540 replies to this topic

[Svidro]

Player A was on a team, player B was on a team. If player A's team won, player A beat player B. Along with a lot of other people. I haven't seen anywhere other than his, admittedly, incredibly poorly worded example that suggests anything other than team averaged ELOs. And he clarified that post fairly quickly.

[Mawai]

Willard Phule, on 17 February 2014 - 07:56 AM, said:

http://mwomercs.com/...79-matchmaking/

Scroll down to the bit with the blackboards on it.

Now, granted, this is OLD information but....from how I understand it....the only thing that has changed is 8 man to 12 man.

It references BOTH how it works to build a team (with everyone's average Elo score) and how you can increase your personal score (by defeating a higher ranked player).

So, herein lies my confusion.

Is it based on a "Team Win?" Because, if this is the case, then all the complaining about "The Matchmaker is trying to ruin my Elo" has some basis in fact. Paul specifically stated that the matchmaker puts together one team that is "favored to win" and one that isn't.

This is a quote from Paul on that string:

Some people also asked to have the description simplified. Here's the summary:

• The Match Maker uses a scoring system to determine if your team is more likely to win or lose based on your team's average Elo rating.
  
• If the Match Maker determines that you're going to lose, but you actually win, then your Elo score is going to go up and the enemy's score is going to go down.
  
• If the Match Maker determines that you're going to win and you actually win, then your Elo score isn't going to change very much (if at all). The same applies to a prediction of loss and you actually lose, your score may drop but it will be slight.
  
• The more games you participate in, the more accurate the Match Maker becomes and you will start seeing that you are playing against people of relative equal skill.

If it's not based on Team win but based on individual performance, why are the experienced players being put into matches with brand new ones? That doesn't make sense, does it? If I have a 1.4 k/d ratio and a new player as a 0 (because this is his first game), how on earth could it possibly be considered "equal?"




That's all I'm getting at here, man.

To use that AWESOME football analogy with PGI's Elo rules applied....

You have the #1 Professional Quarterback in existence. He's got an Elo rating that is based on his lifetime performance (he won some, he lost some, but now he's on the top of the heap). He's got a Football Elo rating of 2200.

Now you have Junior Mechsales...the quarterback of his PeeWee team. He's only been playing for this season, but his team won the neighborhood championship. He's got a Football Elo rating of 1500.

Since the matchmaker is allowed to pick teams +/- 1400, putting these two against each other is PERFECTLY BALANCED.

What's wrong with this scenario? The "New Guy" starting off with a rating that puts him in the match range of the Professional. PGI starts new players off at 1300....which means that with the +/- 1400, they're eligible to play with people in the 2700 range.

Perhaps they need to start newer players off at a lower score as opposed to tweaking the search range. Dunno. That would certainly keep the newer players together but it would screw the experienced players in trying to find a match.

And therein lies yet another problem. Why do the guys with the 2500+ Elos have such a hard time finding a match? Once upon a time, this place was FILLED with people that were part of the Closed and Open Betas. You couldn't swing a cat and not hit 20 of them.

Where have they gone? Elsewhere, that's where. They got sick and tired of all the promises made and broken, problems with the game never getting fixed....PGI paying more attention to dragging in more new, inexperienced people to shell out cash to the beast than the people that TOLD THEM WHAT WAS WRONG and in some cases, HOW TO FIX IT.

By ignoring it's community, PGI has virtually guaranteed stagnation for this game. As they move forward at a snail's pace, trying to put a band-aid over sucking chest wounds, the community continues to dwindle.

No .. and yes,

The team average ELO is calculated based on the individual player ELOs. The matchmaker then calculates a probability of one team defeating the other. IF the team it predicts will win wins the match THEN NO ONE's ELO changes on either the winning or losing side. IF the matchmaker is wrong then the ELO of both teams changes ... the ELO of all players on the team that was predicted to win drops and the ELO of all players on the team that was predicted to lose rises. The change is capped at 50 points and is calculated based on the ELO difference between the two teams.

Individual play in the match DOES NOT matter. You could kill all the opponents and your change in ELO will be no different from any of your team mates.

Summary:
1) Matchmaker calculates the team cumulative ELOs and predicts a winner for the match.
2) If the matchmaker is correct there are NO changes to any ELO values.
3) If the matchmaker prediction is incorrect - the ELO of players on the team predicted to win drops and the ELO of players on the team expected to lose rises.

That's it. Individual performance in a specific match is not relevant and neither is your overall win/loss record. Only matches where the matchmaker is incorrect in its prediction result in ELO changes.

Quoted from the thread you cited:

"At the start of a match, all player's on ONE team have their Elo ratings totaled and divided by 8 (max players). You may realized that this is simply the AVERAGE of a team's total Elo. Team 1's average and Team 2's average are then used to calculate the probability of win (as per the formulas above). If Team 1 beats Team 2, then the appropriate math as above is applied to each player using the probability score calculated by the team averages." ... i.e. everyone gets the same result

P.S. As for placement of new players - I agree that the middle of the range is too high.
P.P.S. The matchmaker only expands the EO matching range when it can't find a match for the player within a smaller range in a reasonable time ... and this applies to folks at both ends of the bell curve.

Edited by Mawai, 17 February 2014 - 09:16 AM.

[Willard Phule]

Svidro, on 17 February 2014 - 08:58 AM, said:

Player A was on a team, player B was on a team. If player A's team won, player A beat player B. Along with a lot of other people. I haven't seen anywhere other than his, admittedly, incredibly poorly worded example that suggests anything other than team averaged ELOs. And he clarified that post fairly quickly.

OK, let's say the entire thing IS based off of how your team performs. That means that the matchmaker is intentionally putting together one team it expects to win....and one team it expects to lose. If you get stomped 10 times in a row, it would be an easy assumption to make that your personal Elo is too high for the playerbase that's online right now....so the matchmaker is intentionally doing what it can to drop your rating to get you closer to everyone else's average at that time.

It also means that all those posts of "the matchmaker is out to get me" has merit. At least, from an outward appearance.

My problem, again, is with the whole concept of using "Elo" for MW:O.

Elo is a system designed for a 2 person, turn based game where both people start out equal. EQUAL. They each start out with 2 Rooks, 2 Knights, 2 Bishops...etc.

MW:O is a 24 player, live action, first person (or Third, depending) shooter where the mechs chosen to play vary not only in tonnage but weapon systems used. How could an "Elo" system possibly be used in a situation like this?

Nevermind. It's not worth arguing about. Here in 6 months or so when there's only 20 or 30 of us left, it'll be a more interesting conversation.

[Sandpit]

Willard Phule, on 17 February 2014 - 08:51 AM, said:

That IS true. Absolutely.

Shame PGI can't set up some sort of "easy to get to" page with links to these groups...which would probably be a great idea. You know...something like the tab/links up at the top (Game, Media, Forums, Community, etc.). Click the link, get some sort of banner-link to a unit that has tutorials, people in their chat to help, etc.

There are a LOT of them in the forum threads but, and I'll use my 12 year old as an example here (though, not a great one), there are a huge number of gamers out there that simply don't bother with forums. If you ever played WoW and surfed theirs....*cringe*. Y'know?

I agree but we don't have that so the community HAS taken it upon themselves to do so. IRC channel and TS are GREAT ways for new players to find quick info and tips and such. So we do have that even if it's not the optimal implementation. Encourage new players to use those tools until PGI implements its own version. This isn't the first game where the community has had to "pick up the slack" so to speak and it probably won't be the last.

[Willard Phule]

Sandpit, on 17 February 2014 - 09:21 AM, said:

I agree but we don't have that so the community HAS taken it upon themselves to do so. IRC channel and TS are GREAT ways for new players to find quick info and tips and such. So we do have that even if it's not the optimal implementation. Encourage new players to use those tools until PGI implements its own version. This isn't the first game where the community has had to "pick up the slack" so to speak and it probably won't be the last.

Nope, the community did a pretty good job back in the MW4 days. But, then, there were LOTS of links to units trying to help on the offical sites, as well.

Maybe someone that's got a lot more tact than I should approach whoever's in charge of the MW:O website and see if you can't come up with something like that.

[Sandpit]

Willard Phule, on 17 February 2014 - 09:32 AM, said:

Nope, the community did a pretty good job back in the MW4 days. But, then, there were LOTS of links to units trying to help on the offical sites, as well.

Maybe someone that's got a lot more tact than I should approach whoever's in charge of the MW:O website and see if you can't come up with something like that.

http://mwomercs.com/...ew-player-help/
http://mwomercs.com/...s-inner-sphere/
http://mwomercs.com/...des-strategies/
http://mwomercs.com/...naments-events/

You can lead a horse to water. The sections are there. Although I agree that stuff needs to be displayed prominently in-game because there's tons of players that never have and never will join the forums.

[Mystere]

[wanderer]

Ngamok, on 17 February 2014 - 08:05 AM, said:

I too sold the Locusts but the V. But I found the SHDs, TDRs, and BLRs, all fun. And some of the WVRs and GRFs.

Wolverines and Griffins actually make nice missile boats. I can pretty much stick 2xML, 2x Streak, LRM 15+ Artemis on any of them and do surprisingly well.

[Svidro]

I don't think the matchmaker designs the teams such that one team will lose. It tries to get them as close as possible, and then the ELO change that it assigns is based on which team had the higher ELO. The matchmaker isn't actually out to get one team.

It also has limited time, severely limited playerbase, and people dropping in groups of atlases, groups of lights and mediums, etc to try to balance as well. Aaaand we see the result. It's not out to get you, really.

"Never attribute to malice that which can adequately be explained by stupidity." and all that.

[Joseph Mallan]

Actually it is probably to limit Students getting hurt by better fighters. We on the other hand are playing a game that plays at war and want to have Sports like Rules.

I have played these game and TT for going on 30 years. Each video/computer game came and went, won awards and eventually lost lime light. TT has stayed viable for 30 years and kept my interest for 27ish. I never played a video game more than 3 months? I also played as a GM for 17 years, Know how many games I won in all those years? i don't expect it to stick Like some of the mainstay PVP games. Why? Cause it has never been a huge draw! The Battletech game and MechWarrior offshoots have survived by die hard fans, not volume sales. If the DEVS thought they were going to break that, then they never should have billed it as CBT Clan Invasion PvP game. That is what drew me here. Had it been Billed as MechCommander I'd never have spent a dime. I didn't like that game, not my cup o tea.

[Joseph Mallan]

Mystere, on 17 February 2014 - 10:27 AM, said:

That is exactly how I am feeling after 14days in a row of 10 hour days mostly on my feet!

[Willard Phule]

Joseph Mallan, on 17 February 2014 - 11:41 AM, said:

Actually it is probably to limit Students getting hurt by better fighters. We on the other hand are playing a game that plays at war and want to have Sports like Rules.

I have played these game and TT for going on 30 years. Each video/computer game came and went, won awards and eventually lost lime light. TT has stayed viable for 30 years and kept my interest for 27ish. I never played a video game more than 3 months? I also played as a GM for 17 years, Know how many games I won in all those years? i don't expect it to stick Like some of the mainstay PVP games. Why? Cause it has never been a huge draw! The Battletech game and MechWarrior offshoots have survived by die hard fans, not volume sales. If the DEVS thought they were going to break that, then they never should have billed it as CBT Clan Invasion PvP game. That is what drew me here. Had it been Billed as MechCommander I'd never have spent a dime. I didn't like that game, not my cup o tea.

I didn't much like MechCommander either, until deployment. I took a bunch of my older games in case I had downtime (I was Artillery...they get a LOT of downtime). Turns out that my laptop just wasn't manly enough to handle 1st person shooters. If you get bored enough, MechCommander starts to look good after a while. Just don't expect it to stay interesting. It was good for a couple of days, that's about it.

Dunno if it ever had any multiplayer support, but the platform would be good if it was used to bring a bunch of people from long distance together to do a turn-based BT type game.

[carl kerensky]

Rebas Kradd, on 26 January 2014 - 10:48 AM, said:

There is still a kernel of truth to them. I'm just not sure what can be done about it at this point.

More than a Kernal...a whole frickin cornfield. The only exaggerations I see are from the white knights. You need to shine the spotlight on the BUGS if you want them fixed. OP got it right.....

[Iskareot]

Mystere, on 26 January 2014 - 11:42 AM, said:

I must be in Elo heaven then because it's been a very fun weekend so far.



As someone who has been playing exclusively solo since the very beginning, I am disputing your 75% figure.




So going into 4-mans is a crutch that you need. And the same holds for running heavies and assaults. We got it.

Its funny because this is the worst I have seen it on a weekend since open beta -

[Mystere]

Iskareot, on 17 February 2014 - 01:28 PM, said:

Its funny because this is the worst I have seen it on a weekend since open beta -

You're quoting me from 3 weeks ago!

But nevertheless, this weekend's games still confirm I am In Elo heaven!

[Craig Steele]

Sandpit, on 17 February 2014 - 08:45 AM, said:

here's the thing

You DO have the Law and just about every other unit out there trying to help new players. We put up guides, tips, threads, TS servers, IRC channels, etc. to help increase the new player experience. You have a select few that like to run around and do nothing but berate those that drop in groups, belong to units, and don't PUG exclusively. So even if it's just 1 new player that listens to that ridiculous propaganda and smear campaign, you have new players that avoid any and all contact with one of the biggest and best resources for learning the game.

As powerful as the community support is though Sandpit, a properly structured and engaging in game intro (with cut scenes, tutorials etc) is a significantly more powerful tool.

If PGI are serious about engaging and retaining new players, leaving it up to the community and hoping the new player navigates that is far from the best system.

Besides, new players should be their lifeblood, that's something they should WANT to have ownership / control over. But as you point out, it's really been left to the community.

[KharnZor]

Mystere, on 17 February 2014 - 02:28 PM, said:

You're quoting me from 3 weeks ago!

But nevertheless, this weekend's games still confirm I am In Elo heaven!

I too had a great time this weekend. Many games were had and nearly all of them close. there was some stomps but hey, thats the way it goes.

[MischiefSC]

KharnZor, on 17 February 2014 - 03:28 PM, said:

I too had a great time this weekend. Many games were had and nearly all of them close. there was some stomps but hey, thats the way it goes.

I had a lot of amazing games. One of the things I like about Skirmish is that when it comes down to a close game there's no easy way out. When it's a one legged Cataphract on one team and a mostly cored, armless Hunchback and a one-legged Stalker with no more LRMs on the other, nobody can go cap to avoid the final showdown.

I had a lot of very, very close games. Some of them went 10 minutes with only 1 or 2 losses on each side, then finally one side made one mistake too many and.... WHOOMPH. It wasn't that it was a stomp even though it ended 12 to 2 - it's that both teams did incredibly well and when things finally went one way or the other it was like a dam breaking.

Even trying to level up Locusts I just can't get into Conquest or Assault anymore. They feel so shallow - no need to commit. The only way to play Skirmish is all in and it's just a lot more enjoyable for me, especially the games that end up 10 to 12 or 11 to 12. The whole match can come down to 1 or 2 hits or misses between mostly crippled mechs.

Good stuff, the weekend was full of them for me.

[Sandpit]

Craig Steele, on 17 February 2014 - 03:22 PM, said:

As powerful as the community support is though Sandpit, a properly structured and engaging in game intro (with cut scenes, tutorials etc) is a significantly more powerful tool.

If PGI are serious about engaging and retaining new players, leaving it up to the community and hoping the new player navigates that is far from the best system.

Besides, new players should be their lifeblood, that's something they should WANT to have ownership / control over. But as you point out, it's really been left to the community.

But instead of doing something to help that community, many are just content to sit there and whine and complain about it instead of actualyl doing something, then complain when someone else gives them advice, tips, etc. on how to work around the shortcomings

[Craig Steele]

Sandpit, on 17 February 2014 - 06:27 PM, said:

But instead of doing something to help that community, many are just content to sit there and whine and complain about it instead of actualyl doing something, then complain when someone else gives them advice, tips, etc. on how to work around the shortcomings

Yeah, I'm not defending or condoning that sort of thing.

But I still have a view that ownership of it belongs with PGI. Until they get serious and take some ownership of something that should be a fundamental building block, I can empathise with peoples frustration.

I guess it's like some of the threads, people have different views because of "he said" or "she said" or "this post" (which was superceded by "that post" but they hadn't seen that one) and that can come out poorly.

But it would be much better for all if the new player experience was organised and streamlined and not left up to the good hearted well intentioned minority who happen to be online when someone needs a hand.