I agree here. I only know ELO systems very superficially and we also don't know how exactly Relic implemented their system, but shouldn't the ELO adjustment be made based on the team's winning chance?
Say you're a top player with only low skill players in your team vs a team of "normal" opponents, giving you a calculated win chance of 50%, shouldn't the ELO rating after the game adjust as much as if this was a match of "worthy" teammates and opponents? It shouldn't take individual ELO into account and punish the top player overly for losing the game.
Asking out of curiosity here, because this is how I always understood the system.
Excellent point.
The ELO system, in and of itself, is just a way of removing x amount of points if you lose and you are the top dog and y amount of points (x>y) if you lose and you were the underdog. The thing is however, that the ELO system makes mathematical sense only if you can prove that the performance of every COH2 player can be modelled by the Normal "Bell Curve" Distribution. A further assumption is necessary because COH2 performance in the above sense is still not measurable. One cannot look at a sequence of clicks, orders or doctrine choice and derive a number to represent that player's skill. Performance can only be inferred from wins and losses. Therefore, if a player wins a game, they are assumed to have performed at a higher level than their opponent for that game. Conversely, if the player loses, they are assumed to have performed at a lower level. To simplify computation even further, Elo proposed a straightforward method of estimating the variables in his model (i.e., the true skill of each player). One could calculate relatively easily from tables how many games players would be expected to win based on comparisons of their ratings to those of their opponents. The ratings of a player who won more games than expected would be adjusted upward, while those of a player who won fewer than expected would be adjusted downward. Moreover, that adjustment was to be in linear proportion to the number of wins by which the player had exceeded or fallen short of their expected number.
Now here is where the retardedness begins in COH2.org:
Normal distribution cannot accurately model the performance of lower ranked players. Every statistical organization worth its salt nowadays uses logistic distribution to model the performance of games. Even FIDE does (the chess cup).
Nobody uses the original ELO system anymore, because contrary to what most people here believe, "get good" cannot overcome basic statistical inadequacies. Most organizations use modified version to weigh heavily on ranking using something called the K-factor.
All that said, I only have 2 questions to relic:
- What kind of modification of ELO is COH2 running?
- Can we make any modifications so as to allow COH3's matchmaking to be more fair?
Relic Entertainment
cblanco ★ 
보드카 중대 
VonManteuffel 
Heartless Jäger 
