Skip to content
Little GolemMake your next move.

Rating stats

Public forum discussion.

Einstein

#1 · YHW

4 May 2011

Hi,

I have constantly monitored the games of the best 50 players in the last year . The idea was a kind of an all-time tournament in which all games of the best 50 against each other has to be considered in order to encourage more people to play against other players, because 4 players in the current top 10 are inactive or other players with a high number of points have only played not more than 100 matches.

The yearly championship seems inappropriate to determine the "best player" because the influense of luck is too high in a single tournament. So that I wanted to monitor all games of the players to get a better estimation.

My stats are including all winning quotes against each other. If you would be interested to take a look on it, please send me a message with your e-mail adress. you would need a current Excel version to view the stats.

After analyzing the stats, the winning quotes of the most players are different but relative close to each other except of 2 signifcant steps between 'the bots' and the 'best players' as well as between the best and the rest - the 'better players'.

It seems so that hanfried_c is really the best bot/player followed by Roro.

The next 'league is given by naive_child_c, Telestes, MarleysGhost, kitaktus, luiser, wurscht, hubbie_c.

While taking the fact into account that the most players in this 'league' are inactive, it is only represented by naive_child_c, kitaktus. So that kitaktus should really be the best active human player.

All other players are in a 'lower league'.

Another but similar rating arisis while using a point system to determine the 'best player' due to the current winning quotes of a specific player against another one in the top 50. According to the points in a tournament, the player get 2 points while the quote against another player is higher than 50% and 0 points if the quote is lower than 50%. The fact that 2 players haven't played against each other and/or the quote is equal (50%) is considered by 1 point.

Roro would reach the averaged points of 1.77 against all players in the top 50 followed by hanfried (1.65 points). So that Roro should be the winner of an all time tournament if the tournament should be finished ever.

Currently, while summarizing all reached points RoRo is in front because that unbelievable number of played games (hanfried gets 1 point for all players that it hasn't played), the 2nd best player would be luiser. Followed by YHW, hanfried_c, John-Marc 'Johnny' McGregor, FatPhil, ypercube, Pawel Grabowski, TUMRAK, Bernhard Herwig, Jonny, opmp, Ray Garrison, kitaktus, ...active players are advantaged while using such a point sytem.

As I saif, if you would be interested in your stats, send me a message. Included are the following players:

hanfried_c

MarleysGhost

Telestes

wurscht

naive_child_c

RoRoRo the Bot

hubbie_c

luiser

kitaktus

YHW

idiot

ypercube

opmp

Eiko Bleicher

Ray Garrison

LeCanardfou

Henrik Sjøl

Jonny

JJ10

Bernhard Herwig

Richard Pijl

kfiecio

FatPhil

Figilano

Dvd Avins

TUMRAK

Pawel Grabowski

Andreas Bleck

Rob

John-Marc 'Johnny' McGregor

Disciple

Tore S

zwack

Manou

Hartmut

Hjallti

fhsIV

Nagy Fathy

Jaakko Pirinen

vic

richyfourtytwo

Looser

Mojmir Hanes

klaashaas

Carroll

wojtex

graff

DukeOfPrunes

Dom

Ralf

Martin Horatschek

duvelman

Luca Bruzzi

bachkiesel

Micco

antony

duplolas

Matteo A.

Diamante

Ragnar Wikman

George

Aganju

kp4to

BigChicken

FCGerard'91

rafi

funtobewith11

conillet

neptun

Charles

VinceS

grozan

#2 · MarleysGhost

5 May 2011

> The next league is given by ...MarleysGhost...

I suspect there's some kind of sample bias here. Early on I may have had a high winning percentage because I had a low rating and was playing rating tournaments.

#3 · somar96

5 May 2011

Weren't I included because of my high timeout rate (March 2010)? I have played quite a lot against the people up there.

#4 · YHW

5 May 2011

The rating/stats are based on all games on your individual 'games' (EWN) page. A difference between a normal lost and a timeout is not considered.

#5 · YHW

5 May 2011

@MarleysGhost

although you have finished only a low number of matches (69) and that you are not active for a year your stats are still one of the best (70% winning quote aggainst all opponents). your number of matches is shown also that you are not active, but you are still in.