"Buchholz" tiebreaker does not implement original Buchholz

[mbergen]

What's implemented as "Buchholz" tiebreaker in https://github.com/Liquipedia/Lua-Modules/blob/main/lua/wikis/commons/Standings/Tiebreaker/Buchholz.lua is not following the original definition of Buchholz, by substracting the losses from the score.
However, this version is apparently used on Dota, R6 and CS.

The original Buchholz (which is used on Chess, AoE in manual fashion) does not do that, and instead only counts wins as 1, draws as 0.5 points.

[hjpalpha]

option 1: rename current buchholz implementation to fakeBuchholz (or something similar) and bot current usage to fit that, after that implement proper buchholz for the buchholz tie breaker

option 2: add an additional tie breaker called properBuchholz or origBuchholz

personally i think option 1 is more appropriate as with that we have the name right instead of calling the fake buchholz buchholz

[mbergen]

fwiw centaur named this variant "esports buchholz": https://discord.com/channels/93055209017729024/1407714004767473685/1437152129721700504
IMO we could go with option 1 and that name.

[ElectricalBoy]

fwiw centaur named this variant "esports buchholz": https://discord.com/channels/93055209017729024/1407714004767473685/1437152129721700504
IMO we could go with option 1 and that name.

tend to agree

[hjpalpha]

fwiw centaur named this variant "esports buchholz": https://discord.com/channels/93055209017729024/1407714004767473685/1437152129721700504 IMO we could go with option 1 and that name.

okay for me

[mbergen]

@Rathoz do you have an opinion about this?

[Rathoz]

Outcome is the same right (relative numbering), if played same amount of opponents?

W=1
D=0 or 0.5
L=-1 or 0

[mbergen]

The outcome is the same if everyone played the same amount of matches, yes.
But for one this isn't always the case, for the other this would mean our display would differ from e.g. official brackets on the occasions where the default variant is used. While it might be acceptable on esports wikis, i don't think chess would accept it.