays.

2.7.2     A radical review of Fischer-Random chess (chess960)

The prestige of past world champion Robert Fischer has helped FR retain the aura of the ideal but impractical “solution”. I believe a rethink is in order. FR makes an all-out attack on the opening phase, intending to destroy all its value, because as was claimed originally, the study of the opening “is based only on memorisation”. It is true that is based on it more than other phases of the game are, but I would add that it is based both in rote memorisation and also in “knowledge” which we would define as the content that can be stored in long-term memory because it relies on patterns and/or techniques. Now, it is important to realise that there is also knowledge in the middle game and endgame: also there are familiar structures, patterns and techniques and the role of long-term memory is also essential in accumulating knowledge about these two phases.

Hence this negative discrimination of the opening phase, singling out its “sinful” support on whole “memory” is misguided and unwarranted in my view, and I believe, in the practical, perhaps unexpressed views of the majority of the chess community. What is different and worrying about the opening phase is its excessive reliance on rote memorisation and the attendant, deleterious effect on how much of the game is played “Over the Board”.

Summing up, I think that the crucial issue is that FR’s goal of completely destroying the value of opening study is highly questionable. Somehow this has not been questioned enough. Restoring the balance and making the study of opening theory mostly based on the build-up of knowledge, hence on an equal footing with the other phases of the game, is the goal we propose and one which fits in a completely natural and consistent way with the game as a whole.

The main issue with FR is: as we know, FR intends to obliterate the value of all opening preparation. For that purpose, K=100 is as good as K=960 (as once stated by Grischuk, and I can only agree), i.e. if FR consisted approx. of 100 initial positions instead of 960 it would also achieve its objective. As there are 960 initial positions, the value of any specific memorised opening preparation (for example, a line prepared by a classical chess player) decreases with the probability of attaining such position, i.e. is divided by K=960.

However, as argued above, FR’s objective is not the better one. The important consequence of all of this is that much lower K’s, in the range of 15-30 are adequate to obtain that “rebalancing” goal. The issue is avoiding the narrowness and predictability of the beginning of the game when two specific players meet. The issue is: will you as a competitive player, find it worthwhile to do a memory refresh just before a game or match (or the day before, or the month before depending on advance time) with a particular opponent, if instead of a fair chance of encountering a particularly sharp line, the odds are much lower?

It is true that FR is the only serious attempt (variant) of chess that is similar enough to the game to have been a contender as an alternative to classical chess. However, it is by now apparent that it is not at all viewed as a potential replacement of chess by the vast majority of chess players. Most people just do not feel that it fulfils the place of chess. Other “wild” chess variants impress even less as substitutes for a majority of chess players, and as regarded as excursions of the imagination, not in the least tempting as a replacement.

We conclude this examination of chess variants with the observation that no variant of chess ever invented has appeared to be close enough to the game of chess itself in the view of the chess community. This empirical fact has been a fundamental source of inspiration for this author when formulating the problem as done below in 3.1.

2.8    future developments of the problem

The problem herein described will only worsen in the future. Current technology already supports expert systems that for a decreasing price will tear apart your game collection and produce a report on all the holes of your opening preparation and a collection of relevant past games and computer lines to exploit them. We can even easily imagine memorisation sessions before the game à la “Clockwork Orange”, where a player attends before the game a well prepared, exhaustive pre-packaged movie of variations based on his/her opponents’ vulnerable lines. While my description is intendedly creepy, technologically this does not entail any innovation.

So, even worse, all this will be done in the future not by hardworking computer-aided preparation from the player by himself or herself, this can and will be outsourced. It is a question of access to the resources, which will only increase “demoralisation” of chess. It is not very uplifting for anyone winning or, even worse, losing in this way. Against this, the chess community seems to be destined to helplessly contemplate the slow decay of interest in top-level chess from chess spectators and the outside world.

We will now endeavour to show that it is possible to solve the key problem while fully preserving the history of the game as we know it.

3      the proposal: neoclassical chess

I propose a new way of playing chess in order to address all the previous problems. I will argue that it is not a mere variant of chess but a full-fledged evolution. It aspires to be an alternative to the game as we currently know it for competitive chess (more precisely, any level of chess in which intensive, opponent-specific opening preparation counts), but also for amateurs of all levels for whom chess is a durable, not casual, undertaking. However, as I will also describe, both Neoclassical and Classical Chess can coexist harmoniously. We use indifferently “Neoclassical chess” or “Nc chess”. In the current paper, I will avoid all formalities and will present intuitively the formulation and conclusions. For a more precise description, or for readers who like statistics or mathematics, please see the companion paper “Foundations of Neoclassical Chess” (which we abbreviate as [FNc]).

3.1    problem formulation

We will try do a first-principles rethink of the issue. We assume that the game of classical chess is the absolute ideal but that as mentioned, the last decades of opening study and later the prevalence of computer-aided opening preparation have compromised in practice its interest as a human competitive activity.

Hence we pose the following problem:

What is the minimal modification of the game of chess that eliminates the problem described in the previous chapter, i.e. diminishes the value of opening preparation based not on understanding but on rote opening memorisation, while preserving both the history and traditions of the game and providing for its future evolution, so that it is acceptable to a large majority of the chess world?

We are not concerned with being original, smart or imaginative in the design of the game. We are not concerned with producing a game where humans have a better chance at beating the computer. All that we consider a distraction. We want to know which is the game most similar, hopefully almost identical, to chess that manages to get rid of the “computer malaise” in the opening.

Thus, this is what we want of the new game:

  1. It is equal to classical chess except in the opening, and with the same flavour throughout.
  1. It is a game of pure skill, not chance. This is exactly as in classical chess, and is essential to the identity and history of the game, see for example the history of chess [Eales].
  1. It integrates the history of the classical game. It must be possible to play a large majority of relevant classical games of the past as a Nc chess game.
  2. It fully preserves the history of the classical game. All the future games of Nc chess must have been possible at any time during the past of the game.
  1. It reflects the contemporaneous and future opening preferences of the human classical game. Neoclassical chess must incorporate the collective opening preferences of players of a sufficiently high skill level.
  1. Universal, easy setup process and start of play (practical condition): The setup of Nc chess should not be substantially more complex than that of chess.

Note that we explicitly want to modify chess as little as possible. We want 4 and 5 because they embody the objective of preserving the history of chess and providing for its future, hence achieving a true evolution of classical chess, and not a mere variant.

But additionally, Neoclassical Chess must fulfil the following condition: sufficient randomness of the set of beginning positions of the new game. That is what will allow us to solve the problem.

3.2    problem solution: neoclassical chess

We define Neoclassical chess of depth D as the following way of playing

A game of chess where the players start the game from the position obtained after D correct full moves (D by White and D by Black), which is taken from a game chosen at random (just before the game) from a database of all relevant contemporary human master games.

This is a general solution that satisfies all we have asked for in the conditions above. This is verified in the companion paper [FNc]. In particular, the important two new conditions 4 and 5 are now satisfied: the ones that enable Neoclassical chess to be an evolution of chess.

Furthermore, we have studied which must be the smallest parameter D that achieves sufficient randomness of the beginning of the game. Then we would be as close to classical chess as possible and would manage to solve the problem: the game which is most similar to chess that rebalances the weight of the opening and bases it on long-term accumulation of knowledge, not on rote memorisation.

This question would seem mostly empirical, but the way humans of sufficient skill play openings has unusual stable statistical properties. After studying this question in the companion paper [FNc], we have found the depth of 3 (starting in the 4th move by White) to be the optimal solution. Depth of 4 moves is not a bad game, but unnecessarily conditioning and farther apart from classical chess.

Hence only 3 moves deep into the randomness of human master play a game very similar to chess emerges: Neoclassical chess.

The paradox of increasing depth in Nc chess is that when the number of possible positions increases sufficiently, the value of memorising preparations for those positions decreases rapidly, as in chess960. The surprise is that at a very low depth (3 moves), the needed rebalancing is obtained: the character and freedom of the classical game is retained, and the way the opening is studied changes radically: we believe that for the better.

A crucial discovery of our investigation is the fortunate[1] fact that strong performance objectives are reached for very low depths. The initial positions of neoclassical chess are often those that define the great openings of today (Spanish, Nimzoindian), or in other cases one or two moves further down the road (as in the French and Sicilian), so that none of these positions are excessively forcing or conditioning on a player’s style, nor they make the player find him/herself playing somebody else’s plan or tactical idea. This is important for the practical quality and attractiveness of neoclassical chess as a solution.

We have also found, and describe in the companion paper [FNc], an important conclusion: Neoclassical chess is “essentially” (i.e. except minor and more complex variations) the unique solution that verifies all the conditions we want.

For mathematicians, uniqueness has prime theoretical importance. For us and the chess community the uniqueness of Nc chess as the solution is crucial. Knowing that it is the only solution facilitates to focus: it means it is the only potentially satisfying alternative to classical chess for chess players.

3.3    key characteristics of solution

One of the most important properties of Neoclassical chess is that when faced with a prospective game with a certain opponent, not much lengthy specific opening preparation is useful or even possible. The only preparation possible is being aware of the opponent opening choices in classical chess and acting accordingly, similar to getting to know the general style of the opponent.

This has the most important consequence: a significant increase in OTB ratios of play as long-line, player-specific preparations are useless.

Why is this so? The expected (average) number of games to be played before a certain opening position appears is the reciprocal of its probability in the opening distribution. Given the choice of depth, for a normal and quite frequent opening, probabilities of 4% mean that 25 games are to be expected to happen with the given colour (or really 50) before it is actually encountered. And as they are independent experiments, even if 21 games pass by and the given opening does not arise, the next occurrence is still 25 games away with those colours, not 4!. However, for a professional or a long-time amateur, those positions are bound to appear sooner or later, so it pays to have a good (or minimum) understanding of plans, pawn structures, themes and early tactical ideas in the lines.

Note that memorisation of long opening lines fights against two winds: the normal memory decay and the effect of continuous chess opening innovation. Also, long-term memory is not boundless: it is now known to “occupy space” in our brain. This forces players to be ruthlessly efficient and learn what they will be able to remember months (or years) after the moment of study. In particular, contrived long lines turn out to be a waste of time. Even if you manage to learn all the possible “openings” (positions from move 3), by the time you would expect them to appear, you will have already forgotten most of what you prepared anyway.

Another consequence is the need for opening preparation to be a broad and shallow endeavour. This induces an increasing richness of knowledge of chess for all players.

Note that the price “we” (the chess community) pay when switching from classical to Nc chess is that we surrender full control of opening choice and the symmetry of the initial position, while crucially all initial Nc chess positions are “balanced” i.e. it is a game of skill, not chance !.

We are also surrendering opening trap preparation. An important question to my mind is: is it so different to prepare an opening ambush with the computer from using supercomputers to analyse an adjourned position?

Simply, I claim that it must be realised that opening “ambushes” and theoretical opening duels are a feature of the human game that have become as pointless in the super-personal-computer era as adjournments have. Few miss now adjournments, and in time I believe few will miss the battle of computer-aided opening ambushes, as opposed to opening study and research per se.

Some may be concerned about the issue of perfection of opening play and opening “progress”. But look at what happens in the endgame with current time controls, or at how engines point out instantly in real time to blunders by the greatest players. Imperfection is already accepted in the human game as an essential characteristic, with very limited time for the complexity of many positions. There seems to be no reason not to accept it as part of the game at its early stage.

Some may be bothered by not starting real play from the initial position. But I believe that most master game positions at move 3 have, after years of collective experience, a very respectful standing. In some cases, they are of such monumental standing that it would seem no lesser a standing than that of the initial position of the pieces. What is odd about starting the game from the position defining the Nimzoindian defense after three moves?

My overall assessment of Neoclassical chess is the following:

We set out to solve a big problem, the computer malaise in the opening phase with all its attendant consequences (remember Anand’s dictum: “the computer has killed the opening phase”). In order to solve it, there would always have to be a price to pay, simply because I have, in our problem formulation, thought of chess as the ideal game. It turns out that not only the problem has been solved for what I argue is a reasonably small price, but that an unexpected bonanza has been found to be thrown in “for free”: the legacy and history of the classical game not only has been left untouched by the solution, but in fact has been substantially enhanced as Nc chess will produce generations of universal chess players. These Nc players, at their more modest level, will be as universal as many of the recent great champions, who play universal opening repertoires and structures, know many of the great chess games of all previous world champions and top contenders etc. Other extra benefits may be a likely decrease in the amount of draws and more creative OTB play in the opening.

It is possible that all this seems a distant concern to amateurs for whom opponent-specific opening preparation is not too relevant. If they are not casual amateurs and will be playing for a long time, they may end anyway playing a very limited set of openings, missing a big part of chess, i.e. instead of travelling the whole world of chess in their lives they miss entire continents (for example, closed positions or as the case may be). They will never know that they could also be good playing other seemingly unfamiliar positions. For them, Nc chess offers an inexhaustible game, as it gently forces them to explore unfamiliar terrain.

For advanced players, the collective switch to Nc chess is similar to the following: in the face of multilateral nuclear proliferation, we find a way to force instantaneous multilateral nuclear disarmament.

In fact, a comment by chess commentator Mig Greengard [Greengard] in 2002 is faintly suggestive of our proposal. Neoclassical chess is less painful though:

“Maybe 20 years from now someone will suggest smacking each player on the head before each game so they can’t remember their preparation…”

The idea behind the Neoclassical chess proposal combines the combinatory explosion which is at the heart of the infinite variety of our game with the very human characteristic of limited memory to bring about the uselessness of “memorising without understanding” long opening lines.

4      neoclassical chess matters

4.1    opening preparation in Nc

When preparing the opening in Nc chess, initially a little time may have to be devoted to each major opening positions, and then to the least frequent ones (a risk minimisation strategy). As a reasonable initial level is attained in most of the position, progress may happen slowly and more focus may be given again to the most usual openings.

The process of opening preparation in Nc chess is similar to applying several layers of lacquered paint. Organic growth of opening knowledge occurs as a slow accumulation through the years. Preparation is a long term endeavour, aided through experience at the board and by studying classic games in any opening.

It also encourages breadth from the beginning. In Nc chess opening prep becomes high-quality prep, it concentrates on the highest values: ideas behind the opening, plans, pawn structures, contours of the possible middle game. In summary, knowledge and chess wisdom instead of rote memorisation. This preparation is more pleasant, has less maintenance costs, and makes you a rounded, better player both in Nc and classical chess.

In terms of lifecycle time studying the opening, Nc chess imposes the need of an initial, broad-based preparation for classical players with narrow repertoires. Once that initial effort is made, one learns “on the job”, free from specific game preparation. You also learn when you study annotated games! But then similar or smaller time is needed in order to maintain a broad-based repertoire, with the advantages of: no or little specific prep for events and opponents (for example, no review morning sessions), much less fraction of waste (forgotten prep) during the years, and prep that is also middle-game prep!.

4.2    The annotated Nc chess game

It seems the same as a classical game of chess, except you will see the first 3 full moves in italics or bold. Then you will know that it is a Nc game, and that OTB will begin much earlier (in high-level games). A gigantic change of incentives that the players have faced in opening preparation is summarised in those italics.

4.3    how players relative strength is affected

Nc chess manages to reward both talent and experience. It also rewards self-confidence in “over the board” powers… hence strong classical players tend to remain strong. Nc chess does not alter at all the major chess hierarchies, as wide differences in skill as measured by results (i.e. ratings) will persist in Nc chess. Top players at classical chess will remain top at Nc chess, with very few exceptions.

The only difference is that very narrow players in terms of opening preparation will remain in their broad category of strength but will initially lose relatively to “broad” (universal) players. In a dynamic world, they will also adjust and part of that relative loss will be mitigated.

Professional players will tend to be favoured versus casual or young players that base their strength in very narrow, tightly controlled repertoires.

Clearly players that play varied openings without much opening preparation will be ideally attracted to Nc chess.

Note that for strong seniors including retired masters, Nc chess is a gift for life. The current predicament is unpleasant for senior players: you may have been a strong player at your level, but without current opening knowledge you may lose a lot of points against well-prepared players and get demotivated. Also Nc chess rewards past broad study of the game for top players in their senior years, and for strong experienced amateurs it will help to keep their motivation, so that the personal “life investment” in their favorite hobby (or passion) keeps yielding enjoyment.

The Nc chess master is truly a master for life.

Young while less experienced players will be fine: in the whole of their life, their time “invested” in chess will be worth more.

4.4    Neoclassical vs Classical chess

I would like to suggest that given the increase in OTB play, Nc chess can only result in more pleasure, inspiration and excitement for the chess world. Also, it would seem that more OTB and less narrowly prepared openings would lead to less draws. At some point that may well find its way in more sponsorship interest. It would seem that monetary value can only increase. It is difficult to suggest to what extent, but it seems that there can be no downside.

Also, it is reasonable to believe that total prize money available will increase: there may be some crossing over, but 2 robust versions enlarge the pot and reinforce mutual interest.

If a robust introduction of Neoclassical chess can only add economic value to the chess world, I think the case for an increase in enjoyment and delight for players and aficionados is even better.

We also suggest that Nc chess brings about an overall better life for the pro players: more pleasure in preparation, less stress and effort, more time to devote to other phases of the game or simply to personal life. If they also play classical, they become better players at it. High-level chess is already a very psychologically taxing sport. I believe Nc chess may well bring about happier chessplayers in their chosen profession, either professional or sportsmen. It si not a leap of the imagination to envision significantly extended duration of professional chess careers. Long careers by choice, not necessity, will be more common. That also means an increased return to the uncertain investment on a chess career for young talent.

Note that as different from classical chess, Nc chess as an exam is comparable to an improvised foreign language oral exam: there is nothing you can do about it. The work has to be already done. Same with improvising a piano piece on demand from an extreme large repertoire: how do you prepare for that concert? The best exam is one that neither you nor anybody else can prepare for.

4.5    transition to Nc Chess

As we have argued, the fact that Nc chess is plausibly the one valid alternative to classical chess for those who like it, may probably give it staying power. However, initially some hurdles may appear in the way to acceptance. Some of them come from the very counterintuitive effect of both statistical illusions and the difficulty of counterfactual arguments.

A possible hurdle is some incredulity on the part of some players that they really do not need to master all the openings. However hard you may explain, the cautious learned attitude of risk aversion will make some shake their heads. They may say: so we wanted less opening theory and now we will have to know all of it!. Fortunately, Nc tournaments will be available that will not affect your classical rating. In fact, many chess students that have loved to review good annotated games even when they were worthless for their opening repertoire, or who have a long chess experience, will instantly feel that they already have a reasonably good grasp of many openings if they avoid sharp critical-lines, and will feel that in Nc chess they fall in inferior positions after the opening less frequently than in classical chess ones. After an appropriate time building their Nc repertoire and plugging the holes, they will be ready for full battle.

Also, luck may have an effect at the beginning of the introduction of Nc chess, if one of the players happens to be an expert on the obtained opening (and part of Nc chess preparation is learning how to deviate before an expert in each important opening). All that has little future as time erodes particular advantages and players adapt.

Another possible hurdle is that some players will fail to recognize that the same familiar initial moves that two top players encounter when playing Nc chess a couple of years after playing by choice these moves in classical chess, is a completely different start to the game. As they had no idea that this opening was going to appear in their game (even the most common opening has only around a 5% chance of appearing at depth 3), the sharp lines they prepared 2 years ago have partly been blunted in memory. Even if that is not the case, it will be easy for the player that feels less confident in a particular line to avoid it, as the possibilities of branching out of that particular line are so many, that it is just not possible for the opponent to have convincing superior preparation against most of them (note that the forgetting factor affects also myriad branches in the line). Both players are by now probably unaware of the latest novelties at this depth. Even the better prepared player, will feel natural risk aversion about embarking on a very complex line without fully remembering all the nuances. Hence they will play what they are able to substantiate OTB, after a much smaller number of moves. In only a few years, a lot of that very long preparation will be forgotten and future generations will thrive in understanding and acquiring long-term knowledge in most of the important openings. Note that what at first thought seemed to be the norm is a situation which is both transitory and unlikely.

Another hurdle is a strong acquired taste and familiarity with some openings, particularly if some players as White only play either 1.e4 or 1.d4. Well, the other side of the coin is that instead of being ground to death and boredom as Black by a cautious player in an unfamiliar opening, they may well be playing with Black an opening with much more active possibilities. I suspect that these preferences are often an acquired taste. But tastes are tastes and some players will choose to live only within the realms of classical chess, and take their weekly or monthly dose of opening memorisation.

In the transition period, it is to be expected that in Nc chess there will be a higher amount of wins and losses, later it will come down but if anything it is likely to stabilize at a higher rate of decisive games, without affecting much the total results or ratings for each player.

4.6    on the harmonious coexistence of neoclassical and classical chess

There will always be classical chess for many players that will still prefer it to Nc chess. I speculate that in the long-run selection pressure will favour Nc chess for competitive play (more enjoyment and money, less pain and boredom), even if the initial inertia and acquired tastes may well delay the process.

Classical chess provides Nc chess with the opening distribution. At the same time Nc chess facilitates the permanent existence of classical chess. Both are mutually reinforcing.

Perhaps in the future, the switch to Nc chess will be seen in retrospect as a minor tweak, not that different from the suppression of adjournments. If in fact, Nc is a success, it will also benefit classical.

But possibly the most common situation is that there will be many players who enjoy both types of chess: they will have both a deep (classical) and a broad (neoclassical) opening repertoire.

4.7    To Nc or Not to Nc, that is the question

An optimal opening strategy for future strong Nc players that are unused to playing classical chess and venture into one classical tournament is to imitate as much as possible, proportionally to their level of strength and preparation time, the opening preparation strategy of current World Champion Magnus Carlsen: a broad repertoire while trying to avoid any crisis in the opening. The aim would be to be left with an equal or perhaps only slightly better position with White, and an acceptable one with Black (perhaps an slight disadvantage, but with possibilities of active play), and then play it from there.

4.8    Training and Self-Diagnosis

Even for chess players mostly interested in classical chess, participating in Nc chess tournament is a great preparation opportunity without affecting rating. I think that Nc chess may be used to refresh and explore other openings without rating worries but in a fully motivating environment. For example, in the case of ambitious players that wonder if their progress has stagnated due to a narrow repertoire that constrains the type of positions to which they are exposed. Nc may also be a good tool for self-diagnosis in the hypercompetitive top-level chess world

The idea would be: Do you know your full potential ? Are you “trapped” in your opening repertoire ? Hence, know thyself as a chess player. You may well play Nc to find out. Go Nc and come back a better classical player

5      finale: a vision for the future


It is now several years after Neoclassical chess has become the accepted, more popular and standard way of playing competitive chess. The international chess community and all the players involved (candidates and current World Champion (WC)) have agreed that for the first time, the World Championship will be held in the neoclassical form, which has resulted in unprecedented sponsorship interest and excitement from the non-chess world. The first game of the World championship match is starting, a particularly exciting match given that the current WC may achieve, by winning, the feat of breaking the longest standing modern record of 15 years of the 13th WC.

The current WC is well into his thirties after more than a decade of sheer dominance of the chess world, while the aspirant, who will turn 20 this year and is believed, for the first time, to have roughly even chances of winning, would become the youngest WC ever.

As the players proceed to the table board and shake hands, the world contemplates the familiar scene of the checkered board, the chess pieces in the same initial position more than 500 years afterwards, its seemingly eternal quality that chessplayers have ended up perceiving as part of the natural order. The main arbiter approaches the table and hands both players two sealed envelopes which cointain, in figurine algebraic notation, the first few moves of today’s game, those of one game chosen at random from all standard-rate games in the last few years of grandmaster practice: a representative mainstream beginning of the old game.

Both players open their envelopes and peek into identical single sheets of paper, containing each the start to the game: three moves for each player. For a few seconds, only they and the main arbiter are privy to those initial moves, while the world of chess, through websites and internet TV, and the spectators that pack the playing hall, watch transfixed in expectation of the beginning.

In these few seconds, the players adjust mentally and embrace the initial nudge of the hand of destiny, the slightest hand that leaves them from now on sovereign as ever of their fate at the board.

The arbiter starts the chess clock and the players, as in any game of classical chess has always been done, begin to put out the initial moves, in sheer expectation of the watching world as the moves unfold. The third move of the WC moving his white King’s bishop signals that the opening of the first game of the match is going to be a Ruy López. Seeing White’s move, and while summoning all will and mental strength in preparing for the start of the battle, the aspirant is distracted for a split second and instinctively reaches out to the King’s Knight, indicating that the opening will be in fact a Berlin Defense, only to stop mid-air, in the briefest of pauses. While a barely audible shudder is felt from some spectators in the playing hall, the aspirant’s hand abruptly moves to the right-hand board corner, grabs the Queen’s Rook’s pawn, and confidently pushes it one square forward.

It will truly be a Spanish, after all.

Madrid, May 2015

6      references

[Greengard] Mig Greengard discussed Chess960 in “Mig on Chess, #177, (2002)

as quoted in:

[Kramnik] V. KRAMNIK and IAKOV DAMSKY, “My life and games”, p. 11. Everyman Chess (2000)

[Mindsports] Christian Freeling and Ed van Xon, : “Why do great players make poor inventors ?”

[Nunn] J. NUNN, “Grandmaster Chess Move by Move”, p. 270. Gambit Publications (2005)

[Shiller] ERIC SHILLER “Why Fischer-Random is not the future of chess”, web entry in

7      Acknowledgements

My wife Mercedes was a helpful sounding board for some of the core ideas in this paper while enduring the process of its production. My brother Jaime F. Bobadilla did detailed reviews of several versions of the manuscript, proposing improvements in both content and presentation and influencing my final thinking on the matters herein described.

8      About the Author


Gabriel F. Bobadilla[2]

He received a Master’s degree in Economics and Finance, a two-year full-time graduate program, from CEMFI (Centro de Estudios Monetarios y Financieros, Foundation of the Bank of Spain, Class of 1998), with 1st of class honors (“Premio Extraordinario”). He did his master’s thesis on stochastic models of the term structure of interest rates (the yield curve).

He also holds a Dr.- Ing. degree in Applied Mathematics from Universidad Politécnica de Madrid (U.P.M.), with a doctoral thesis on matrix models of linear dynamical systems, and a Master of Science in Electrical Engineering from Caltech (California Institute of Technology, Pasadena, USA, 1990), where he had the support of a Fulbright scholarship, studying robust control. He first graduated as an Industrial Engineer (U.P.M., Madrid, 1989), with specialization in control theory, electronics and computer science, achieving 1st of class honors and being awarded the distinction to the top nationwide graduate in his studies (“Primer Premio Nacional de Terminación de Estudios”). Since 2001 he holds the CFA (Chartered Financial Analyst) designation from the CFA Institute.

Gabriel F. Bobadilla works as an investment management professional for a Spanish asset management company and family office. He manages financial assets, with a special focus on alternatives.

[1] In truth, his is not a “lucky” accident: again the combinatorial explosion essential to the variety of the game, makes an increase in depth of one mere additional full move go a long way in terms of “performance”.

[2] The name of the author in its original Spanish form is Gabriel Fernández de Bobadilla


No Replies to "Neoclassical Chess: Main article"

    Got something to say?

    Some html is OK