One of most common problems emerging at the round robin tournaments is what tie-break should be used to rank players (or teams) level on game points. Multitude of methods are in use but one called "The Sonneborn-Berger" or just "Berger" is by far the most popular. It is counted as a sum of scores of all opponents a player defeated plus half of scores of all players he has drawn with. I asked myself once why is it called "Sonneborn-Berger" and what was the origin of the system. I had to do some research on the issue in the dust-covered volumes of old chess magazines. As is happened (not only in this case, I believe) the history proved not to be the fair judge.

The system was actually invented by Viennese master of Czech origin Oscar Gelbfuhs (1852-1877) who, while he participated in the tournament in Vienna in 1873, proposed that the final ranking of the players should be judged according (more or less) to what we use to call "Sonneborn-Berger" nowadays. He believed that final order should be decided by the sum of points of beaten opponents plus half of a sum of points of opponents held to a draw; this divided by number of games played (which makes no difference anyway in an all-play-all tournament where everybody plays the same number of games). That player scoring less game points could be placed above another player scoring more game points, was actually a further merit for him, as he thought casual losses vs patzers should not bias player's overall score by so much. This was perhaps justified to some extent at his time since the skills of best players in a typical round robin tournament used to differ immensely from those of the dawdlers. Almost the same idea was raised by Hermann Neustadtl (1962-1909), a doctor from Prague in a letter he sent to the "Chess Monthly" editorial office in 1882. Basing on Gelbfuhs' ideas he only amended to remove the "number of games" factor, actually becoming the first man to formalise the method of tie-breaking in its contemporary shape.

Gelbfuhs and Neustadtl's ideas were rigidly criticised by William Sonneborn (1843-1906), the bank officer from London, in the columns of selfsame "Chess Monthly" magazine some four years after Neustadtl's letter. He considered the proposed system as defective since it allowed players be ordered inconsistently with the game point order. That's why he suggested that Neustadtl's score should be augmented with the square of number of game points scored by a player. The system however was quite complex and not very transparent and as a matter of fact it did not contribute much to the essence of the problem. This is why it never gained popularity and was quickly forgotten. Of course the system was forgotten, but not Sonneborn's name!

Now, what is Johann Berger (1845-1933, see photo), the Austrian player, theoretician and editor doing around here? The truth is, he added virtually nothing creative to the discussion, just blatantly added his own name to that of Sonneborn only to popularise the Neustadtl's system (successfully, as it turned out after decades) as the "Sonneborn-Berger"! I was not surprised at all hearing it given that the Americas are named after Amerigo Vespucci, a man to be first to describe a continent (bugging it with many lies by the way!), and not e.g. Columbus who put a foot on the American soil years before Vespucci did.

Anyway, to cut a long story short (as Jeffrey Archer would love to say):
- what we know today should be named as "Gelbfuhs-Neustadtl" rather than S-B;
- Sonneborn put a dose of criticism on an idea named after him at the end of the day and proposed his own system, given up soon;
- Berger was crafty enough to add his name to the discussion, smart enough to popularise it, and lucky enough to bear shorter name than Sonneborn, which is why the system's name is commonly truncated to "Berger" - a name that contributed least...

C'est la vie!

Yours,
Wojtek

* * *