Morphy, Fischer, Alpha Zero, and How the Adaptation of Chess Theory Has Blinded Us

Sort:
GrahaMasterBuckner

Over time chess strategy has changed. I am only 25 years old- I grew up in the post-Kasparov: human-trained machine chess era. A short disclaimer. I do not doubt myself that all below mentioned players would be outplayed by a machine of the present. Stockfish- even after Alpha Zero's loss- is great because it will never tactically blunder against a human (and its strategic vision is also more than adequate). Humans simply cannot play full 50+ move games of chess without inaccuracies as these silicon monsters will consistently do game-after-game-after-game. In this post I will present examples of how chess theory has changed and even folded-on itself over the decades and what it means about the world's best players of all time.
Firstly we begin with Morphy. For you skeptics, and it shall be a great many, I will start by quoting Fischer 
"‘Paul Morphy was a great chess player, a genius. I’m a little embarrassed when I think that I’ve got a [commemorative] stamp and I’m still even alive. As far as I know he hasn’t got a stamp, at least not in America, and I think it is really outrageous that he hasn’t got a stamp. Nonentities have got a stamp. A few years ago, I think it was a hundredth anniversary of the birth of Capablanca, the Cuban world chess champion, and the Cubans sent me, I think, 30 stamps that they had published and printed about Capablanca just in one year. They already had stamps in earlier years of Capablanca. Morphy, I think everyone agrees, was probably the greatest genius of them all, and he’s never gotten a single stamp." I add that in comparing Capablanca to Morphy- Fischer considered Capablanca to be possibly the greatest player of all time (also with Morphy and Tarrasch)
Another quote sums up as Fischer saying that while he may have had worse openings- Morphy even in his day would have demolished anyone (maybe excluding himself) in a set match. His theory being that even if he lost a couple games- he would adapt and simply outplay the inferior opponents he faced. I always found it ironic that people assumed that 'chess theory was worse then' simply because other great players say that. On a relative scale, I hear great players say certain openings are bad while other great players make "revivals" and "refutations" of such openings and lines. All such refutations and revivals were created by players, relatively, far, far weaker in playing strength to Paul Morphy (for the most part- although Carlsen tends to believe he would win fairly straightforwardly but it would be quite a lot more difficult that others might believe- to sum up his own words {do you see a pattern, most world champions are pretty confident that they are the best ever}). If you doubt my arguments I charge you- try plugging in some Morphy games into Stockfish. You will find that not only was he one of the most accurate players of all time (albeit what is thought of as being slightly less advantageous opening play {mostly by inferior opponents!!!})- but also that he surprises Stockfish quite a bit more often per recorded game than, say, any given Carlsen game. Now we move on to my next point- what else has surprised the best known chess engines/therefore best player of our royal game on the planet?
The answer is Alpha Zero. Not only was Alpha Zero's match win the most dominant ever seen in the history of computer chess, but it was the most shocking. All human-trained engines play a very specific style for the most part. Anti-tactics. Anti-disadvantages. Small strategic advantages and ridiculous tactical vision that leads to mostly draws. Well- alpha zero scored 28 wins and no losses in a 100 game match with this world-leading foe. That result alone is breathtaking. Even more insane was the style of its play. We hear often about grandmasters reviving old openings and lines thought at one point unplayable. This was blind ignorance of chess. For a 2200 level player never able to play a man of Morphy's talent to state "this is bad because of this".. Well.. The players Morphy was obliterating were of their level and above (often doing so.. WITHOUT A ROOK.) So this statement of fact is preposterous. They certainly cannot determine what his next moves would be nor how they might react in such a game. They can spew off knowledge from stockfish- but Morphy would not play like Stockfish, and their preparation would be null. Also in the advent of Alpha Zero, stockfish seems less relevant as far as.. objective truths in chess.. does it not? Another disclaimer, no, I do not argue nor do I believe Morphy or any human could best a modern engine in a match. They are too precise. Too tactical and deep, too sharp. They do not err the way we humans do. It is due to this fact and their own play that we humans have "refuted" certain lines or even styles of play. My grand point here is, plug in an immortal game (specifically "the immortal draw" into stockfish, and watch how accurate they play- and at times how shocked the engine is on many moves as it realizes chess truths it never considered.. {ie. the king is sometimes quite safe without castling depending on the nature of the position, and that often times sacrificing a pawn for development or activity is absolutely okay or even correct.}) If nothing else Alpha Zero backs this idea, as it abandoned pawns and even PIECES for the sake of activity (shocking the computer and obliterating it in endgames!!!"). The above mentioned fact is merely one example of how the highest-level chess-playing-entity on the planet plays in a style vastly similar to the romantic era, while welding machine-like precision into grasp. Immediately after thousands of grandmasters have abandoned and ridiculed this style of chess.
For my final exmaple: Fischer. As I have heard masters of the 2200-2300 level, even experts and class A players tell me that Morphy's openings were so bad they would outplay him easily (freaking ridiculous- I wish I could spawn him to disprove this, Fischer would agree). Most players, even many rating systems (remember Elo isn't necessarily the most accurate- yet most widely-accepted rating system) put Fischer's play, within a year or so, (1968-69) as the best play ever witnessed by man: the most accurate, the most precise. Now consider that years later after Fischer's prime Kasparov said of the 1992 championship "they were playing chess of 1972". His opinion is one of, genius he was, "they were inferior because they didn't know what we do now". On a certain scale this is true. But just as Fischer said. In a set match, do you honestly believe the best players in the world wouldn't immediately find how their openings were less advantageous, or dubious, and immediately change it to fit the current style? I assure you they would. Fischer even assures you, they would. They were simply the best human players around- and no player with superior knowledge of "more modern theory" could outplay them in the vast complexities of the game they mastered better than all the rest around them. (Keep in mind, during their tenures, other top players were constantly trying to refute their play.. and losing to them). Would they lose a few to new ideas they hadn't previously considered or played against? Certainly, but almost undoubtedly they would regroup and fortify their own ideas and others' with unimaginable accuracy.
So, in conclusion. We see that these players were the best in the world, and that only they could adequately reflect their accuracy in analysis. If you doubt me, again, I charge you to plug in these Fischer games (68-69) and Morphy games (almost all of the recorded ones) into stockfish and watch how surprising this style of play can be to the silicon monster..  Then watch how surprised it is with Alpha Zero's play (it played many a gambit thought previously "unplayable" by completely inadequate master analysis). These great players, not just Morphy and Fischer but any, and all of them possess pieces of chess truth- OBJECTIVE chess truths. Truths only the players themselves could attest to- and not inferior tactical strategists, who would be and were obliterated by them. Just today, I heard an expert player tell me "{exf4 followed by} g5 was not the best way to combat the king's gambit as black" (an outrageously opinionated and relative statement, as Capablanca played that line and stockfish still to this day prefers that line). Still note, said player was above my level, and as such I have no right nor desire to "refute" the king gambit declined, I only suggest that the KGA is absolutely playable for black. And countless times I have heard masters and experts speak of truths they have not yet discovered, but only heard of.. (one master who analysed Alpha Zero claimed it refuted e4... despite the fact that in the 1300 games it played.. 100 in the ruy lopez and 100 in the french defense: it lost NONE and won countless times- in those e4 openings... while it did lose something like 24 games of those 1300 total: not to bad). Overall, my argument is: next time you hear a player refute something, next time you hear from a human that "this is not the best way to play chess", when referring to a strategy, opening, or line, remember that some of the world's best wood-pushers would have fiercely disagreed. Remember that such style of play won them world titles and posthumous recognition in the game of kings and queens, amongst champions and amateurs. FIND out what works and what doesn't for yourself. This is what our new champion Alpha Zero does (hardware advantage not applicable to the point, as the feat still resulted in the greatest chess play we have ever seen as humans). Listening to what others think, no matter how good they are, will only hamper your tactical vision. Learn from your mistakes, learn from others' successes. But never blindly trust the word of any chessplayer- even Carlsen- as even he plays this computeristic safe style of avoiding tactics and gaining small advantages(that is not to say- if Carlsen tells you a tactic doesn't work... I'd go ahead and take his word for it unless you also can compete at the 2800+ level). You can't blame him, after all Stockfish was the best. But now, who knows who's strategy fares the finest, the purest, the most accurate and advantageous- it could even have been Morphy's; who are we mere mortals to challenge his ideas long-lost in the grave. Chess is forever beautiful in its intricate complexity.

WilliamShookspear
GrahaMasterBuckner wrote:

Over time chess strategy has changed. I am only 25 years old- I grew up in the post-Kasparov: human-trained machine chess era. A short disclaimer. I do not doubt myself that all below mentioned players would be outplayed by a machine of the present. Stockfish- even after Alpha Zero's loss- is great because it will never tactically blunder against a human (and its strategic vision is also more than adequate). Humans simply cannot play full 50+ move games of chess without inaccuracies as these silicon monsters will consistently do game-after-game-after-game. In this post I will present examples of how chess theory has changed and even folded-on itself over the decades and what it means about the world's best players of all time.
Firstly we begin with Morphy. For you skeptics, and it shall be a great many, I will start by quoting Fischer 
"‘Paul Morphy was a great chess player, a genius. I’m a little embarrassed when I think that I’ve got a [commemorative] stamp and I’m still even alive. As far as I know he hasn’t got a stamp, at least not in America, and I think it is really outrageous that he hasn’t got a stamp. Nonentities have got a stamp. A few years ago, I think it was a hundredth anniversary of the birth of Capablanca, the Cuban world chess champion, and the Cubans sent me, I think, 30 stamps that they had published and printed about Capablanca just in one year. They already had stamps in earlier years of Capablanca. Morphy, I think everyone agrees, was probably the greatest genius of them all, and he’s never gotten a single stamp." I add that in comparing Capablanca to Morphy- Fischer considered Capablanca to be possibly the greatest player of all time (also with Morphy and Tarrasch)
Another quote sums up as Fischer saying that while he may have had worse openings- Morphy even in his day would have demolished anyone (maybe excluding himself) in a set match. His theory being that even if he lost a couple games- he would adapt and simply outplay the inferior opponents he faced. I always found it ironic that people assumed that 'chess theory was worse then' simply because other great players say that. On a relative scale, I hear great players say certain openings are bad while other great players make "revivals" and "refutations" of such openings and lines. All such refutations and revivals were created by players, relatively, far, far weaker in playing strength to Paul Morphy (for the most part- although Carlsen tends to believe he would win fairly straightforwardly but it would be quite a lot more difficult that others might believe- to sum up his own words {do you see a pattern, most world champions are pretty confident that they are the best ever}). If you doubt my arguments I charge you- try plugging in some Morphy games into Stockfish. You will find that not only was he one of the most accurate players of all time (albeit what is thought of as being slightly less advantageous opening play {mostly by inferior opponents!!!})- but also that he surprises Stockfish quite a bit more often per recorded game than, say, any given Carlsen game. Now we move on to my next point- what else has surprised the best known chess engines/therefore best player of our royal game on the planet?
The answer is Alpha Zero. Not only was Alpha Zero's match win the most dominant ever seen in the history of computer chess, but it was the most shocking. All human-trained engines play a very specific style for the most part. Anti-tactics. Anti-disadvantages. Small strategic advantages and ridiculous tactical vision that leads to mostly draws. Well- alpha zero scored 28 wins and no losses in a 100 game match with this world-leading foe. That result alone is breathtaking. Even more insane was the style of its play. We hear often about grandmasters reviving old openings and lines thought at one point unplayable. This was blind ignorance of chess. For a 2200 level player never able to play a man of Morphy's talent to state "this is bad because of this".. Well.. The players Morphy was obliterating were of their level and above (often doing so.. WITHOUT A ROOK.) So this statement of fact is preposterous. They certainly cannot determine what his next moves would be nor how they might react in such a game. They can spew off knowledge from stockfish- but Morphy would not play like Stockfish, and their preparation would be null. Also in the advent of Alpha Zero, stockfish seems less relevant as far as.. objective truths in chess.. does it not? Another disclaimer, no, I do not argue nor do I believe Morphy or any human could best a modern engine in a match. They are too precise. Too tactical and deep, too sharp. They do not err the way we humans do. It is due to this fact and their own play that we humans have "refuted" certain lines or even styles of play. My grand point here is, plug in an immortal game (specifically "the immortal draw" into stockfish, and watch how accurate they play- and at times how shocked the engine is on many moves as it realizes chess truths it never considered.. {ie. the king is sometimes quite safe without castling depending on the nature of the position, and that often times sacrificing a pawn for development or activity is absolutely okay or even correct.}) If nothing else Alpha Zero backs this idea, as it abandoned pawns and even PIECES for the sake of activity (shocking the computer and obliterating it in endgames!!!"). The above mentioned fact is merely one example of how the highest-level chess-playing-entity on the planet plays in a style vastly similar to the romantic era, while welding machine-like precision into grasp. Immediately after thousands of grandmasters have abandoned and ridiculed this style of chess.
For my final exmaple: Fischer. As I have heard masters of the 2200-2300 level, even experts and class A players tell me that Morphy's openings were so bad they would outplay him easily (freaking ridiculous- I wish I could spawn him to disprove this, Fischer would agree). Most players, even many rating systems (remember Elo isn't necessarily the most accurate- yet most widely-accepted rating system) put Fischer's play, within a year or so, (1968-69) as the best play ever witnessed by man: the most accurate, the most precise. Now consider that years later after Fischer's prime Kasparov said of the 1992 championship "they were playing chess of 1972". His opinion is one of, genius he was, "they were inferior because they didn't know what we do now". On a certain scale this is true. But just as Fischer said. In a set match, do you honestly believe the best players in the world wouldn't immediately find how their openings were less advantageous, or dubious, and immediately change it to fit the current style? I assure you they would. Fischer even assures you, they would. They were simply the best human players around- and no player with superior knowledge of "more modern theory" could outplay them in the vast complexities of the game they mastered better than all the rest around them. (Keep in mind, during their tenures, other top players were constantly trying to refute their play.. and losing to them). Would they lose a few to new ideas they hadn't previously considered or played against? Certainly, but almost undoubtedly they would regroup and fortify their own ideas and others' with unimaginable accuracy.
So, in conclusion. We see that these players were the best in the world, and that only they could adequately reflect their accuracy in analysis. If you doubt me, again, I charge you to plug in these Fischer games (68-69) and Morphy games (almost all of the recorded ones) into stockfish and watch how surprising this style of play can be to the silicon monster..  Then watch how surprised it is with Alpha Zero's play (it played many a gambit thought previously "unplayable" by completely inadequate master analysis). These great players, not just Morphy and Fischer but any, and all of them possess pieces of chess truth- OBJECTIVE chess truths. Truths only the players themselves could attest to- and not inferior tactical strategists, who would be and were obliterated by them. Just today, I heard an expert player tell me "{exf4 followed by} g5 was not the best way to combat the king's gambit as black" (an outrageously opinionated and relative statement, as Capablanca played that line and stockfish still to this day prefers that line). Still note, said player was above my level, and as such I have no right nor desire to "refute" the king gambit declined, I only suggest that the KGA is absolutely playable for black. And countless times I have heard masters and experts speak of truths they have not yet discovered, but only heard of.. (one master who analysed Alpha Zero claimed it refuted e4... despite the fact that in the 1300 games it played.. 100 in the ruy lopez and 100 in the french defense: it lost NONE and won countless times- in those e4 openings... while it did lose something like 24 games of those 1300 total: not to bad). Overall, my argument is: next time you hear a player refute something, next time you hear from a human that "this is not the best way to play chess", when referring to a strategy, opening, or line, remember that some of the world's best wood-pushers would have fiercely disagreed. Remember that such style of play won them world titles and posthumous recognition in the game of kings and queens, amongst champions and amateurs. FIND out what works and what doesn't for yourself. This is what our new champion Alpha Zero does (hardware advantage not applicable to the point, as the feat still resulted in the greatest chess play we have ever seen as humans). Listening to what others think, no matter how good they are, will only hamper your tactical vision. Learn from your mistakes, learn from others' successes. But never blindly trust the word of any chessplayer- even Carlsen- as even he plays this computeristic safe style of avoiding tactics and gaining small advantages(that is not to say- if Carlsen tells you a tactic doesn't work... I'd go ahead and take his word for it unless you also can compete at the 2800+ level). You can't blame him, after all Stockfish was the best. But now, who knows who's strategy fares the finest, the purest, the most accurate and advantageous- it could even have been Morphy's; who are we mere mortals to challenge his ideas long-lost in the grave. Chess is forever beautiful in its intricate complexity.

You put a lot of effort into this, so, of course, it goes hitherto unnoticed. Such is the way of Chess.com.

You raise some interesting points, some of which I agree with, some of which I do not. I agree with your point that letting other people do the thinking for you is avoiding the soul of chess, and remarkably ineffective, not to mention disempowering. I agree that you should take everything you hear at face value, and look at it yourself, and come to your own conclusion. Especially at my level, you can play all sorts of weird and wacky things and still come out on top by superior play.

That said, your point about Alpha Zero... Seems a bit naïve. One has to take into account the disparity in the Stockfish vs Alpha Zero match. Alpha Zero was running on a Google supercomputer, the magnitude of which I do not know, but what I DO know is that it makes the 44-core TCEC supercomputer look like a toy in comparison. (The Google supercomputer used "tensors", which are especially suited to neural networks.) I do not know what Stockfish was run on, but I highly doubt it was a google supercomputer. Not to mention, Stockfish was not allowed to use any databases, which is an intrinsic part of its strength. So, while Alpha Zero was very impressive, I do not put much stock into the 28-0 score, and I am confident that Alpha Zero would have won that match even if it exclusively played 1. a3 as white.

Also, the masters claiming that they could crush Morphy's openings, while their condescension rankles, may not be far off. You see, while Morphy was an absolutely brilliant player, the facts were, he was living in an age where chess knowledge was far diminished. So while a 1600 rated player would easily get annihilated by Morphy's "inferior openings" because of his brilliant vision, I think that a 2200 rated player, with reasonable opening knowledge, could probably stymie Morphy, because of our vast improvement over the past 100 years.

I agree that club players banging their heads against what the best 20th move is to gain an advantage in the Exchange Ruy Lopez are wasting their time, but I think knowing the theory, and whether things are preferred or not, is a valuable thing to know. Opening theory is a tool, like endgame databases. And you can only use your tools as well as your ability allows you. So PLAY that +/- opening as Black, but you'd better have a plan to bamboozle your opponent, because assuming inferior play from your opponent is a one way street to frustration.