The Golden Age of Chess
by Bill Forster
This little article is the result of a thought experiment, followed by a little recreational computing to follow through.
The top players today play a lot. Carlsen and Firouzja might play as many games in an online bullet session as Alekhine and Capablanca played in decades of rivalry. No one can possibly keep up with all the chess that's played today. One of the charms of looking back at a different and slower time is that you get the feeling that you can almost know all of the top level chess of that time. For example, early in the twentieth century the top players competed in a series of famous tournaments that still resonate today, perhaps because there were comparitively few of them. The idea occurred to me to collect all the top games of this golden age, from just the best of the best.
Here is the plan I came up with. Start with Alekhine, Capablanca and Lasker, the three great world champions of their age. Collect all games played by the big three themselves, plus all games played by all players who played all three titans at least once in their career. Now we have all the rivals and challengers of the 1910s, 20s and 30s. Spoiler alert: I got 62 players this way. Next start a reducing process. First eliminate all games featuring anyone out of the group. Then for each player in the group count which other players in the group they faced in their career. By definition, Lasker Capablanca and Alekhine faced everyone. Eliminate the player who faced the least number of other players. Another spoiler alert: the first player eliminated was the famous composer and pianist Sergei Prokofiev, an expert player who played the big three and only the big three!
Finally, repeat this process successively removing players and the games they featured in until there are only players who have all played each other at least once.
I was very happy with the way the whole thing worked out. Without any need to cheat or tweak things, I ended up with the nice round number of twenty players. All twenty players had played at least 150 games with others in the group. (Some of the last players eliminated had played many less). It would be hard to quibble that the group of players blindly selected by the algorithm represent a certain era of chess more or less perfectly. Perhaps I'd like to sub out Sir George Thomas for Carl Schlecter, but I don't make the rules. Actually I do make the rules, but they are rules. Schlecter misses out because there are a few players he didn't play, because he really belongs to an earlier era. Great players like Euwe, Reshevsky, Fine and Botvinnik also miss out for the opposite reason (they really belong to a later era).
Below I present the results of the experiment as a kind of "whole of career" tournament. Placings in this tournament are determined by percentage score rather than absolute score, because obviously each player played a different number of games. Just in case it's not absolutely clear, the only games included, the only games that count here, are games played within this group of twenty players. I used my personal Chessbase "Big database" from 2007, mainly because they are very professional and consistent in their naming. I am sure a few extra games have emerged in the last 15 years. I'm also sure no extra games have been played in the last 15 years! :-) I ended up with a reasonably manageable collection of 2757 games. Because it is a subset of a commercial database I am not comfortable about sharing it on the server. If there's any interest I might repeat the exercise with a non-commercial alternative database hopefully generating very similar results. I suspect the top placings in my tournament would not be affected by a few different games. Happily our three heroes take the top three positions, and Lasker's well known statistical superiority is confirmed.
| Player | Total | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 Lasker, Emanuel | 69.3% +96 -22 =74 133/192 | +2 -6 =16 | +3 -1 =4 | +2 -1 =4 | +0 -1 =1 | +4 -1 =2 | +1 -0 =1 | +4 -0 =1 | +3 -1 =3 | +5 -1 =2 | +0 -0 =2 | +1 -0 =3 | +18 -4 =8 | +12 -2 =11 | +4 -0 =2 | +1 -0 =0 | +26 -4 =7 | +0 -0 =1 | +8 -0 =5 | +2 -0 =1 | |
| 2 Capablanca, Jose Raul | 64.8% +97 -21 =138 166/256 | +6 -2 =16 | +9 -7 =33 | +1 -1 =7 | +5 -0 =6 | +3 -0 =6 | +3 -0 =6 | +2 -0 =1 | +5 -0 =9 | +5 -0 =2 | +0 -0 =2 | +2 -2 =8 | +2 -1 =3 | +21 -4 =29 | +5 -1 =2 | +6 -1 =4 | +9 -1 =1 | +1 -1 =0 | +4 -0 =0 | +8 -0 =3 | |
| 3 Alekhine, Alexander | 63.6% +149 -50 =165 231.5/364 | +1 -3 =4 | +7 -9 =33 | +8 -4 =2 | +9 -3 =9 | +6 -0 =5 | +4 -1 =10 | +3 -3 =4 | +10 -2 =14 | +38 -16 =40 | +4 -1 =5 | +3 -2 =10 | +9 -1 =2 | +7 -0 =7 | +3 -1 =5 | +7 -0 =6 | +4 -2 =2 | +11 -0 =4 | +4 -0 =0 | +11 -2 =3 | |
| 4 Rubinstein, Akiba | 57.5% +133 -79 =147 206.5/359 | +1 -2 =4 | +1 -1 =7 | +4 -8 =2 | +7 -6 =9 | +4 -1 =9 | +4 -3 =6 | +3 -3 =10 | +10 -6 =12 | +13 -12 =10 | +4 -0 =11 | +15 -11 =8 | +8 -0 =12 | +11 -9 =15 | +9 -4 =8 | +5 -1 =4 | +5 -3 =0 | +6 -0 =6 | +15 -7 =6 | +8 -2 =8 | |
| 5 Nimzowitsch, Aaron | 56.0% +81 -52 =107 134.5/240 | +1 -0 =1 | +0 -5 =6 | +3 -9 =9 | +6 -7 =9 | +1 -0 =7 | +2 -3 =8 | +1 -1 =6 | +5 -3 =15 | +5 -5 =4 | +2 -2 =3 | +13 -4 =13 | +5 -2 =4 | +6 -5 =8 | +5 -1 =5 | +3 -1 =3 | +3 -0 =2 | +10 -1 =0 | +5 -2 =0 | +5 -1 =4 | |
| 6 Maroczy, Geza | 53.9% +72 -52 =132 138/256 | +1 -4 =2 | +0 -3 =6 | +0 -6 =5 | +1 -4 =9 | +0 -1 =7 | +2 -0 =10 | +3 -1 =11 | +2 -3 =18 | +7 -4 =3 | +2 -3 =3 | +6 -3 =7 | +3 -2 =11 | +11 -4 =10 | +2 -1 =8 | +1 -3 =5 | +9 -5 =5 | +2 -3 =2 | +10 -0 =8 | +10 -2 =2 | |
| 7 Vidmar, Milan | 53.2% +45 -34 =92 91/171 | +0 -1 =1 | +0 -3 =6 | +1 -4 =10 | +3 -4 =6 | +3 -2 =8 | +0 -2 =10 | +2 -0 =4 | +5 -3 =12 | +5 -1 =5 | +1 -0 =3 | +4 -4 =10 | +1 -1 =1 | +3 -3 =6 | +2 -2 =1 | +2 -0 =5 | +5 -1 =0 | +0 -0 =1 | +4 -2 =1 | +4 -1 =2 | |
| 8 Teichmann, Richard | 52.1% +57 -49 =85 99.5/191 | +0 -4 =1 | +0 -2 =1 | +3 -3 =4 | +3 -3 =10 | +1 -1 =6 | +1 -3 =11 | +0 -2 =4 | +3 -0 =7 | +0 -0 =1 | +0 -0 =2 | +9 -4 =8 | +5 -8 =2 | +7 -6 =14 | +1 -2 =0 | +1 -0 =0 | +4 -5 =4 | +1 -0 =1 | +16 -6 =8 | +2 -0 =1 | |
| 9 Tartakower, Saviely | 51.4% +109 -97 =217 217.5/423 | +1 -3 =3 | +0 -5 =9 | +2 -10 =14 | +6 -10 =12 | +3 -5 =15 | +3 -2 =18 | +3 -5 =12 | +0 -3 =7 | +7 -11 =8 | +4 -3 =18 | +17 -12 =28 | +4 -2 =6 | +7 -0 =10 | +16 -14 =17 | +9 -3 =10 | +3 -2 =2 | +5 -1 =12 | +5 -3 =6 | +14 -3 =10 | |
| 10 Bogoljubow, Efim | 50.4% +129 -126 =119 188.5/374 | +1 -5 =2 | +0 -5 =2 | +16 -38 =40 | +12 -13 =10 | +5 -5 =4 | +4 -7 =3 | +1 -5 =5 | +0 -0 =1 | +11 -7 =8 | +3 -5 =6 | +18 -12 =9 | +6 -3 =3 | +2 -2 =4 | +17 -7 =4 | +5 -0 =4 | +1 -2 =0 | +15 -6 =9 | +3 -1 =3 | +9 -3 =2 | |
| 11 Gruenfeld, Ernst | 49.7% +36 -37 =102 87/175 | +0 -0 =2 | +0 -0 =2 | +1 -4 =5 | +0 -4 =11 | +2 -2 =3 | +3 -2 =3 | +0 -1 =3 | +0 -0 =2 | +3 -4 =18 | +5 -3 =6 | +4 -5 =11 | +2 -1 =6 | +1 -1 =3 | +3 -5 =8 | +3 -0 =3 | +0 -1 =1 | +2 -2 =11 | +2 -0 =1 | +5 -2 =3 | |
| 12 Spielmann, Rudolf | 47.2% +111 -134 =171 196.5/416 | +0 -1 =3 | +2 -2 =8 | +2 -3 =10 | +11 -15 =8 | +4 -13 =13 | +3 -6 =7 | +4 -4 =10 | +4 -9 =8 | +12 -17 =28 | +12 -18 =9 | +5 -4 =11 | +6 -6 =6 | +6 -8 =14 | +12 -16 =12 | +2 -1 =3 | +4 -2 =1 | +7 -2 =8 | +9 -5 =8 | +6 -2 =4 | |
| 13 Tarrasch, Siegbert | 46.9% +77 -94 =106 130/277 | +4 -18 =8 | +1 -2 =3 | +1 -9 =2 | +0 -8 =12 | +2 -5 =4 | +2 -3 =11 | +1 -1 =1 | +8 -5 =2 | +2 -4 =6 | +3 -6 =3 | +1 -2 =6 | +6 -6 =6 | +13 -7 =18 | +2 -5 =4 | +1 -1 =3 | +9 -5 =3 | +4 -0 =3 | +14 -5 =8 | +3 -2 =3 | |
| 14 Marshall, Frank | 46.7% +115 -145 =190 210/450 | +2 -12 =11 | +4 -21 =29 | +0 -7 =7 | +9 -11 =15 | +5 -6 =8 | +4 -11 =10 | +3 -3 =6 | +6 -7 =14 | +0 -7 =10 | +2 -2 =4 | +1 -1 =3 | +8 -6 =14 | +7 -13 =18 | +4 -2 =6 | +5 -0 =1 | +35 -25 =17 | +2 -1 =3 | +13 -10 =8 | +5 -0 =6 | |
| 15 Reti, Richard | 46.4% +81 -101 =97 129.5/279 | +0 -4 =2 | +1 -5 =2 | +1 -3 =5 | +4 -9 =8 | +1 -5 =5 | +1 -2 =8 | +2 -2 =1 | +2 -1 =0 | +14 -16 =17 | +7 -17 =4 | +5 -3 =8 | +16 -12 =12 | +5 -2 =4 | +2 -4 =6 | +3 -3 =1 | +4 -0 =1 | +4 -7 =6 | +3 -2 =2 | +6 -4 =5 | |
| 16 Thomas, Sir George | 38.0% +27 -65 =66 60/158 | +0 -1 =0 | +1 -6 =4 | +0 -7 =6 | +1 -5 =4 | +1 -3 =3 | +3 -1 =5 | +0 -2 =5 | +0 -1 =0 | +3 -9 =10 | +0 -5 =4 | +0 -3 =3 | +1 -2 =3 | +1 -1 =3 | +0 -5 =1 | +3 -3 =1 | +1 -0 =1 | +1 -3 =1 | +2 -2 =2 | +9 -6 =10 | |
| 17 Janowski, David | 37.9% +71 -134 =56 99/261 | +4 -26 =7 | +1 -9 =1 | +2 -4 =2 | +3 -5 =0 | +0 -3 =2 | +5 -9 =5 | +1 -5 =0 | +5 -4 =4 | +2 -3 =2 | +2 -1 =0 | +1 -0 =1 | +2 -4 =1 | +5 -9 =3 | +25 -35 =17 | +0 -4 =1 | +0 -1 =1 | +1 -0 =0 | +12 -8 =5 | +0 -4 =4 | |
| 18 Saemisch, Fritz | 37.6% +31 -76 =74 68/181 | +0 -0 =1 | +1 -1 =0 | +0 -11 =4 | +0 -6 =6 | +1 -10 =0 | +3 -2 =2 | +0 -0 =1 | +0 -1 =1 | +1 -5 =12 | +6 -15 =9 | +2 -2 =11 | +2 -7 =8 | +0 -4 =3 | +1 -2 =3 | +7 -4 =6 | +3 -1 =1 | +0 -1 =0 | +2 -2 =4 | +2 -2 =2 | |
| 19 Mieses, Jacques | 35.6% +57 -134 =76 95/267 | +0 -8 =5 | +0 -4 =0 | +0 -4 =0 | +7 -15 =6 | +2 -5 =0 | +0 -10 =8 | +2 -4 =1 | +6 -16 =8 | +3 -5 =6 | +1 -3 =3 | +0 -2 =1 | +5 -9 =8 | +5 -14 =8 | +10 -13 =8 | +2 -3 =2 | +2 -2 =2 | +8 -12 =5 | +2 -2 =4 | +2 -3 =1 | |
| 20 Yates, Frederick | 33.9% +39 -111 =74 76/224 | +0 -2 =1 | +0 -8 =3 | +2 -11 =3 | +2 -8 =8 | +1 -5 =4 | +2 -10 =2 | +1 -4 =2 | +0 -2 =1 | +3 -14 =10 | +3 -9 =2 | +2 -5 =3 | +2 -6 =4 | +2 -3 =3 | +0 -5 =6 | +4 -6 =5 | +6 -9 =10 | +4 -0 =4 | +2 -2 =2 | +3 -2 =1 |
Finally I would like to emphasise that it's a mistake to forget the chess of the past just because there's so much good chess being played now. Some people feel that the old masters played a comparitively boring style of chess that's just not worth bothering with. But in my opinion that's just not true. The old masters produced any number of sparkling and exciting games. I picked basically one random game from the 2757, one I hadn't seen before (I don't think) and I think it bears out my theory.
Capablanca, Jose Raul - Bogoljubow, Efim
London BCF Congress 1922
1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.O-O Be7 6.Re1 b5 7.Bb3 d6 8.c3 O-O 9.d4 exd4 10.cxd4 Bg4 11.Be3 Na5 12.Bc2 Nc4 13.Bc1 c5 14.b3
Moves are clickable
All of this is still seen today, although of course it is a sideline as White normally plays h3 earlier to prevent ...Bg4 14...Na5 (14...Nb6 is more common today ) 15.Bb2 Nc6 16.d5 Nb4 17.Nbd2 Nxc2 18.Qxc2 Re8 19.Qd3 h6 20.Nf1 Nd7 21.h3 Bh5 22.N3d2 Bf6 23.Bxf6 Qxf6 24.a4 c4 25.bxc4 Nc5 26.Qe3 bxa4 27.f4! White's plan involves a big space advantage on the kingside. Black has obvious chances on the other wing 27...Qe7 28.g4 Bg6 29.f5 Bh7 30.Ng3 Qe5 31.Kg2 Rab8 32.Rab1 f6 33.Nf3 Rb2+ 34.Rxb2 Qxb2+ 35.Re2 Qb3 36.Nd4
36...Qxe3 (36...Qxc4 was better, eg 37.Rc2? Qxd5! ) 37.Rxe3 Rb8 38.Rc3 I must admit I don't understand this move. My computer wants to play it too, so unsurprisingly perhaps Capablanca probably knew what he was doing! 38...Kf7 39.Kf3 Rb2 40.Nge2 Bg8 Black's main problem has been that his bishop was buried and needed a bunch of moves to escape, this becomes a bigger problem as pieces are exchanged, so Bogoljubov resolves to start getting the bishop back in the game. 41.Ne6! A monster obviously, and if Black exchanges it off White has the option of permanently excluding the bishop from the game. 41...Nb3 Now I think I understand 38.Rc3. Capablanca realised that the knight wasn't staying on c5 forever 42.c5! A decisive breakthrough? 42...dxc5 43.Nxc5 Nd2+ 44.Kf2 Ke7 ( The Computer wants to play 44...Nb1 but even then it finds some truly insane tactics to win the game for White. For example 45.Rc1 a3 46.d6 a2 47.Ne6 threatening Rc7+ and Re7 mate 47...Ke8 48.Rc7 Bxe6 49.fxe6 Rb8 (only move) 50.Rxg7 Kf8 51.Rf7+ Kg8 52.Rb7 Rxb7 53.d7 a1=Q 54.e7
A remarkable position indeed. White has given up a queen and a rook but Black is utterly helpless! Of course Capablanca didn't see all of this. But it is clear he did back his famous positional judgement, which told him the united forces of his rook, two good knights and united central pawns were better than Black's slightly discombobulated equivalents, and the silicon monster seems to back this up. ) 45.Ke1 Nb1 46.Rd3 a3 47.d6+ The situation clarifies. White is faster 47...Kd8 48.Nd4 Rb6 49.Nde6+ Bxe6 50.fxe6
Pawns abreast on the 6th rank, almost as good as on the 7th rank in the computer variation! 50...Rb8 51.e7+ Ke8 52.Nxa6 Nc7 is coming next, so Black resigns 1-0