Test Tournament

Organization: EGO
Place: Utrecht, Ducosim Convention
Date: February 15, 2003
Experiment "GIPF with handicap"
Format: Swiss system (double rounded) , 3 x 2 rounds, 15 minutes per player/game
Participants: 6


Handicap system for GIPF ?

The aim is to find out whether it is possible to set up a handicap system for GIPF (i.e. similar to the one that is used in go). To start the tests, playing with less or more pieces than the regular 18 per player was restricted to minimally 15 and maximally 21 pieces per player.

Rating difference / handicap
50-100 : 1 piece
101-200 : 2 pieces
201-300 : 3 pieces
301- ... : 4 pieces

Participants
Patrick Van de Perre (1765)
Angelo Wentzler (1475)
Boris Ammerlaan (1465)
Fred Kok (1415)
Hennes Passman (1312)
Jan Willem Knopper (1302)

Round 1
Patrick Van de Perre (16 pcs) - Fred Kok (20 pcs) : 1-0
Angelo Wentzler (17 pcs) - Jan Willem Knopper (19 pcs) : 1-0
Boris Ammerlaan (17 pcs) - Hennes Passman (19 pcs) : 0-1

Round 2
Fred Kok (20 pcs) - Patrick Van de Perre (16 pcs) : 0-1
Jan Willem Knopper (19 pcs) - Angelo Wentzler (17 pcs) : 1-0
Hennes Passman (19 pcs) - Boris Ammerlaan (17 pcs) : 0-1

Round 3
Patrick Van de Perre (16 pcs) - Angelo Wentzler (19 pcs) : 0-1
Boris Ammerlaan (17 pcs) - Jan Willem Knopper (19 pcs) : 1-0
Hennes Passman (19 pcs) - Fred Kok (17 pcs) : 1-0

Round 4
Angelo Wentzler (19 pcs) - Patrick Van de Perre (16 pcs) : 1-0
Jan Willem Knopper (19 pcs) - Boris Ammerlaan (17 pcs) : 0-1
Fred Kok (17 pcs) - Hennes Passman (19 pcs) : 0-1

Round 5
Angelo Wentzler (17 pcs) - Boris Ammerlaan (18 pcs) : 0-1
Hennes Passman (20 pcs) - Patrick Van de Perre (16 pcs) : 0-1
Jan Willem Knopper (18 pcs) - Fred Kok (17 pcs) : 1-0

Round 6
Boris Ammerlaan (18 pcs) - Angelo Wentzler (17 pcs) : 1-0
Patrick Van de Perre (16 pcs) - Hennes Passman (20 pcs) : 1-0
Fred Kok (17 pcs) - Jan Willem Knopper (18 pcs) : 0-1

   Name                  ELO   Score
1. Boris Ammerlaan       1465    5
2. Patrick Van de Perre   1765    4
3. Hennes Passman        1312    3
   Angelo Wentzler       1475    3
   Jan Willem Knopper    1302    3
6. Fred Kok              1415    0

Some first impressions:

- First conclusion: Patrick Van de Perre didn't win the tournament, although he was by far the strongest player. He played with, respectively, 3 and 4 pieces less than his opponents, yet managed to win 4 games. Hennes Passman, according to his rating the weakest player, succeeded in winning 3 games.
- The games were played with 15 minutes per player/game, and that appeared not to be sufficient for reliable tests. Some players - in particular the less experienced - struggled more with the time than with their respective opponents. Ranking tournaments are normally played with minimally 20 minutes per player. Players should be able to play close to their best to come to somewhat reliable results and for that purpose more time per player/game is needed.
- For this tests, the valour of one piece was estimated to be 100 elo-points. Ask for there opinion after the tournament, most of the participants thought that the compensation of one piece for each 100 elo-points difference was too much.
- Many (MANY !!!) more games will have to be played by many (MANY !!!) more players to find out the approximate valour of a piece in terms of Elo-points. And even if we find that out some day, we'll still have to wait and see whether it will make sense to translate that knowledge into some sort of handicap system. Anyway, Angelo Wentzler has already announced to go on with the tests within the GIPF-club in Eindhoven.