An Unscientific Observation

Class Players Then and Now

     I was disappointed at the number of draws in the recently concluded London tournament. It seems nobody could beat anybody else. The results looked like an engine correspondence tournament! 
     Out of curiosity I looked at some older tournaments like Linares and the draw percentage in those events was pretty high, too. Then I went back to AVRO, 1938 where the best players in the world battled it out: Keres, Fine, Botvinnik, Euwe, Reshevsky, Alekhine, Capablanca and Flohr. Out of 56 games they drew 57 percent. Things have gotten worse, but then with big money at stake and the players being better, you can't expect much else. 
     Everybody knows you can't compare the players of the past against the players of today based on ratings. Fischer's rating, for example, only tells us how he performed against his contemporaries, not how well he would have performed in, say, the London tournament.
     Mark Glickman of the USCF’s ratings committee said, “You get into dangerous territory when estimating ratings of the past.” He added, “It is possible to say how much better Capablanca was than his contemporaries. If it were possible to put Capablanca in a time machine and transport him to this time, it would be virtually impossible to predict how strong he would be. Modern players have access to computers and games databases so they know more than their predecessors and are better than them, but the argument is that for a fair comparison, a player from another era would have to have the same resources as players of today and really, nobody knows what would happen."  Also, see the article Who was The Strongest Chess Player of All Time? 
     It makes sense to me that today's players are playing at a much higher level because of a variety of factors, but what about class players? Are today's 1800s better than 1800s of 50 years ago? 
     The other day I played over a rather ordinary game by a couple of 1800 rated players in a weekend tournament and it seemed to be pretty decently played. The game was a 51 move English opening featuring a minor piece ending and I got to wondering how they might compare to 1800s of 50 years ago. 
     Of course, like comparing grandmasters and world champions of different eras, there is no way to tell for sure. I selected four random games by players rated in the 1800s from recent tournaments and four games played by players from the 1960s that I know were also rated in the 1800s. The only criteria I had were that they had to be about the same length (45-60 moves) and similar openings, English and QP openings. My guess is that had I selected highly tactical openings the results would have been much different! 
     What I looked at was: 1) how soon they departed the Fritz 12 opening book, 2) how many changes in evaluation there were that Stockfish 6, at 10 seconds a move, recommended that fell between 0.25 and 0.50 and 3) moves where the evaluation jumped more than half a Pawn. 
     Although this survey wasn't very accurate, what I found was there wasn't that much difference in percentages between the 1800s of today and 60 years ago. The average move where the book was left was move 7. That seemed a bit early, but when I checked about a dozen of my OTB games from the 1960s they left the book between moves 5 and 12. If nothing else, it tells me that maybe we average players do spend too much time on the openings because most of our games are not going to follow the book very far anyway.
     Twelve percent of the moves were evaluated at being over a half Pawn worse and 15 percent were between a quarter and a half Pawn worse than Stockfish's recommended move. So in a 50 move game, about 6 moves will be serious mistakes while 7-8 moves, while not bad, are not the best either. 
     A quick observation was that those 4-6 moves per game where the evaluation jumped over a half a Pawn were usually because of a tactic that was missed. Way back when, the study of tactics wasn't emphasized like it is today, but today's emphasis on tactical study doesn't seemed to have helped the average player much. We still can't see them! 
     It also suggests to me that despite the glut of chess material that's available to us today, usually at high prices, we average players have not, unlike world championship class players, gotten much better. So, instead of paying $25-30 for a “modern” chess book or even more for training DVDs, one might as well buy Dover reprints of old chess books on Amazon for a fraction of the price; they'll give the same results.