76 replies to this topic

[Jman5]

Ilwrath, on 20 December 2012 - 02:28 PM, said:

Sounds pointless and boring.

Tell that to all the 8-man groups who went back to pugstomping.

Quote

in any case haters gonna hate and smurfs gonna smurf. it'll still be worlds better than what we have now, which is nothing.

Sure. I am not trying to say we should stop phase 3 of the ELO. I think it will be a wonderful addition to the multiplayer experience. I made this post in the hopes that we can all go into it with both eyes open. Too often, the problems seem to emerge only after a feature is released, so I wanted to have a conversation beforehand while there is still time. Perhaps someone will say something the developers never considered. Either way, it doesn't hurt.

Edited by Jman5, 20 December 2012 - 02:40 PM.

[Major Cockburn]

I could get a ringer account built and running a trollapult in an hour or 2 while watching "educational" movies (pgi why does your filter * out porn?)

[Zylo]

Garth Erlam, on 20 December 2012 - 02:13 PM, said:

Paul has mentioned after Phase 3 is out, we'll look at what happens and add additional statistics, like K/D ratio. So this team with a 50% win percentage and 50/1 K/D is going to stand out in our statistics gathering, right?

So then they'd have to not just lose, but die (often) to hide that. I'm fairly certain that the vast, vast, vast majority of players wouldn't bother.

Is there going to be any restriction in matchmaker to try and match pre-made team sizes as well as stopping the current sync-drop problem that exists for small groups and lone-wolves in the current matchmaking system?

I think a large number of players in the small-group and lone-wolf matches would prefer to see something done to stop sync-drop 4+4 teams. ELO will mean nothing if this sync-dropping of 4+4 into small-group/lone-wolf random matches is allowed to continue.

[p00k]

I'm not sure OP realizes how math works.

Yes you can do all those things you mentioned. There is absolutely no reason to.

Sure you can intentionally tank your ELO rating by throwing a few matches. To what end? So you can fight weaker opponents, and win, and inflate your win/loss? except you had to deflate it first in order to inflate it, so in the end you're right back where you started.

Sure you can intentionally tank your team's average ELO rating. Keep in mind, a new player's rating is 1300, not 0. So yes, you can drag down your team average ELO rating by partnering with newbies. Only, now you have a newbie on your team, and your average didn't drop that much. Sure you can team up with 4 newbies, and drop your team average down a lot. Only, now you have 4 newbies on your team. You'll be facing teams whose best players may not be as good as you, but their worst players are probably better than your dead weight. Seems like a decent match-4 good players and 4 newbs vs 8 average players.

Yes, you can smurf and create a new account. Only, that new account won't have all the nice mech skills, cbills to buy mechs, it won't have the mc your main account does, and after you smurf a few times, if you win your ELO rating rises. At which point you then...make a new acct? And grind cbills/xp again? uhh, go right ahead

[QuantumButler]

My major concern with ELO is just that the p-layerbase won't be large enough to support all the fragmentation it will bring, finding an 8man match is already a exercise in frustration at times and ELO will only make it worse.

Edited by QuantumButler, 20 December 2012 - 03:08 PM.

[Cerlin]

These abuses are things for just bored players it seems. I would love to fight good players If I am good enough too. The new incentive system rewards fighting well and hard and losing in the end. I just dont want to have no weight limit (current 8v8) so I avoid that for now. And smurfing is in every game and again who cares? sometimes you want to play suboptimal builds or just mess around, why shouldnt people be able to do that? It is a F2P game after all.

[Major Cockburn]

p00k, on 20 December 2012 - 03:03 PM, said:

I'm not sure OP realizes how math works.

Yes you can do all those things you mentioned. There is absolutely no reason to.

Sure you can intentionally tank your ELO rating by throwing a few matches. To what end? So you can fight weaker opponents, and win, and inflate your win/loss? except you had to deflate it first in order to inflate it, so in the end you're right back where you started.

Sure you can intentionally tank your team's average ELO rating. Keep in mind, a new player's rating is 1300, not 0. So yes, you can drag down your team average ELO rating by partnering with newbies. Only, now you have a newbie on your team, and your average didn't drop that much. Sure you can team up with 4 newbies, and drop your team average down a lot. Only, now you have 4 newbies on your team. You'll be facing teams whose best players may not be as good as you, but their worst players are probably better than your dead weight. Seems like a decent match-4 good players and 4 newbs vs 8 average players.

Yes, you can smurf and create a new account. Only, that new account won't have all the nice mech skills, cbills to buy mechs, it won't have the mc your main account does, and after you smurf a few times, if you win your ELO rating rises. At which point you then...make a new acct? And grind cbills/xp again? uhh, go right ahead

like I said (maybe in the other thread where its being discussed) it would take like 2 h to get a fully set up trollapult with the new cadet system

[Phelan Ward-Kell]

I believe we're seeing a failure of logic on this thread.

In any event with this system, there's 15 variables (the other players in the game) that our attempted jerk player has to account for.

If he's predicted to lose, and loses, his score drops by only a few points.
If he's predicted to win, and wins, his score raises by only a few points.
If he's predicted to win, but loses, he could drop by 50 points
If he's predicted to lose, but wins, he could raise by 50 points.

If everything goes according to plan, than sure he could always drop against people he's expected to win against, at which point if he loses than his score drops by a good-sized chunk.
But because he has 7 teammates that can still win against 8 players, he has a chance to get the less desirable outcome.
As his ELO is dropping, so are the ELO's of the people he faces, continuing to maintain the variables.

In a completely flawed system, this jerk player could continually crash his ELO with little effort, but the variables are absolutely not in his favor.


Now, this still begs the question of why would this guy spend a week to reduce his ELO to 'n00b' level, only to see it jump back up in a fraction of the time.
This is equivalent to torturing yourself endlessly just so you can get one sparse moment of pleasure, only to have to torture yourself again for an obscene amount of time. There's no satisfaction, or accomplishment in this.


I also think the logic for why someone would make a new account to "farm" CBills with the cadet bonuses is flawed.
You're suggesting people would just keep making new accounts, farming the CBills that they CANNOT transfer to another account, then abandoning that account to do it on another. They don't get to transfer the CBills or accomplishments to another account, they continuously have to start over from scratch, and they continuously have to fight in trial mechs. At best, they gain a guy that can use his low ELO for 8-mans with his unit. At worst, they **** away their life banging their head against the wall.
This is equivalent to burning your *** so you can't sit down, waiting for it to heal, then doing it again.

Edited by The Phigment, 20 December 2012 - 04:12 PM.

[Silra]

ELO ranking in any and all formats will be abused by someone somewhere. So far I have not found any matchmaking in any game that could not be abused by some determined individual for their own amusement. I don't personally care about it one way or another, as it will be a minority thing.

What I am fearful for is all the sillyness coming with ranking systems.
- "Your mech isn't optimized."
- "Stop using XXX, it's garbage."
- "Get YYY next time."

Replace the above sentences with bad grammar and remove about half the letters for them to be internet friendly.
Then add in a lot of random rage messages when someone blames the rest of the team's builds, mech choices, playstyles and living arrangements for losing the match and thus causing him to lose ELO.

Just like any stat in any given game, ELO is going to matter.

[SpiralRazor]

Jman5, on 20 December 2012 - 12:56 PM, said:

For anyone who hasn't seen it: Paul Inouye has been kind enough to outline some of their plans for Phase 3 Matchmaking:

http://mwomercs.com/...65#entry1626065
http://mwomercs.com/...51#entry1639451


Creating a fair matchmaking system is always a struggle especially when you factor in a portion of your playerbase will do everything in their power to abuse it to their advantage.

A lot of it is still unknown, but I thought I would point out some of the issues PGI will have to consider before they release this to the public.

Purposefully sinking your ELO to fight noobs and win more.

In some games the term is called Lossbotting. Players will spam lose matches until the matchmaker starts placing them against legitimately bad players. He then wins 10-20 games in a row and then follows up by spam losing 10-20 in a row. Once they've made it to noob-ville, they maintain a 50% win percentage so they never leave.

Bringing in the dedicated loss botter(s) to sink your team's overall ELO average

So your team is starting to face off against the real opponents and you're struggling to get a win. What do you do? You all blame cheesy builds and bring in your buddy with the ELO score of zero to lower your team's overall average. Ideal scenario would be to bring in enough zero ELO people sink it down to 1300 where the noobs start. The fact that you have teams of 8, means it's easy to subtly switch one or two players in and out to keep your ELO artificially low.

If I were PGI, I would consider creating some sort of fudging effect where you take the highest and lowest rated players and give them less impact on the average.

Smurfing

Smurfing is the term for creating a fresh account so that your ELO is back to noob-level. You then face off against a bunch of legitimate scrubs and wipe the floor with them. I honestly have no idea how PGI could combat this, but it's something they should keep in mind. If this becomes rampant, it will just ruin the beginner experience.

Incentives to Abuse the Matchmaker

Basically, the more of these that exist, the more pressure there will be to abuse matchmaker:

• More Money/Experience for beating a higher rated team: if the game rewards upsets, people will artificially create upsets by lowering their team's ELO.
  
• Achievements/Fluff Rewards: In Starcraft 2, there were people who lost games to bring their ELO to the lowest point possible where only the ultra terrible and other farmers existed. They then traded wins with each other to farm achievements and portraits for their accounts. This trivialized many of the impressive achievements out there. If MWO creates stuff like bobbleheads for 1000 wins or something. People will farm them.
  
• Letting really good players match against mediocre players: If you have 4 really good players paired with 4 really bad players. They will almost always win against 8 mediocre players. be careful about people stacking teams.
  
• 4 man premade rated the same as 4 pugs: If a team of 4 premades are rolling an 8 man pug and are treated the same as 4 random pugs, you're going to have trouble with fair matchmaking. There should be some sort of temporary boost to their ELO to face them against slightly tougher PUGs. Of course ideally the matchmaker would try to split premades in the search queues up evenly and against one another.

Jman5, on 20 December 2012 - 12:56 PM, said:

For anyone who hasn't seen it: Paul Inouye has been kind enough to outline some of their plans for Phase 3 Matchmaking:

http://mwomercs.com/...65#entry1626065
http://mwomercs.com/...51#entry1639451


Creating a fair matchmaking system is always a struggle especially when you factor in a portion of your playerbase will do everything in their power to abuse it to their advantage.

A lot of it is still unknown, but I thought I would point out some of the issues PGI will have to consider before they release this to the public.

Purposefully sinking your ELO to fight noobs and win more.

In some games the term is called Lossbotting. Players will spam lose matches until the matchmaker starts placing them against legitimately bad players. He then wins 10-20 games in a row and then follows up by spam losing 10-20 in a row. Once they've made it to noob-ville, they maintain a 50% win percentage so they never leave.

Bringing in the dedicated loss botter(s) to sink your team's overall ELO average

So your team is starting to face off against the real opponents and you're struggling to get a win. What do you do? You all blame cheesy builds and bring in your buddy with the ELO score of zero to lower your team's overall average. Ideal scenario would be to bring in enough zero ELO people sink it down to 1300 where the noobs start. The fact that you have teams of 8, means it's easy to subtly switch one or two players in and out to keep your ELO artificially low.

If I were PGI, I would consider creating some sort of fudging effect where you take the highest and lowest rated players and give them less impact on the average.

Smurfing

Smurfing is the term for creating a fresh account so that your ELO is back to noob-level. You then face off against a bunch of legitimate scrubs and wipe the floor with them. I honestly have no idea how PGI could combat this, but it's something they should keep in mind. If this becomes rampant, it will just ruin the beginner experience.

Incentives to Abuse the Matchmaker

Basically, the more of these that exist, the more pressure there will be to abuse matchmaker:

• More Money/Experience for beating a higher rated team: if the game rewards upsets, people will artificially create upsets by lowering their team's ELO.
  
• Achievements/Fluff Rewards: In Starcraft 2, there were people who lost games to bring their ELO to the lowest point possible where only the ultra terrible and other farmers existed. They then traded wins with each other to farm achievements and portraits for their accounts. This trivialized many of the impressive achievements out there. If MWO creates stuff like bobbleheads for 1000 wins or something. People will farm them.
  
• Letting really good players match against mediocre players: If you have 4 really good players paired with 4 really bad players. They will almost always win against 8 mediocre players. be careful about people stacking teams.
  
• 4 man premade rated the same as 4 pugs: If a team of 4 premades are rolling an 8 man pug and are treated the same as 4 random pugs, you're going to have trouble with fair matchmaking. There should be some sort of temporary boost to their ELO to face them against slightly tougher PUGs. Of course ideally the matchmaker would try to split premades in the search queues up evenly and against one another.

Mm...sorry what? None of the standard abuses of games like LoL really work in this format.....Big deal, you can smurf in trial mechs unless you want to spend some money...umm...ok.

[Almeras]

I've read about the ELO I too share some concerns;

1. I've seen people in other games (mainly in APB) deliberately drop their ELO just to stomp nubs. For what ever reason they get a kick from trolling.

2. I think their should be an adequate upper cap to ELO. If every drop turns out to be a tooth and nail fight match they are going to get burnt out. Phelan Wolf once said "It would be the battle of a knife against the grind stone". An upper cap would give units the chance for a few easy matches (or you'll end up with smurf'ing groups).

3. If ELO was made a public stat it would encourage elitism and cheese builds.

[Phelan Ward-Kell]

Silra, on 20 December 2012 - 04:02 PM, said:

What I am fearful for is all the sillyness coming with ranking systems.
- "Your mech isn't optimized."
- "Stop using XXX, it's garbage."
- "Get YYY next time."

We already see this with people flaunting KDR and such.

To those who tell someone else their mech isn't optimized, they may be right. But mostly because it isn't optimized for that person.
The greatest thing about this game, is that one person may do really amazing with a really powerful burst damage design, but somebody with a more balanced design can still out do them. It comes down to the skill of the pilot and how well designed the mech is to THEIR play style.

[justin xiang]

Jman5, on 20 December 2012 - 12:56 PM, said:

For anyone who hasn't seen it: Paul Inouye has been kind enough to outline some of their plans for Phase 3 Matchmaking:

http://mwomercs.com/...65#entry1626065
http://mwomercs.com/...51#entry1639451


Creating a fair matchmaking system is always a struggle especially when you factor in a portion of your playerbase will do everything in their power to abuse it to their advantage.

A lot of it is still unknown, but I thought I would point out some of the issues PGI will have to consider before they release this to the public.

Purposefully sinking your ELO to fight noobs and win more.

In some games the term is called Lossbotting. Players will spam lose matches until the matchmaker starts placing them against legitimately bad players. He then wins 10-20 games in a row and then follows up by spam losing 10-20 in a row. Once they've made it to noob-ville, they maintain a 50% win percentage so they never leave.

Bringing in the dedicated loss botter(s) to sink your team's overall ELO average

So your team is starting to face off against the real opponents and you're struggling to get a win. What do you do? You all blame cheesy builds and bring in your buddy with the ELO score of zero to lower your team's overall average. Ideal scenario would be to bring in enough zero ELO people sink it down to 1300 where the noobs start. The fact that you have teams of 8, means it's easy to subtly switch one or two players in and out to keep your ELO artificially low.

If I were PGI, I would consider creating some sort of fudging effect where you take the highest and lowest rated players and give them less impact on the average.

Smurfing

Smurfing is the term for creating a fresh account so that your ELO is back to noob-level. You then face off against a bunch of legitimate scrubs and wipe the floor with them. I honestly have no idea how PGI could combat this, but it's something they should keep in mind. If this becomes rampant, it will just ruin the beginner experience.

Incentives to Abuse the Matchmaker

Basically, the more of these that exist, the more pressure there will be to abuse matchmaker:

• More Money/Experience for beating a higher rated team: if the game rewards upsets, people will artificially create upsets by lowering their team's ELO.
  
• Achievements/Fluff Rewards: In Starcraft 2, there were people who lost games to bring their ELO to the lowest point possible where only the ultra terrible and other farmers existed. They then traded wins with each other to farm achievements and portraits for their accounts. This trivialized many of the impressive achievements out there. If MWO creates stuff like bobbleheads for 1000 wins or something. People will farm them.
  
• Letting really good players match against mediocre players: If you have 4 really good players paired with 4 really bad players. They will almost always win against 8 mediocre players. be careful about people stacking teams.
  
• 4 man premade rated the same as 4 pugs: If a team of 4 premades are rolling an 8 man pug and are treated the same as 4 random pugs, you're going to have trouble with fair matchmaking. There should be some sort of temporary boost to their ELO to face them against slightly tougher PUGs. Of course ideally the matchmaker would try to split premades in the search queues up evenly and against one another.

I think most of the jack ***** who would try these type of tactics (if they could) generally suck at the game on par or below most noobs... I'm not too worried about it.

[Major Cockburn]

justin xiang, on 20 December 2012 - 04:28 PM, said:

I think most of the jack ***** who would try these type of tactics (if they could) generally suck at the game on par or below most noobs... I'm not too worried about it.

nah, I know a bunch of pretty competent players who find this kind of **** highly amusing.

[Corvus Antaka]

no doubt some people will try this. However, losing is a lot less money, and losing badly more so. people dropping their ELO's intentionally like this really I don't see the massive benefit.

Maybe we should pay high ELO players more than low ones? that might be some incentive right there

[IamTheEggMan]

Please tell me how this makes any sense to you? Player wants nothing more then to win, and so loses to achieve this?

[Jman5]

IamTheEggMan, on 20 December 2012 - 05:42 PM, said:

Please tell me how this makes any sense to you? Player wants nothing more then to win, and so loses to achieve this?

They lose on purpose in quick succession then when they play "for real" they only match against easy opponents. I'm surprised so many people aren't familiar with this because it's pretty common in competitive online games

[Crazycajun]

Garth Erlam, on 20 December 2012 - 02:13 PM, said:

Paul has mentioned after Phase 3 is out, we'll look at what happens and add additional statistics, like K/D ratio. So this team with a 50% win percentage and 50/1 K/D is going to stand out in our statistics gathering, right?

So then they'd have to not just lose, but die (often) to hide that. I'm fairly certain that the vast, vast, vast majority of players wouldn't bother.

Garth...man i like you. .. but you are severely under estimating these kinda people... i've seen it in action.. been victim to it..

the point me and jman are trying to get across is this..

say ur a steady team getting a good run going...the elo points are comen in .. then all of a sudden you get 3 + or more matchs in a succession or in broken bits by teams < and you can tell em > that are purposely losing to lower the elo to go < i'll steal the phrase for the right reason now> pug stomp newb pugs ..

this not only affects them..but it WILL affect you to ... not to much if you win..But HORRIBLY if you lose..and you won't wana lose..trust me.

They dont care bout win/loss ... they care bout the profits behind everything..and there is and always will be some form of profit better then some meaningless w/l ratio ..

it gets worst in the tourneys ... YOU might not have seen it.. but i have .. and it sucks.

god forbid if your doing 4 man drops and get slammed with 3-4 of these kinda players in your grp vs 8 people who do care... you're screwed..

Edited by Crazycajun, 20 December 2012 - 06:16 PM.

[p4r4g0n]

Sorry to revive an old thread but there's something here people are missing or avoiding.

Smurfing has zero benefits for a Lone Wolf or solo PUGger other than for sh*ts and giggles. The real benefit to smurfing will lie with the merc corps and maybe Faction military units when they contest planetary control.

When groups contest for a planet, one side's top tier team with 7 players playing their own accounts + one playing a specially created account (that has been levelled up enough to own 1 decent optimised mech but with a nerfed ELO) could be enough to allow this team to be matched against the opposition's second / third tier team.

How effective this will be depends on the the band (i.e. the upper & lower ELO limit) used by the matchmaker in matching units for a fight AND how the system will treat a "failed to find match" when planetary control is being contested.

If the MM band is so wide that it is virtually impossible not to find a match then this avoids the scenario above but the roflstomps will continue in the broader game universe and makes ELO meaningless. However, If the MM is designed to only apply this wider band in planetary conquest and just makes the closest possible match, then we will have hugely mismatched battles.

This would like end up with a few corps whose top tier players will dominate all the contestable planets. Possibly will be less of an issue with Faction planets due to the differences in benefits and restrictions on inter-faction warfare determined by lore.

The only possible way to avoid this that I can see is to make planetary conquest require a significant number of battles that cannot all be fought by a handful of top tier teams in a corp / faction. Then again, PGI may come up with some game mechanic to deal with this.

Just a thought.

Edited by p4r4g0n, 29 January 2013 - 12:16 AM.

[Novawrecker]

PANZERBUNNY, on 20 December 2012 - 01:16 PM, said:

That's a lot of energy being spent on thinking about things you can't effect, like people making new accounts and people losing on purpose, if they are really determined.

Play with people you know. Problem solved.

Many issues are solved by you acting.

To JIGOKU with you and your Harey logic!
(btw, sadly it won't work. It makes too much sense, ergo, it will frazzle the mass' minds. They can't handle the truth. hehe)