

Revert Mm To Non-Elo...
#1
Posted 27 February 2013 - 08:21 PM
http://cloud-2.steam...543075FD55CD61/
#2
Posted 27 February 2013 - 08:25 PM
Unless those are 2 8drops fighting each other that should never happen.
#3
Posted 27 February 2013 - 08:28 PM
It should take tonnage into account before Elo ranking.
#4
Posted 27 February 2013 - 08:33 PM
It's one thing to be good, and these guys weren't bad, but it's another thing to know you can mass drop the Doesn't Die Chassis into a game and half of the time have it be a free win thanks to the terrible matchmaking...and the fact Atlas' just don't die..ever.
Edited by PapajIGC, 27 February 2013 - 08:39 PM.
#5
Posted 27 February 2013 - 08:37 PM
It's worrisome.
#6
Posted 27 February 2013 - 08:45 PM
Edited by PapajIGC, 27 February 2013 - 08:45 PM.
#7
Posted 27 February 2013 - 08:47 PM
#8
Posted 27 February 2013 - 08:51 PM
The winning team had a lot of tonnage, but they were apparently also terrible.
#9
Posted 27 February 2013 - 08:57 PM
Edited by PapajIGC, 27 February 2013 - 08:59 PM.
#10
Posted 28 February 2013 - 02:02 AM
Stop panicking whenever the new thing they add doesn't work perfectly. I swear, people have completely forgotten that we're doing this to TEST game systems, FIND the problems, and then TEST the fix.
"Oh my god, we installed metal detectors at the embassy, and now we have to deal with knives and guns being brought in! Throw those things in the closet until we fix our security procedures, this is ridiculous!"
No, it's not a perfect metaphor, but yours isn't a perfect argument, so it'll do.
Edited by Renthrak, 28 February 2013 - 02:02 AM.
#11
Posted 28 February 2013 - 03:36 AM
I never drop alongside 3 trial cicadas anymore.
Sometimes you get bad results with _any_ randomized system. ELO is far better than just matching on weight, because honestly, weight doesn't say much at all.
I'd much rather be in a group with 7 skilled jenner pilots vs 8 skilled D-DC's than in a group with 7 completely newbish trial Atlas D pilots vs the 8 skilled D-DC's. If my copilots are skilled at least we can use some tactics and make the best of the situation.
#12
Posted 28 February 2013 - 06:20 AM
Also, I never said dump it entirely, I said dump it until it is ready to be properly implemented. They had no problem doing this for 2-3 days before the 50/50 weekend. And honestly, if the 50/50 weekend wasn't enough data (it might not be precise data because a lot of that weekend was "cheese") then their playerbase is really a lot worse off than we thought. If that is the case, then they need to do something to either increase it or maintain it. Getting dropped against 4, 5, or even 6 D-DCs when your team consistently has 2 or 3 assaults (if you're lucky) is a waste of time and I'd rather have to re-enter a match queue than flail hopelessly about as I see a curtain of death approach me from afar knowing there is nothing that can be done.
The whole "but it's beta" bit is getting old too. It was in closed beta for 3 months and has been in open beta for close to 5 now. For the amount of money the development team has gotten coupled with the length of time the game has been in beta, there should be no excuse why changes cannot even be run thru simulations based on the "live" matchmaking history to see what new matchmaking systems work better than others. I suppose we should all be lucky PGI caved under pressure and let everybody keep their progress from open beta onward; any other game would've scrapped it all and refunded any cash items back to the accounts. But hey, they had to keep people playing somehow, it would've been pointless to otherwise if they're development/maintenance cycle is this slow (the only fast thing about that recently has been the gaurenteed map + hero mech + chassis every month (which has only been done in February and scheduled to happen for March and the forseeable future - we'll just have to see if they keep their promises)
Edited by PapajIGC, 28 February 2013 - 06:20 AM.
#13
Posted 28 February 2013 - 08:50 AM
PapajIGC, on 28 February 2013 - 06:20 AM, said:
Let's be honest, dictating other people's reasons for playing the game is rude.
Quote
Which is very hard to do if they can't test it. You know, in, like, an open beta.
And skill also carry matches. Without ELO you can also end up stomped, so since none of them ensures you won't get stomped, why not just dump both until we have a perfect system?
Quote
Yeah because developing a system for simulating matches based on the results of all previous games won't take resources at all. And 8 months of player base testing isn't that much you know. It's not that I'm a PGI fanboy (I've critiziced them in several other threads for stuff) but 8 months of testing isn't that much. It's an open beta. You knew it was an open beta when you began playing. Open betas are for testing stuff. Now they are testing stuff.
And you know what? Plenty of people prefer ELO. I mean, if your hypothesis was correct, that their weight-based matchmaking system is more even than their skill-based matchmaking system, I should end up with less even matches now and get stomped much more - my prefered 'mechs are spider, cicada and raven - but I've never had as many even matches as now.
EDIT: And yeah, I think it's bad of PGI not to do a full reset when they move to the full version, but that's not really for this topic.
Edited by Stringburka, 28 February 2013 - 08:51 AM.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users