Guild Wars 2

I have some backgrounds in statistics. Not any real strenuous statistics beyond checking for heteroskedacity and doing Chi, Z, T-tests and running linear regressions.

There seems to be some variance of points people have on leaderboard with same number of wins and roughly the same winning %. I’m curious as to if MMR is producing these results.

I don’t really care too much about leaderboards, but think point system may need looking into as it seems RNG-ish on a just mere glance.

There’s actually a table for it somewhere, the number of points won/lost depends on:

a) the teams respective winning expectation (which appears to be based off mmr and party size)
b) how much score difference there was at the end of match (losing 0-500 will make you lose more than 400-500)

^this is from an official anet source even tho they don’t articualte on how the winning expectation is calculated exactly.

so basically if they have roughly the same number of games and win% but different points one of them won against teams considered stronger or with less score difference at the end of match when it lost (most likely a combination of both)

Justin,

And how well is the Expected Win function performing against real previous game scores/win data as you back-simulate the games that took place?

Justin ODell.9517:

Thanks adozu! Here are the links to go along with what you were referring to.

adozu.6398:

this is from an official anet source

http://wiki.guildwars2.com/wiki/PvP_Matchmaking_Algorithm#Ladder

adozu.6398:

even tho they don’t articualte on how the winning expectation is calculated exactly.

http://wiki.guildwars2.com/wiki/PvP_Matchmaking_Algorithm#Match_Prediction

Cool, I was just looking for that info. Thanks Justin.