OxfordCroquet Logo

Dr Ian Plummer

Donation Button

http://oxfordcroquet.com/tech/nel-cb/index.asp
Technical
Cross-Block Peer Games are Bad

by Louis Nel
March 2001

I've seen numerous tournaments in which the usual block games (i.e. Round Robin format) are supplemented by giving each player an extra "cross-block" game against a similarly skilled opponent chosen from another block. At first glance this may seem a good idea, but it is not. It undermines the purpose of block play to sort players into the appropriate pecking order according to performance level.

Consider two idealized blocks of 5 players, with accurate Grades as shown:

Block A

Block B

A1

2400

B1

2400

A2

2300

B2

2300

A3

2200

B3

2200

A4

2100

B4

2100

A5

2000

B5

2000

In ordinary block play (i.e. Round Robin format applied to each block separately), A1 has the highest probability to win Block A, not because the format gives him preferential treatment, but because he happens to have the highest skill level. The symmetric Round Robin format gives every player an equal opportunity to prove his superior performance level to the extent that it is superior.

Now consider what happens if these same players were to play under the following extreme form of block play: the Single-peer-game format.

A1 to play one game against B1
A2 to play one game against B2
A3 to play one game against B3
A4 to play one game against B4
A5 to play one game against B5

and these are the ONLY block games to be played.

In this format, every player has a 50% probability of winning his only block game and thus to end in first place (possibly tied). This will surely not give an appropriate ordering of players according to performance level.

While no one would dream of using this Single-peer-game as a standalone format, it is often added in tournaments to ordinary block play format. When that is done, its influence on the results is camouflaged and overshadowed by the ordinary block games, but its influence remains present. Its thrust is toward equalizing winning probability irrespective of performance level. This is an inappropriate influence. It could result in play-off positions which undeservedly reward or punish players.

One cannot fault the motivation underlying these isolated cross-block games: to let players have more games or to ensure that every player gets to play at least one winnable game. But there are better ways to accomplish that. One example is the Swiss format, where the stronger players quickly gravitate upward and play each other while the weaker players gravitate downward and play each other, with nobody getting eliminated. A future article will say more about the Swiss format.

A cross-block format (where every player in block A plays every player in block B), where the players in block A are almost equal to the players in block B, becomes close to the situation where a peer-game is added to an ordinary block. It is therefore likewise undesirable.

Louis

Author: Louis Nel
All rights reserved © 2010


Updated 28.i.16
About, Feedback
oxfordcroquet.com/tech/nel-cb/index.asp
on www.oxfordcroquet.com
Hits: 4849