PDA

View Full Version : Sonneborn-Berger tie-break in Vega

Andrew Hardegen
30-12-2019, 05:09 AM
I have just noticed what seems to be a small bug in Vega's calculation of Sonneborn-Berger scores. Specifically, half-point byes are treated as wins against the virtual opponent, rather than draws.

13.15.2. For tie-break purposes a player who has no opponent will be
considered as having played against a virtual opponent who has
the same number of points at the beginning of the round and
who draws in all the following rounds. For the round itself the
result by forfeit will be considered as a normal result.

This gives the formula:

Svon = SPR + (1 – SfPR) + 0.5 * (n – R)

where for player P who did not play in round R:
n = number of completed rounds
Svon = score of virtual opponent after round n
SPR = score of P before round R
SfPR = forfeit score of P in round R

Here is a Vega output (Vega 8.2.1):

2019 Southern Suburbs Championship
Cross Table at round 7 sorted by score

Pos NAME Rtg T Fed Pts | 1 2 3 4 5 6 7 | DirE Mwns MBlk Buc1 BucT S-B
---------------------------------------------------------------------------------------------------------------------------------------------------
1 Hardegen,Kathryn 2045 WFM WA 6.5 | +B12 +W11 +W5 =B2 +W4 +B6 +W7 | 0.5 6 3 27.5 31.0 27.75
2 Ferozkohi,Ihsan 2309 WA 6.5 | +W18 +BYE +B6 =W1 +W10 +B4 +B5 | 0.5 6 3 27.0 30.0 26.75
3 Holland,Dennis 1957 WA 5.0 | +W7 =B19 +W13 -B4 =W15 +B16 +W12 | 0.0 4 3 22.0 24.5 17.00
4 Kuan,Han Sen 2105 WA 4.5 | =BYE +B17 +W19 +W3 -B1 -W2 +B15 | 0.0 4 3 28.0 30.5 17.50
5 Hare,Tim 2175 WA 4.5 | +B9 +W10 -B1 +W11 +W8 =B15 -W2 | 0.0 4 3 27.5 31.0 16.25
6 Calado,Edward 2070 WA 4.5 | +W23 +B16 -W2 =B15 +W19 -W1 +B10 | 0.0 4 3 25.5 27.5 12.75
7 Prunty,Jon 1508 WA 4.0 | -B3 -W9 +B24 +W22 +B11 +W8 -B1 | 1.0 4 4 24.5 25.5 10.50
8 Ayling,Simon D 1922 WA 4.0 | +B14 -BYE +W9 +W12 -B5 -B7 +W16 | 0.0 4 3 23.5 26.5 13.50
9 Gao,Oscar 1596 WA 3.5 | -W5 +B7 -B8 -W17 +B20 +W22 =B11 | 0.0 3 4 21.5 23.5 10.25
10 Soutar,Gavin 1805 WA 3.5 | +BYE -B5 +W23 +W14 -B2 =B12 -W6 | 0.0 3 3 25.5 27.5 10.25
11 De Heer,Alex 1643 WA 3.5 | +W22 -B1 +W17 -B5 -W7 +B13 =W9 | 0.0 3 3 25.0 27.0 10.25
12 Sheridan,Wallace 1515 WA 3.5 | -W1 +B22 +W16 -B8 +BYE =W10 -B3 | 0.0 3 3 25.0 27.0 9.75
13 Andrew,Christopher-Jack 959 WA 3.5 | =BYE +W20 -B3 -B16 +W21 -W11 +B18 | 0.0 3 3 20.5 23.0 11.50
14 Hu,Tommy 1500 AUS 3.5 | -W8 +B18 +W21 -B10 =BYE -- +W20 | 0.0 3 2 20.5 23.0 11.50
15 Geronimo,Brent 1778 WA 3.5 | -W17 +B21 +BYE =W6 =B3 =W5 -W4 | 0.0 2 2 24.5 27.0 12.50
16 McAteer,Ian 1736 WA 3.0 | +B24 -W6 -B12 +W13 +B17 -W3 -B8 | 0.0 3 4 23.5 24.5 7.50
17 Brown,Amy 1248 WA 3.0 | +B15 -W4 -B11 +B9 -W16 -W18 +B24 | 0.0 3 4 21.0 22.0 8.00
18 Simonds,Marcus 1632 WA 3.0 | -B2 -W14 -B20 +BYE +W25 +B17 -W13 | 0.0 3 3 20.5 21.5 5.50
19 Gunawan,Tjermin (W) 1668 WA 2.5 | +B25 =W3 -B4 +W20 -B6 -- -- | 0.0 2 3 24.0 25.0 6.00
20 Gough,Dylan 1688 WA 2.5 | =W21 -B13 +W18 -B19 -W9 +BYE -B14 | 0.0 2 3 18.5 20.5 6.25
21 Geronimo,Harry 1131 WA 2.5 | =B20 -W15 -B14 +BYE -B13 -W24 +BYE | 0.0 2 3 16.5 17.5 4.75
22 Zhang,Emily 596 WA 2.0 | -B11 -W12 +BYE -B7 +W24 -B9 -BYE | 0.0 2 3 19.5 20.5 3.00
23 Gunawan,Marco (W) 1582 WA 2.0 | -B6 +W24 -B10 +W25 -BYE -- -- | 0.0 2 2 19.5 20.5 2.00
24 Morrison,Robert 1136 WA 1.0 | -W16 -B23 -W7 -BYE -B22 +B21 -W17 | 0.0 1 3 17.0 19.0 2.50
25 Howard,Bryan (W) 1129 WA 1.0 | -W19 +BYE -BYE -B23 -B18 -BYE -- | 0.0 1 2 16.5 18.5 2.50

Generated by Vega - www.vegachess.com
(copy licensed to: Perth Southern Suburbs Chess Club - AUS)

#4 (Kuan, Han Sen) had a half-point bye in Round 1 which, at the close of the tournament, should be treated as a draw against a virtual opponent with score 3.5/7. We can manually calculate Kuan's Sonneborn-Berger score as follows:

S-B = 0.5 * 3.5 + 1 * 3.0 + 1 * 2.5 + 1 * 5.0 + 0 * 6.5 + 0 * 6.5 + 1 * 3.5 = 15.75.

Whereas Vega gives S-B = 17.5, which is what we would get if we replace the Round 1 term in the above sum by 1 * 3.5. Other examples within this tournament also show this to be happening.

forlano
08-01-2020, 07:17 AM
Hello,

Without to see the file and how the tournament was set is difficult to say anything. However, before to do that sum there is a preliminary task that should be done and it seems you have not performed:
You need to ADJUST the score used for the tie-break. In fact there are many unplayed games and all need to be considered DRAW. Only then you can do the sum.
For example Gunawan has 3.5 tiebreack score and not the real 2.5, and so on.

Here http://www.vegachess.com/ns/node/22 you can find a powerpoint that shows how to apply the rule (bottom page kallithea rule).
Unfortunately the document is in Italian, however from slide 8 the example should be understood without problems.

Please send me the file of that tournament and I'll inquire better.
Thanks,
Luigi

Bill Gletsos
21-01-2020, 10:51 AM
This has been fixed in version 8.2.3 released 19th January.

21-01-2020, 09:50 PM
This has been fixed in version 8.2.3 released 19th January.Hi Bill,
The version available for download from http://www.vegachess.com/ns/node/4 is still 8.2.2. How does one go about getting 8.2.3? I am using Windows 64 bit. Thanks.

UPDATE: I just saw a Vega thread with latest releases where this query might better belong, but it does not answer my question.

Bill Gletsos
21-01-2020, 10:01 PM
Hi Bill,
The version available for download from http://www.vegachess.com/ns/node/4 is still 8.2.2. How does one go about getting 8.2.3? I am using Windows 64 bit. Thanks.

UPDATE: I just saw a Vega thread with latest releases where this query might better belong, but it does not answer my question.Why do you claim it is still 8.2.2?