Truth behind malak's event
The truth is that there's no date for his return, but every new day, the swgoh bot rolls a random number that has a small chance to trigger the malak event. So like, you can either see him tomorrow th...
Forum Discussion
"Drazhar;c-1914018" wrote:
Oh, does it? Matching a 2.8M GP player with a 3.5M one is just fair then, is it? Clearly based on GP, that's clear. These two GP will surely also have the same pertinent GP. Guaranteed.
The first matchmaking algorithm made total GP vs total GP the primary indicator, whether or not ships were a part of the GA. This had much outcry because ships were unusable yet counted against you. The most common user suggestion was to exclude ship GP if ships aren't a part of the GA.
The second matchmaking algorithm made character GP vs character GP the primary indicator when ships were not a part of the GA. This had much outcry because it counted useless fluff at the bottom of the roster and encouraged practices like mod stripping. The most common user suggestion was to compare the top of the rosters based on how many teams you place to weed out all of the fluff.
Now the third algorithm DOES EXACTLY WHAT WE ASKED. It compares top of roster to top of roster. Why is matching a 2.8M GP player with a 3.5M player fair? Because the top of their rosters are similar and the bulk of that difference is probably a combination of ships, fluff at the bottom of the roster, mods stored on irrelevant toons, and random abandoned G7-8 toons.
And just like the last two algorithms had breakdowns, this one does too. But it's not 2.8M players facing 3.5M players. It's when you get up to that 3.5M+ range where that pertinent GP range is pretty much maxed and the difference becomes a ballooning roster of powerful toons below the cutoff. A valid criticism, but that will be iterated on in future takes on the matchmaking algorithm.
The matchmaking algorithm as it presently exists works as advertised. It's not that it doesn't care about GP. Rather, it interacts with GP in a sensible way that has its own issues but works better than comparing pure total GP to pure total GP.
