|
Paul Morphy, Spring Hill College, 1854
This is a highly unusual article for me. I have no love for chess engines. Additionally, I have severe reservations about trying to compare modern players with 19th cenutry players. It has always been my opinion that Romantic players played as much for beauty as for results, seeing the Combination as the pinnacle of chess. This focus would seem to put them on unfair footing when comparing them to Scientific or Modern players.
Many pundits and patzers alike will often take to the message board and denigrate the 19th century warriors are far inferior to today's players. Across the board (so to speak) this seems like common sense. So it came as a surprise to me, a shock really, when chess.com member SteveCollyer contacted me with some computer analysis he ran on some of Morphy's games.
Steve Collyer has used his engine skills to expose computer cheaters on this site and on other chess sites. The process he uses, following certain criteria, is to compare each move in a game against the top four moves of those selected by computers and comparing the results against a model. He has used this same process to examine the play top GMs.
In explaining the methodology, Mr. Collyer told me, "The system used is for this analysis is a pretty fast quad-core pc with Houdini 1.5a 4x CPU, rated about 3200 Elo . . . The games were then auto-analysed for top 1, 2, 3 & 4 engine choice moves once out of the onboard 4m game database. . . ."
He added, "Cumulative results for the very best modern OTB Super GM's when using this technique are consistently Top 1 match =60%, Top 2 match =75%, Top 3 match =85% & Top 4 match just under 90% for sample sizes of typically 700-1000 non-database moves."
One would expect Morphy and his opponents to match significantly lower against the computer in their games than later players and certainly than modern players. But this isn't the case at all.
Mr. Collyer chose three of Morphy's strongest match opponents from his 1858-9 European tour, Adolf Anderssen, Daniel Harrwitz and John Löwenthal. From the matches with these opponents, he chose 23 games:
"Each game has a minimum 35 moves. This is so that a reasonable sample of non-theory moves can be generated in the more closely contested games.
There are 23 Morphy games in the batch. I didn't cherry-pick Morphy's best games - those that meet this basic selection criteria are all here, including several draws & a few losses.
I selected these because they were played against Morphy's strongest opponents during his crushing European tour when Morphy was at the peak of his powers. "
So, the only games culled were those that didn't comply with pre-established criteria.
Once again:
The very best modern OTB Super GM's with all their opening prep & engine use to aid them achieve match rates of around
Top 1 Match = 60%
Top 2 Match = 75%
Top 3 Match = 85%
Top 4 Match = 90%
and most are a good few % points lower when using this methodology.
Here are Morphy's and his opponents' results:
|
Löwenthal, Johann Jacob (Games: 8)
Top 1 Match: 188/373 ( 50.4% ) Morphy: 196/370 ( 53.0% )
Top 2 Match: 250/373 ( 67.0% ) Morphy: 263/370 ( 71.1% )
Top 3 Match: 286/373 ( 76.7% ) Morphy: 293/370 ( 79.2% )
Top 4 Match: 311/373 ( 83.4% ) Morphy: 315/370 ( 85.1% )
Anderssen, Adolf (Games: 8)
Top 1 Match: 168/349 ( 48.1% ) Morphy: 202/349 ( 57.9% )
Top 2 Match: 238/349 ( 68.2% ) Morphy: 266/349 ( 76.2% )
Top 3 Match: 271/349 ( 77.7% ) Morphy: 290/349 ( 83.1% )
Top 4 Match: 290/349 ( 83.1% ) Morphy: 304/349 ( 87.1% )
Harrwitz, Daniel (Games: 7)
Top 1 Match: 137/295 ( 46.4% ) Morphy: 155/295 ( 52.5% )
Top 2 Match: 193/295 ( 65.4% ) Morphy: 202/295 ( 68.5% )
Top 3 Match: 224/295 ( 75.9% ) Morphy: 225/295 ( 76.3% )
Top 4 Match: 242/295 ( 82.0% ) Morphy: 242/295 ( 82.0% )
Morphy, Paul (Games: 23)
Top 1 Match: 553/1014 ( 54.5% ) Opponents: 493/1017 ( 48.5% )
Top 2 Match: 731/1014 ( 72.1% ) Opponents: 681/1017 ( 67.0% )
Top 3 Match: 808/1014 ( 79.7% ) Opponents: 781/1017 ( 76.8% )
Top 4 Match: 861/1014 ( 84.9% ) Opponents: 843/1017 ( 82.9% )
Composite scores:
Top 1 Match: 1046/2031 ( 51.5% )
Top 2 Match: 1412/2031 ( 69.5% )
Top 3 Match: 1589/2031 ( 78.2% )
Top 4 Match: 1704/2031 ( 83.9% )
|
For comparative purposes, here are results posted by modern players, using the same game selection and analysis criteria:
|
Petrosian-Spassky 1960 WC:
Petrosian (Games: 16)
Top 1 Match: 309/625 ( 49.4% )
Top 2 Match: 430/625 ( 68.8% )
Top 3 Match: 491/625 ( 78.6% )
Top 4 Match: 530/625 ( 84.8% )
Boris Spassky (Games: 16)
Top 1 Match: 325/622 ( 52.3% )
Top 2 Match: 420/622 ( 67.5% )
Top 3 Match: 470/622 ( 75.6% )
Top 4 Match: 513/622 ( 82.5% )
Karpov-Kasparov 1984/85 WC:
Garry Kasparov (Games: 21)
Top 1 Match: 342/659 ( 51.9% )
Top 2 Match: 478/659 ( 72.5% )
Top 3 Match: 545/659 ( 82.7% )
Top 4 Match: 580/659 ( 88.0% )
Anatoly Karpov (Games: 21)
Top 1 Match: 363/657 ( 55.3% )
Top 2 Match: 474/657 ( 72.1% )
Top 3 Match: 529/657 ( 80.5% )
Top 4 Match: 567/657 ( 86.3% )
Karpov-Kasparov 1996 WC:
Anatoly Karpov (Games: 14)
Top 1 Match: 228/426 ( 53.5% )
Top 2 Match: 302/426 ( 70.9% )
Top 3 Match: 344/426 ( 80.8% )
Top 4 Match: 371/426 ( 87.1% )
Garry Kasparov (Games: 14) }
Top 1 Match: 228/429 ( 53.1% )
Top 2 Match: 297/429 ( 69.2% )
Top 3 Match: 343/429 ( 80.0% )
Top 4 Match: 375/429 ( 87.4% )
Karpov-Kamsky 1996 WC:
Gata Kamsky (Games: 16)
Top 1 Match: 359/699 ( 51.4% )
Top 2 Match: 487/699 ( 69.7% )
Top 3 Match: 567/699 ( 81.1% )
Top 4 Match: 602/699 ( 86.1% )
Anatoly Karpov (Games: 16)
Top 1 Match: 373/700 ( 53.3% )
Top 2 Match: 501/700 ( 71.6% )
Top 3 Match: 573/700 ( 81.9% )
Top 4 Match: 616/700 ( 88.0% )
Hikaru Nakamura 2014/2013 games vs Super GM's:
Hikaru Nakamura (Games: 20)
Top 1 Match: 504/871 ( 57.9% ) Opponents: 441/869 ( 50.7% )
Top 2 Match: 641/871 ( 73.6% ) Opponents: 624/869 ( 71.8% )
Top 3 Match: 708/871 ( 81.3% ) Opponents: 700/869 ( 80.6% )
Top 4 Match: 764/871 ( 87.7% ) Opponents: 752/869 ( 86.5% )
Fabiano Caruana 2014/2013 games vs Super GM's:
Fabiano Caruana (Games: 20)
Top 1 Match: 560/1018 ( 55.0% ) Opponents: 508/1019 ( 49.9% )
Top 2 Match: 749/1018 ( 73.6% ) Opponents: 694/1019 ( 68.1% )
Top 3 Match: 834/1018 ( 81.9% ) Opponents: 802/1019 ( 78.7% )
Top 4 Match: 887/1018 ( 87.1% ) Opponents: 854/1019 ( 83.8% )
Vladimir Kramnik 2014/2013 games vs Super GM's:
Vladimir Kramnik (Games: 20)
Top 1 Match: 535/938 ( 57.0% ) Opponents: 523/936 ( 55.9% )
Top 2 Match: 701/938 ( 74.7% ) Opponents: 720/936 ( 76.9% )
Top 3 Match: 776/938 ( 82.7% ) Opponents: 785/936 ( 83.9% )
Top 4 Match: 819/938 ( 87.3% ) Opponents: 826/936 ( 88.2% )
|
And from the later 19th century Scientific players:
Lasker-Steinitz 1894 WC
Emanuel Lasker (Games: 18)
{ Top 1 Match: 395/747 ( 52.9% )
{ Top 2 Match: 523/747 ( 70.0% )
{ Top 3 Match: 617/747 ( 82.6% )
{ Top 4 Match: 657/747 ( 88.0% )
Wilhelm Steinitz (Games: 18)
Top 1 Match: 389/744 ( 52.3% )
Top 2 Match: 525/744 ( 70.6% )
Top 3 Match: 586/744 ( 78.8% )
Top 4 Match: 629/744 ( 84.5% )
|
How one interprets this data will probably differ from person to person, but there is no doubt that Houdini has an appreciation for Morphy's play.
|
