37 replies to this topic

[BlueStrat]

The PSR system in MWO has many problems, one of which is the low population's effect on the matchmaker algorithms for solo-PUG and how that affects PSR calculations based on the baked-in assumptions the PSR algorithm makes about players.

When the que has fewer players it starts to drop filters (Tier, weight class, etc). When MM starts putting almost everyone into one 'bucket' that invalidates all of the player's PSR scores that match as PSR is calculated with the assumption that players won't be matched against others who have significantly different levels of skill & experience.

[Exilyth]

Imagine this:

The moment you press the 'quick play' button, someone jumps through your window, chains a bag of potatos to your legs and forces you to run a marathon against trained athletes.

That's what matchmaking can feel like at times.


In an ideal world, we'd have a wide spread of tiers, so you could e.g. be tier 30 and get matched against a tier 35 and it would still be balanced because the difference in skill would be smaller than a current tier 3 being matched against a current tier 4.

But then we're back at 'if we had more players' and 'positive bias PSR' ...

*sigh*

[Jackal Noble]

Big chain of potatoes. love it.

[Escef]

To be honest, I think the worst potatoes in solo-QP are people that rode into tiers 1 and 2 on the backs of coordinated groups. A little coordination goes a long way into covering up people's shortcomings in gunnery or mechbuilding. And it's easier to be aggressive when you know you have 3 or 4 friends at your side. But these guys in solo-QP? Oh, gawds... Atrocious builds, horrible awareness, no idea how to use their builds, and about as aggressive as a quadriplegic sloth.

[El Bandito]

I am not gonna complain much cause population is low and there is no quick fix to it. Also, potatoes are present on both sides, so in some twisted way the teams are still even-ish. I just have to tryhard and swing the victory towards our side. That's all there is to it.

[Nightbird]

There are quick fixes, just use a better measurement of skill than tiers and ensure both teams have an equal amount of it.

[Escef]

Nightbird, on 18 March 2018 - 05:25 PM, said:

There are quick fixes, just use a better measurement of skill than tiers and ensure both teams have an equal amount of it.

And that's going to increase population how?

[Nightbird]

Escef, on 18 March 2018 - 05:38 PM, said:

And that's going to increase population how?

Stomps happen because of unbalanced teams, repeated stomps reduce population.

[Escef]

Nightbird, on 18 March 2018 - 05:45 PM, said:

Stomps happen because of unbalanced teams, repeated stomps reduce population.

On the contrary, balanced teams make stomps more likely. In teams were everyone is equally valuable, losing 1 player is bad, and can end up snowballing into a stomp. In teams where not every player is equal, sure, losing your worst player doesn't hurt much, but losing your best is even worse than losing a player on the everyone is equal team.

Stomps are the result of one team making a mistake and the other exploiting it. Close games, however, are the result of BOTH teams making a lot of mistakes.

As for repeated stomps reducing the population, it can disenfranchise some people, absolutely. But balancing teams does not eliminate stomps, it just makes it harder to recover from a mistake.

In short, stomps are normal.

[Nightbird]

Escef, on 18 March 2018 - 06:11 PM, said:

On the contrary, balanced teams make stomps more likely. In teams were everyone is equally valuable, losing 1 player is bad, and can end up snowballing into a stomp. In teams where not every player is equal, sure, losing your worst player doesn't hurt much, but losing your best is even worse than losing a player on the everyone is equal team.

Stomps are the result of one team making a mistake and the other exploiting it. Close games, however, are the result of BOTH teams making a lot of mistakes.

As for repeated stomps reducing the population, it can disenfranchise some people, absolutely. But balancing teams does not eliminate stomps, it just makes it harder to recover from a mistake.

In short, stomps are normal.

LOL!!!!

[Escef]

Nightbird, on 18 March 2018 - 06:34 PM, said:

LOL!!!!

Laugh all you like. Go and look at pretty much EVERY conversation on this topic. You'll find people more knowledgeable than I am telling you the exact same thing.

[Nightbird]

Escef, on 18 March 2018 - 06:47 PM, said:

Laugh all you like. Go and look at pretty much EVERY conversation on this topic. You'll find people more knowledgeable than I am telling you the exact same thing.

Stomps are inevitable due to the randomness of the game, but having balanced teams making stomp more likely? Goodness. You gave me a good laugh, thank you sir!

[Krivvan]

Nightbird, on 18 March 2018 - 06:49 PM, said:

Stomps are inevitable due to the randomness of the game, but having balanced teams making stomp more likely? Goodness. You gave me a good laugh, thank you sir!

He's actually partially right, at least in a pub queue context, but it should perhaps be amended to say that balanced teams with good players on both sides can make a stomp more likely, as they're more likely to directly exploit the first mistake the other side makes (and with pub queue, there will often be mistakes everywhere).

It may be flawed to argue that an imbalanced team inherently makes a stomp less likely though. A team full of good players is likely to still stomp a team that isn't. But the fact that stomps are natural due to the nature of the game isn't wrong. Very even teams are likely to see many stomps, and close games in pub queue are indeed likely to occur more due to mistakes made on both sides rather than a perfect balance of skill.

So I'd argue the whole "what makes a stomp more likely" is more specifically about how you distribute the top 1% of players throughout the teams than how balanced a match is overall.

All this starts to go out the window when you talk about the dynamics of big groups vs. small or other big groups and tonnage restrictions and etc. So it mostly applies to solo queue.

That was a bit rambly though so I think the only real takeaway is that you should still often expect stomps even with perfectly balanced teams.

Edited by Krivvan, 18 March 2018 - 08:12 PM.

[Kiiyor]

It's.. it's not just the MM responsible for bad matches. There's a whole host of factors involved, and the MM has far less influence on the outcome of a match than people seem to realize.

The MM can't control mechs dropping with loadouts that don't compliment each other... (though it would be nice if it did)

It also can't control the tactics people use - defensive players can rack up great scores if a match is more stationary, whereas aggressive players can do well if they force the team into constant movement. The two don't always mix...

The MM can't prevent people dropping in bad robots.

It also can't prevent people from dawdling around spawn, or splitting up - or exposing themselves to a lot of focus fire at the start of a match, or getting out of position, or failing to move with the team, or failing to shoot down UAV's, or failing to deal with light mechs harrying assaults, or bunching up together and eating strikes, or going into the basement on HPG and wandering out the exits single file to die bewildered and irrationally angry at the rest of the team, or failing to identify high threat mechs, or failing to call targets...

...all of which have a far greater impact on the outcome of a match than the way the MM matches someone's XP bar with someone else's...

The MM is more of a convenient scapegoat than a villain, IMHO.

[Dragonporn]

PSR is inaccurate system, and there's nothing PGI can do about it. In fact, pretty much any system of any game that uses some parameters for skill calculation between players are quite inaccurate as well, providing very superficial comparison and matching of players.

Example: I play StarCraft 2 for pretty long time already. Game has around ~10k players online at average time I play. For 1v1 matchup, it takes less or around one second to find opponent for me. Every opponent I've met so far, judging by parameters, has almost the same MMR I have, and matched in the same leagues (Plat/Diamond). But it doesn't mean anything since 50% of my matches are stomps, either of me or my opponent. On the post game screen, I see that my opponent has ~400 APM with insane order precision and great knowledge of builds/mechanics, while other opponent barely gets to ~80 APM and make some brutal mistakes (viewable in replay), which player with such MMR and league simply cannot make. This isn't exclusively mine experience, it is very common for pretty much any SC2 player.

So, PSR and tier system in MWO in reality isn't half bad, and more importantly, with solo queue system and overall gameplay specifics, PSR gets pretty much irrelevant. There's simply no way to calculate and have even matchups for many various reasons. And yes, stomps happen absolutely everywhere. Complaining about it is silly.

[PocketYoda]

If both teams have the same level of skill no stomps aren't normal.. if the teams are bias one way or another yes they are, but that by definition is a bad game and the populations leave..

Stomps should not be normal if you want a healthy population.. until PSR and matchmaker gets properly fixed the people will continue to leave.. Until then it'll be what we have now, depressing garbage..

Luckily i only have to wait till april..

Edited by Samial, 18 March 2018 - 11:01 PM.

[justcallme A S H]

Krivvan, on 18 March 2018 - 08:03 PM, said:

He's actually partially right, at least in a pub queue context, but it should perhaps be amended to say that balanced teams with good players on both sides can make a stomp more likely, as they're more likely to directly exploit the first mistake the other side makes (and with pub queue, there will often be mistakes everywhere).

To be honest the more good names in a game I see, the closer it usually is to 12-6 or above.

If there is only one or two notable players in the entire 24, 2 good players on one side almost guarantee's a 12-2.

Of course there are examples where this isn't the case and the tide turns the other way, but for the most part if I see a good 3-4 players on each side it is, generally, a decent game as you have 8 players that won't make silly mistakes and can affect a decent change on a battle even 2-3 mechs down.

[Krivvan]

justcallme A S H, on 18 March 2018 - 11:12 PM, said:

If there is only one or two notable players in the entire 24, 2 good players on one side almost guarantee's a 12-2.

My opinion may also be impacted by the fact that, usually playing a Light, I rely quite a bit on how solid the core of the team is. I can't be the one to initiate a push for example, or take a position that leads to the rest of the team repositioning.

Edited by Krivvan, 18 March 2018 - 11:18 PM.

[justcallme A S H]

Yeah maybe that is it, you can't really dictate much in a light, more reliant on what else the team does.

Legit talking about this... Just has THIS GAME bunch of names on both sides and was reasonably solid with 2-3 'mini battles' going on simultaneously. Was actually a solid one (we were down 0-2 IIRC early-ish and a disco). I actually thought we would've lost and with the names on the other side, we prob should've.

[El Bandito]

BlueStrat, on 18 March 2018 - 04:06 PM, said:

The PSR system in MWO has many problems, one of which is the low population's effect on the matchmaker algorithms for solo-PUG and how that affects PSR calculations based on the baked-in assumptions the PSR algorithm makes about players.

Meh, even hugely popular FPS like Overwatch has just as many stomps. Good thing the matches end fast. I do think PSR was implemented wrongly.