![](https://static.mwomercs.com/forums//public/style_images/master/icon_users.png)
![](https://static.mwomercs.com/img/house/merc-corps.png)
So, Contrary To Popular Belief, The Immortal Assault Mech Is No Replacement For Good Mech-Fu! (Or How To Survive When The Matchmaker Screws You Big!)
Started by Bishop Steiner, Feb 11 2014 06:24 PM
52 replies to this topic
#41
Posted 12 February 2014 - 11:20 AM
I was in an oddly similar, yet completely different match the other day.
My team was a pretty even distribution. We had a couple of lights, a couple of mediums, a bunch of heavies, and a couple of assaults. My lance (4-man with buddies) was Cataphracts and Jagermechs.
The other team had an Orion, 2 Cataphracts, a K2, a Jager, 2 Cicadas, and 5 lights. I think it was 4 Embers and a Jenner. They were seriously underweight compared to us, but I didn't do the math to figure out the exact numbers.
The thing is... the other 2 lances on my team sucked. Bad. The 7 enemy "lights" (counting the Cicadas) rolled through one of them almost before we'd started and then came after my lance. The other puggies got steamrolled by the enemies with weight, though they did significant damage.
My lance went defensive, worked together, and traded a Jagermech for 4 enemies. The remaining 3 "lights" ran off right about the time the remaining enemies with weight arrived only to be obliterated by my lance (which lost another Mech in the process). Then the 3 "lights" returned and my lance's 2 remaining Mechs beat them handily.
So on drop it looked like a serious weight-advantaged ROFLstomp. But that didn't survive initial contact - it suddenly looked like a light swarm stomp instead.
Ultimately, though, teamwork prevailed. My lance held together better than any of the enemy lances, and our teamwork won the day. It wasn't weight. I don't think it was Elo (the other team's players seemed pretty good). We just out-cooperated the other team.
My team was a pretty even distribution. We had a couple of lights, a couple of mediums, a bunch of heavies, and a couple of assaults. My lance (4-man with buddies) was Cataphracts and Jagermechs.
The other team had an Orion, 2 Cataphracts, a K2, a Jager, 2 Cicadas, and 5 lights. I think it was 4 Embers and a Jenner. They were seriously underweight compared to us, but I didn't do the math to figure out the exact numbers.
The thing is... the other 2 lances on my team sucked. Bad. The 7 enemy "lights" (counting the Cicadas) rolled through one of them almost before we'd started and then came after my lance. The other puggies got steamrolled by the enemies with weight, though they did significant damage.
My lance went defensive, worked together, and traded a Jagermech for 4 enemies. The remaining 3 "lights" ran off right about the time the remaining enemies with weight arrived only to be obliterated by my lance (which lost another Mech in the process). Then the 3 "lights" returned and my lance's 2 remaining Mechs beat them handily.
So on drop it looked like a serious weight-advantaged ROFLstomp. But that didn't survive initial contact - it suddenly looked like a light swarm stomp instead.
Ultimately, though, teamwork prevailed. My lance held together better than any of the enemy lances, and our teamwork won the day. It wasn't weight. I don't think it was Elo (the other team's players seemed pretty good). We just out-cooperated the other team.
#42
Posted 12 February 2014 - 11:51 AM
ELO is a decent gathering force to get matches going. The problem, though, lies in the fact that we've got a very small population of players from which to pull for a game at any one time. This is lessened by the fact that each player can have up to 4 different ELO scores. But, that is rendered moot when you can only queue up one mech at a time. Furthermore, the ELO isn't adjusted based on your mech's total tech tier and/or their efficiencies. And, if that isn't enough, when you begin the game, your ELO score doesn't start at 0.
Granted, this is all just my opinion but what needs to happen is:
Finally, having multiple mechs, each with their own scores, in the queue would really help the match maker. At this point, you'd have so many mechs available from which to choose, even the limiter that is the weight cap woudn't impact the speed of matches. Furthermore, it would promote even more balanced matches with ELO variances being in the single digits and it would prevent arm locked trial mech newbies from hopping in games with veterans.
Granted, this is all just my opinion but what needs to happen is:
- ELO needs to be wiped for everyone and have everyone start at 0
- All PGI offered trial mechs would have a permanent ELO score of 0 (or whatever)
- ELO needs to be centered on individual mech variants and not weight class
- ELO needs to be adjusted up or down based on the mech's total efficiencies (tech could be a bit of a problem)
- Players need to be able to queue up as many mechs at a time as they wish
Finally, having multiple mechs, each with their own scores, in the queue would really help the match maker. At this point, you'd have so many mechs available from which to choose, even the limiter that is the weight cap woudn't impact the speed of matches. Furthermore, it would promote even more balanced matches with ELO variances being in the single digits and it would prevent arm locked trial mech newbies from hopping in games with veterans.
#43
Posted 12 February 2014 - 11:56 AM
I would have kicked yer ends !!!
or maybe not
or maybe not
![:D](https://static.mwomercs.com/forums/public/style_emoticons/default/tongue.png)
#44
Posted 12 February 2014 - 12:03 PM
warner2, on 12 February 2014 - 04:37 AM, said:
It isn't popular belief that the assault is a replacement for good mech-fu.
What made you think it was?
Only really bad players in Atlases think they are going to do well because they weigh 100 tons. Speaking of which, there was such a player on the opposite team, along with a © Stalker that did 0 damage and some dude in an A1 that did 5.
This is just another example of the lottery that is PUG matches.
What made you think it was?
Only really bad players in Atlases think they are going to do well because they weigh 100 tons. Speaking of which, there was such a player on the opposite team, along with a © Stalker that did 0 damage and some dude in an A1 that did 5.
This is just another example of the lottery that is PUG matches.
If my memory serves, the Stalker was ECM'ed out by one of the 3L's, and the A1 was a SRM6 Splat cat.... so they really had little chance to shoot anything.
#45
Posted 12 February 2014 - 12:05 PM
Bishop Steiner, on 12 February 2014 - 09:34 AM, said:
I've also been on the other end of the Steiner Scout roll. Had a team with 7 Atlases and one mech below the heavy range absolutely crush my underweight medium/light company, because they parked in a nice wide open space and wouldn't be budged. Even when you tried to swarm into them, they didn't mind a little friendly fire because they had the armor to shrug it off.
One thing about Skirmish mode...... no cap is good most of the time, but it also means you can't use it to force the oppositions hand. (Then again, I have also had to try to dislodge a DDC and BoomJager company off our cap in 12 man..... that was just ugly)
You mean like this:
http://steamcommunit...s/?id=218284961
Number of assaults on my team = 0, heavies = 3. Number of assaults on enemy team = 6, heavies = 2.
#46
Posted 12 February 2014 - 12:57 PM
Trauglodyte, on 12 February 2014 - 11:51 AM, said:
And, if that isn't enough, when you begin the game, your ELO score doesn't start at 0.
Elo doesn't start at zero. Elo starts at whatever you've defined to be "average" for your implementation. Starting it at 0 wouldn't solve anything - it would just change the midpoint of the rating scale for MWO.
Quote
All PGI offered trial mechs would have a permanent ELO score of 0 (or whatever)
Bad, bad, bad idea. Or are you actually implying that you think that an expert in a trial Mech is no better than a complete noob in a trial Mech? Elo doesn't rate Mechs, it rates player performance.
Quote
ELO needs to be centered on individual mech variants and not weight class
This would certainly provide the potential for increased accuracy of Elo ratings, but it would take far longer for Elo ratings to arrive at their correct values. Given the seemingly low population of players in MWO, I think this would actually be counter-productive.
Quote
ELO needs to be adjusted up or down based on the mech's total efficiencies (tech could be a bit of a problem)
No no no. It appears that you don't understand what Elo ratings actually are. They are not and should not be adjusted based on the loadout of a Mech. You could theoretically have a different Elo rating for every player for every build of every Mech, but that's not a realistic system. But you should not "adjust" ratings based on loadout.
It's bad enough that PGI "adjusts" ratings for new players. Doing that causes their ratings to take longer to reach their correct values.
#47
Posted 12 February 2014 - 01:46 PM
ELO rating, in my mind, is just a bad way of doing things. The ladder system that Blizzard used was much better. We can agree to disagree and I won't take away from your thoughts. But, starting at the mid point is a horrible way of doing things because it assumes that all players are average AT that point. Day 1 newbie in a champion mech isn't average at all. It's exactly why they're arm locks and using weapons beyond their max ranges or firing LRMs without locks. These players don't belong anywhere near games with me, you, or anyone else and it is part of the reason why the newbie experience is so harsh. And while I understand your feelings on not deducting from someone's ELO based on their efficiencies, they directly contribute or hold back the performance of the player because they're not on par with the mechs used by everyone else. ELO needs to match players together based on skill and mech performance...doing anything other than that is a hinderance to everyone involved.
#48
Posted 12 February 2014 - 02:04 PM
Blizzard's ladder system was Elo-based.
The main difference (as I understand it) was that Blizzard had a separate queue for new players. However, they still started with "average" Elo ratings and not 0 ratings. After some number of games they took their Elo ratings with them and were moved into the "you're not new anymore" queues and from then on they had to sink or swim on their own.
Not all newbies are bad. I assure you that I was not bad in my first game of MWO. I wasn't as good as I am now, but I was far from bad. I'm guessing you weren't bad either. Newbies start out with an "average" Elo rating so that they can easily adjust up or down to get to their correct rating quickly. If you are good you will rapidly move above the average rating. If you are bad then you will rapidly move below it.
If you are truly average, then yes you will have to constantly deal with new players in all of your games. Get better.
![:D](https://static.mwomercs.com/forums/public/style_emoticons/default/smile.png)
Not all newbies are bad. I assure you that I was not bad in my first game of MWO. I wasn't as good as I am now, but I was far from bad. I'm guessing you weren't bad either. Newbies start out with an "average" Elo rating so that they can easily adjust up or down to get to their correct rating quickly. If you are good you will rapidly move above the average rating. If you are bad then you will rapidly move below it.
If you are truly average, then yes you will have to constantly deal with new players in all of your games. Get better.
![:(](https://static.mwomercs.com/forums/public/style_emoticons/default/wink.png)
#49
Posted 12 February 2014 - 02:46 PM
If I remember correctly, and it has been a long time, I want to say that everyone started at 1000. So, I guess that I was wrong about WoW. For some reason, I was thinking that everyone started at 0. Maybe that was the rated BGs? I don't know.
You are correct that not all new players are bad. But, I've seen some doozies that'll just make you cry for the future of humanity. Thing is, I've got to play a million times better than normal to offset them. It isn't an issue of getting better cause I play with a lot of the big boys. Having arm locked bads in my team offsetting either me or the people I'm grouped with, though, kills my gaming experience. I make mistakes and I own up to them but I can't make up a group of 6 mechs that total 200 damage. So, I have to ask why they're in the game with me when they should be laddered up with people of their quality so that they can learn the game and get better. That is why I said that being allowed to queue up multiple mechs to artificially boost the population would go a long way to helping out.
I still don't agree with starting people at the midrange. I think that everyone should start from zero and build up. Either that or ELO drop/gain needs to be bigger and stronger. But, I will confess that I don't much understand the system all that well.
You are correct that not all new players are bad. But, I've seen some doozies that'll just make you cry for the future of humanity. Thing is, I've got to play a million times better than normal to offset them. It isn't an issue of getting better cause I play with a lot of the big boys. Having arm locked bads in my team offsetting either me or the people I'm grouped with, though, kills my gaming experience. I make mistakes and I own up to them but I can't make up a group of 6 mechs that total 200 damage. So, I have to ask why they're in the game with me when they should be laddered up with people of their quality so that they can learn the game and get better. That is why I said that being allowed to queue up multiple mechs to artificially boost the population would go a long way to helping out.
I still don't agree with starting people at the midrange. I think that everyone should start from zero and build up. Either that or ELO drop/gain needs to be bigger and stronger. But, I will confess that I don't much understand the system all that well.
#50
Posted 12 February 2014 - 03:03 PM
The real problem is complex, but can be simplified thusly:
1. There aren't enough people playing MWO.
2. Because there aren't enough players, the matchmaker has to look at much too wide of a band of Elo ratings in order to put together matches in a reasonable amount of time.
3. Because the matchmaker is forced to use a very wide Elo-band to create matches, you end up with average-rated players (which includes new players) in matches with experts.
I suspect, but of course have no proof, that PGI can't (or isn't willing to) create a separate queue for cadet-bonus players because there either aren't enough of them, or aren't enough total people to make the separate queues work.
FYI - Elo is a zero-sum system. If you start everyone at 0, that just means that 0 becomes average, negative numbers are below average, and positive numbers are above average. New players must start at the average value or it isn't an Elo system. PGI started everyone at 1300 for MWO, so that's average.
1. There aren't enough people playing MWO.
2. Because there aren't enough players, the matchmaker has to look at much too wide of a band of Elo ratings in order to put together matches in a reasonable amount of time.
3. Because the matchmaker is forced to use a very wide Elo-band to create matches, you end up with average-rated players (which includes new players) in matches with experts.
I suspect, but of course have no proof, that PGI can't (or isn't willing to) create a separate queue for cadet-bonus players because there either aren't enough of them, or aren't enough total people to make the separate queues work.
FYI - Elo is a zero-sum system. If you start everyone at 0, that just means that 0 becomes average, negative numbers are below average, and positive numbers are above average. New players must start at the average value or it isn't an Elo system. PGI started everyone at 1300 for MWO, so that's average.
#52
Posted 12 February 2014 - 05:44 PM
??? elaborate for those of us on Android phones and can't use flash player
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users