EGF Winning Statistics
The table shows statistics on winning a game with the opponent that is one (G+1), two (G+2), three (G+3) and four (G+4) grades stronger than the player whose strength is idicated in the first column. The data were collected from all events included in GoR tournament database.
G + 1 G + 2 G + 3 G + 4
G Nw Ng Pw Nw Ng Pw Nw Ng Pw Nw Ng Pw --- ----------------- ----------------- ----------------- ----------------- 20K 811 2455 33.0 543 1632 33.3 166 715 23.2 85 437 19.5 19K 406 1040 39.0 158 526 30.0 69 237 29.1 31 164 18.9 18K 543 1286 42.2 267 695 38.4 146 407 35.9 37 160 23.1 17K 481 1065 45.2 373 830 44.9 99 266 37.2 38 184 20.7 16K 625 1359 46.0 278 659 42.2 86 284 30.3 46 198 23.2 15K 637 1501 42.4 354 854 41.5 147 393 37.4 57 197 28.9 14K 582 1288 45.2 305 770 39.6 97 278 34.9 69 254 27.2 13K 631 1424 44.3 340 774 43.9 167 408 40.9 57 191 29.8 12K 684 1505 45.4 531 1244 42.7 157 373 42.1 45 217 20.7 11K 899 1916 46.9 388 952 40.8 114 318 35.8 43 178 24.2 10K 1176 2623 44.8 568 1410 40.3 135 501 26.9 87 356 24.4 9K 1186 2638 45.0 567 1351 42.0 162 498 32.5 52 277 18.8 8K 1260 2845 44.3 637 1636 38.9 152 504 30.2 61 282 21.6 7K 1474 3298 44.7 660 1718 38.4 157 510 30.8 60 271 22.1 6K 1725 4036 42.7 753 1892 39.8 207 640 32.3 75 350 21.4 5K 1989 4461 44.6 918 2355 39.0 188 689 27.3 89 454 19.6 4K 2279 5041 45.2 913 2327 39.2 216 806 26.8 68 418 16.3 3K 2430 5660 42.9 938 2859 32.8 184 848 21.7 52 387 13.4 2K 2583 6354 40.7 852 2692 31.6 135 780 17.3 39 421 9.3 1K 3117 7884 39.5 750 2951 25.4 196 1302 15.1 53 627 8.5 1D 2780 7399 37.6 930 3752 24.8 237 1657 14.3 46 860 5.3 2D 2199 6357 34.6 722 3164 22.8 162 1405 11.5 20 544 3.7 3D 2051 5928 34.6 715 3556 20.1 118 1511 7.8 1 159 0.6 4D 1469 4807 30.6 355 2661 13.3 10 303 3.3 0 0 - 5D 914 3287 27.8 35 559 6.3 0 0 - 0 0 - 6D 103 694 14.8 0 0 - 0 0 - 0 0 -
Number of games included in statistics: 155189
Notation: Nw - number of wins Ng - number of games Pw = Nw/Ng
An important thing to note here is that these statistics are inherently unreliable!
Because we use the McMahon system so much, there is a strong sorting effect at work in these statistics. If, for example, a 1k plays against a 2d, then in a McMahon tournament the 1k has already won two more games than the 2d. So probably it is a strong 1k against a weak 2d. This effect inflates all these numbers significantly.
The only area where this effect does not apply is at the top, because players that are above the top bar in a tournament will play each other regardless of their rank, as they start with the same McMahon score. This effect is very noticeable in the table above, where the percentages suddenly start dropping significantly among (stronger) dan players.
EDIT: Other effect at work that bias these statistics:
- 7D and 6D have been used in the past for professional players or very strong Asian amateurs, who might be two or three stones stronger than the average European amateur 6D and 7D players
- 20K is the lowest rank accepted by the database, and also contains the results of players 21K and weaker.
The most reliable statistics are probably those between players in the range 2D-5D, where the players are almost guaranteed to be above the bar, but without interference from the above effect of strong Asian amateurs, so that either effect has a relatively small impact.