Loading...
  OR  Zero-K Name:    Password:   

MM rank is misleading

7 posts, 227 views
Post comment
Filter:    Player:  
sort

5 years ago
Currently the stated rank of an MM game is based on the rank of the highest ranking player. This leads to ridiculously misleading titles like this: http://zero-k.info/Battles/Detail/782721

Notice how the other player only has a 2% chance of victory. You see "rank supergiant" and go "oh this might be fun to watch", but no.

Heck even more relevant is the fact that the matchmaker is putting together games with a 2+ rank difference between players. I understand the wait time issue but this is getting old. I can't imagine these games being any more fun to play than they are to watch, especially for the lower ranking player. It doesn't even provide any real learning experience because the skilled player basically just screws around (assuming they don't just crush the newb in 5 minutes).

Games with 1 rank difference are ok, but 2 ranks is totally one-sided.
+1 / -0
5 years ago
Yah. I've had a few of those lately, unfortunately.

I'd be curious to know why this happens, but I would bet that it's because the community is really small, so you can sometimes be matched against people way below or above your league or you cannot play at all.

We all have different ways to handle this, but I'd rather get crushed and learn than to indefinitely wait in MM.
+1 / -0

5 years ago
You seem to be assuming that it's the noobs who are waiting around forever for MM, but given the number of noob-high rank games I would be willing to bet it's the other way around. MM seems to prioritize whoever has been waiting the longest, so it's basically setting up low ranked players to get bashed.
+0 / -0
5 years ago
If I am right, then it works both ways. If you're significantly above or below average, you'll have to wait or be matched at the extremes.
+0 / -0

5 years ago
I mean that's true, but statistically oranges and reds way outnumber silvers and up.

White = top 10%
Orange (depending on which orange, it's stupid that there are two) = top 40-60%

Since yellow is top 20%, doing the math we find that oranges make up 50% of the population while silver and up is only 10%. You're outnumbered 5 to 1.
+0 / -0


5 years ago
quote:
Since yellow is top 20%, doing the math we find that oranges make up 50% of the population while silver and up is only 10%. You're outnumbered 5 to 1.

Shouldn't that be 60%? The interval between the 20th percentile (yellow), and 80th (dark red) is 60 percentile points for both oranges.

If we're talking orange that is actually named Orange (Subgiant, top 40%), then that's only 20%.
+0 / -0

5 years ago
Actually I forgot to account for yellows. Yellows are within one rank of silver so that matchup isn't an issue. That leaves a 40% gap between the lowest orange and the lowest yellow. It's only 60% if you also want to count reds. That's maybe not a bad idea either since reds and dark oranges are only 1 rank apart as well.

Did I mention how much I hate this two-orange color scheme? :P Dark orange is clearly not red no matter how hard you squint. I mean we could always make brown dwarfs, I dunno? Brown?

And I still think at the very least that MM games should be ranked by the lowest ranked player rather than the highest. I'm sick of sifting through red dwarf games that are labeled "supergiant", or worse, "neutron star".
+0 / -0