How strong were Fischer and Morphy?

As he often does, Nakamura has incited alot of controversy with his comments in a reddit AMA that Fischer could not compete with today's top players because the game has so changed due to computers.

Both Grandmasters David Smerdon and Ben Finegold have weighed in with interesting blog posts (Finegold's and Smerdon's). GM Finegold, who I respect and whose commentary I greatly enjoy, states that he "has never been so mad." He asserts that both Fischer and Morphy would destroy almost all of their GM opponents. GM Smerdon predicts a 6-0 Fischer-Nakamura whitewashing once Bobby has incorporated computers in his preparation.

This controversy is not new. Chess players have debated the relative playing strengths of the great players of the past for as long as chess has had great players of the past. For a long time, there was no objective means of determining how strong players such as Capablanca, Morphy, and Charousek really were. However, today chess engines provide a valid, objective standard by which to assess player's from bygone errors.

Rating Inflation (or the lack thereof)

Dr. (and IM) Kenneth Regan has received much well-deserved publicity for his work in this area and the associated area of identifying cases of cheating in chess. Dr. Regan and I both live in Buffalo, NY, and I have been happy to have the opportunity to discuss his research with him several times. It is truly inspiring work, and his methods are extremely logical and understandable.

Dr. Regan has published several papers and written frequently about his methods. His papers and writings are available at his website (http://www.cse.buffalo.edu/~regan/). To summarize his methods very briefly, he has compared player performance to an objective standard (Rybka 3). He compares the moves played to the engine's top choices and determines a valued difference between the player's and the engine's choice. In doing so he ignores the opening moves, and he ignores differences in winning moves. For instance, a mate in 4 or a completely winning endgame are equally good - something that Tactics Trainer could learn from ;)

Using these techniques, Dr. Regan with Guy Haworth looked at whether or not there has been rating inflation since the introduction of the ELO system. He determined that there has been practically no inflation (Regan and Haworth 2011). A 2600 player today performs similarly to a 2600 player from 1960.

To quote from Regan and Haworth, "In the 1970’s there were only two players with ratings over 2700, namely Bobby Fischer and Anatoly Karpov, and there were years as late as 1981 when no one had a rating over 2700 (see Wee 2000). In the past decade there have usually been thirty or more players with such ratings. Thus lack of inflation implies that those players are better than all but Fischer and Karpov were."

I think the first thing to note is how strong Fischer and Karpov were compared to their peers. Today, Fischer and Karpov at their peak would still be among the best players in the world. Fischer's peak is particularly impressive. At his best, he was near 2800, long before another player crossed 2700. This is in favor of GM Finegold's and Smerdon's specific arguments about Fischer's strength at the expense of Nakamura's. Fischer was fantastically strong (especially compared to his peers), and at his peak, he would be an excellent match for any of today's players. However, he would be no means dominate any of them, and he would likely be a significant underdog against Carlsen today.

Another thing to note is simply how strong today's top players are. It is easy to overlook a player like Sutovsky or Bologan in the clustered market of players, but these players are easily as strong as many of the past greats, and their games well reward study.

Many persons (including GM Smerdon) seem to assert that rating inflation is a given. They say that the presence of so many 2700+ players today as compared to the past is a clear indication of inflation. After all, Movsesian couldn't really be stronger than Spassky could he? In fact, the top players of today are simply stronger than the top players of the past. This shouldn't be surprising. There are more players than ever; they are starting younger than ever, and they have a mountain of prior knowledge to build upon. Dr. Regan's research clearly shows that the notion of rating inflation is largely incorrect.

How strong were the early masters?

Dr. Regan has also been able ask the question, "what is a player's expected rating given a players performance?" He has used this research laregly to address cheating scandals. He has been able to feed a suspected cheater's tournament games into his system and determine whether or not that player's performance was compatible with his rating. He provides an example in a New York Times article article about his work (http://www.nytimes.com/2012/03/20/science/a-computer-program-to-detect-possible-cheating-in-chess.html?ref=science&_r=0).

"At the Canadian Open last year, a player whose rating was 2100 (a candidate master) beat three international masters, whose ratings are usually at least 300 points higher. After analyzing the games, Dr. Regan said, “I was able to prove that his intrinsic rating was in the 2100s and the international masters had just played poorly.”

Such a system can also be used to calculate ratings for great players from the past such as Morphy, Steinitz, Lasker, Capablanca, Alekhine, etc. How strong were these players? Sadly, Dr. Regan has not published research regarding the intrinsic performance of the early masters, but I hope it is forthcoming. However, Chesszen.com has used similar methods to claculate a player's performance and they estimate Morphy's rating in the low 2300's. Capablanca is estimated as playing on a high 2500 level. While these ratings do not seem to be calculated with nearly the rigor that Dr. Regan's research may provide, I believe they are generally accurate. Morphy was probably around 2300. I believe the early masters such as Lasker, Rubinstein, Pillsbury, etc. were probably right around 2500. I believe many subjective guesstimates of these player's strength are inflated because because their masterpieces are reviewed and not the entirety of their games. These guesses are in line with Dr. Regan's well established findings that "human skill at chess has been improving over time" (Regan, Haworth, and Macieja 2011). My own suppostition is that eventually, one will be able to show a fairly linear increase in the playing strength of top players from 2300 around the time of Morphy to 2400 in the time of Steinitz to 2500 in the time of Lasker, 2600 in the time of Botvinnik and Smyslov, 2700 for Fischer and Karpov in the 70's, 2800 for Kasparov in the 90's and possibly 2900 for Carlsen in the 10's.

There is also interesting anecdotal support for the notion that the early masters were not objectively on the level of today's top players from GM John Nunn (my favorite chess author), who dedicated one chapter (The Test of Time) of his excellent John Nunn's Chess Puzzle Book to the Karlsbad 1911 tournament. He says...

"I decided to take two tournaments, one from the historical past and one recent, and analyse all the games in the two tournaments looking for serious errors. Since I wanted a fairly large sample, I chose tournaments containing a considerable number of games. My historical tournament was Karlsbad 1911 (325 games). This event seemed to have the qualities I was looking for: top players such as Alekhine, Nimzowitsch, Schlechter and Rubinstein, together with only moderately familiar names such as Perlis and Fahmi (I did not want to restrict my assessments to the very top) and a tournament book by a well-known player (Vidmar) to help me find errors." and further on...

"I had no particular preconceptions about what the results of this search would be. Like most contemporary grandmasters, I was familiar with all the standard textbook examples from the early part of the century, but I had never before undertaken a systematic examination of a large number of old games. I was quite surprised by the results. To summarize, the old players were much worse than I expected. The blunders thrown up by Fritz were so awful that I looked at a considerable number of complete games 'by hand', wondering if the Fritz results really reflected the general standard of play. They did."

"In order to be more specific about Karlsbad, take one player: Hugo Stichting ( 1874-1916). At Karlsbad he scored 11 1/2 / 25 or 'minus 2', as they say these days - a perfectly respectable score. Having played over all his games at Karlsbad I think that I can confidently state that his playing strength was not greater than Elo 2100 (BCF 187) - and that was on a good day and with a following wind."

The players of Carlsbad 1911

Legacy

Many players, such as GM Finegold in his post, get a slight tough heated when it is suggested that the great players of the past are simply not as strong as their modern counterparts. I cannot understand this mentality. In no other sport that I am aware of do people defend the past greats as being truly on par with modern athletes. Pundits seem to accept that modern football players would dominate the football players of the 50's. There have simply been too many improveements in diet, training, and technique for it to be otherwise. The same is true of chess. Today's players have a magnificent foundation of theory in the opening, middlegame, and endgame. They have been exposed to more tactics, more great games, and more exceptional positions than players of the past could have ever hoped to. How could we reasonably expect that Capablanca could have objectively played on the same level as Carlsen, Caruana, or Kamsky?!

In other sports, it is accepted that the objective discrepancies in skill across the ages do not in any way diminish the legacy of the great players. They are the pillars of their time. Babe Ruth and Cy Young are still considered two of the greatest players of all time even though no one would argue that they could time travel to the modern era and achieve the same level of baseball success.

Babe Ruth playing for the Red Sox

Morphy, Steinitz, Lasker, Pillsbury, Marshall, Capablanca, Rubinstein, Nimzowitsch, Alekhine, Flohr, Botvinnik, Averbakh, Tal and so many other masters have produced countless brilliancies and made enormous contributions to the global understanding of chess. They are well worthy of our esteem, gratefulness, and celebration regardless of their objective playing strength.

Talent

Another argument to make is that players such as Morphy, Capablanca, and Fischer were the greatest chess talents of all time. Their success relative to their peers indicates that they had a remarkable gift for chess, and had they been born in the modern era, they would have achieved all that today's top players have achieved and perhaps more. This is obviously a fanciful argument, but in principle, I agree. These players deserve recognition for being as talented, and perhaps more so, than today's talents such as Carlsen. After all, it is arguably at least as an enormous achievement to be the first 2300 strength player as it is to be the first 2900 strength player in history.

Empowerment

Many seem to find the notion that the great players of the past weren't objectively all that strong relative to today's players disempowering. On the contrary, I find it enormously empowering. People tend to judge their skills by their peers. I constantly hear great players (experts, masters, international masters, and even grandmasters) be excessively self-depecrating. It is as though, if you are not on the level of Aronian or Carlsen, you cannot have legitimate insight into a position, or you cannot produce a true brilliancy. I find it enormously empowering to think that Philidor, Morphy, and Andersson were not even close to GM strength, and yet their moves and ideas have enthralled chess players for generations. It inspires me to believe that while I may not be as strong as Carlsen, perhaps I too can produce a game that will last through the ages.

Private lessons available - see http://www.chess.com/coach/sam-copeland for details! For interested students, I offer one FREE annotated game.