52 replies to this topic

[LTC Kilgore]

Prototelis, on 05 July 2019 - 03:05 PM, said:

It was implied homie; same as when anyone blames the matchmaker exclusively for a long streak of losses.

but, you said the term "only consistent factor". Well, the "only consistent factor" in the Op's wins is the OP. You cannot make the claim of one factor being what decides the outcome of a match for one particular outcome, then dismiss that very same factor when the outcome is different.

[Prototelis]

Of course I can.

If you lose more often than you win it is a reasonable assumption you aren't contributing much to those wins on average.

Edited by Prototelis, 06 July 2019 - 01:11 PM.

[LTC Kilgore]

Prototelis, on 06 July 2019 - 01:06 PM, said:

Of course I can.

If you lose more often than you win it is a reasonable assumption you aren't contributing much to those wins on average.

thats just it, an assumption. It assumes that the matchmaker is making perfectly balanced teams every time. it assumes that all other 23 players in the drop are playing to their abilities. It assumes that the OP is playing below his rating in every loss. It also assumes that in each win, the OP is playing at or below his rating.

All those assumptions are seriously flawed.

[Prototelis]

Except that we have large sample sizes.

[Azhrael]

@ Prototelis.

Your valiant and unrelenting effort to avoid the point of the topic is dully noted, you can move along now. Thank you.


Now, i also noticed this weird streak for the last 3 days or so. We also had pretty high number of " Newbies " of the order of asking what the objectives meant, aimlessly walking in random directions, ignoring the overall direction of the enemy spawn, ridiculous builds, etc etc.

Just for reference, im T2, coming back after a very long hiatus, but still consistently pulling around 300 average score....i was of the idea i shouldnt be matched with new people...?

I mean, of course there could be the odd UBER NOOB that reachs T3 and STILL cant remember the maps i guess....but c'mon.

Still, so many people not landing a single volley, dying while running away from the enemy, half the team consistently pulling 2 digits scores......i dont know, its really strange.

That last one brings me to the infamous " Its ppl farming the Event " argument.:

Hm...So you are trying to tell me that people purposedly make an effort to do the minimal score possible in order to get Bags more quickly. Then quit the match, grab another chassis and repeat? That's the argument? Ok here's whats painfully ****** with that :

1) You want to win things for your game quickly, but you evidently dont want to play it . (¿?¿?¿)

2) You have what? a week? 2 weeks to finish a regular challenge? whats the hurry?

3) By purposedly throwing the game, you lose potential money.

4) " " " ", you lose potential xp.

5) " " " ", you destroy your own ranking, something that you will sorely regret when/if you want play seriously later.

In short, when you sum up all of the above and others i may miss, you find yourself saving some time to gain the bags, but losing an awful lot in the mid/long term, This is true for ANY FPS game, like ever.

So, unless most of the current playerbase is totally ****** ( In the scope of me and several other echoing these concerns ) , there seems to be something really weird going on with Match Making.

Tnx for reading.

[Prototelis]

Azhrael, on 07 July 2019 - 09:43 PM, said:

@ Prototelis.

Your valiant and unrelenting effort to avoid the point of the topic is dully noted, you can move along now. Thank you.

No.

Quote

Tnx for reading.

TL;DR

[Nightbird]

I changed the MM deliberately to mess with you and only you, so no one else will believe you. Muahahaha

[Kubernetes]

LTC Kilgore, on 05 July 2019 - 01:44 PM, said:

if you play at a high level and lose, its your fault?

To some extent, yes. Most matches come down to the performance of the best 2-3 players on each side. If you're one of those guys and you die prematurely, that's probably the biggest factor in a loss.

[Azhrael]

Kubernetes, on 08 July 2019 - 04:14 PM, said:

To some extent, yes. Most matches come down to the performance of the best 2-3 players on each side. If you're one of those guys and you die prematurely, that's probably the biggest factor in a loss.

I would say "In very limited " cases, really. When you talk about High level playing, individual overall skill gives way to individual skill in a role. A simple example would be you are a very good light pilot, but a terrible shooter. If you are playing in your role, you will do fine, if for some reason you end up as a Dakka Assault, you ll do poorly.

In fact, the ability to "single-handle carry " a game diminishes greatly the higher the overall quality of both teams.

Put in another way : in a low tier game you may have a match where your 2 best pilots survive to the end, vs the worst 5 on the other team, making possible the " carry "

Not so in a high tier game, where all things being equal, your 2 survivors are as good as the enemy's 5.


In short, in a top Tier match, every player contributes an equal part to the 100% efficiency of the team, assuming each one is playing to the best of its abilities. A single error is not devastating, a High Tier team is able to adapt its tactics to compensate for it.

In a low Tier match, the oppossite occurs, where a couple players could be much better than the rest, or another couple brought ineffective builds, or chassis they are not familiar with, etc etc. Every team, in average, will have individual players contributing MORE to the 100% effectiveness not just because they are better, but if only for bringing an usable mech when someone else dont, knowing the best map approaches, using minimap intel, etc.

[Kubernetes]

Azhrael, on 08 July 2019 - 04:46 PM, said:

In short, in a top Tier match, every player contributes an equal part to the 100% efficiency of the team, assuming each one is playing to the best of its abilities.

That's not even close to being true. In any given T1 match there are several players who contribute significantly more than the average player. If you think about the sports statistic VORP (Value over replacement player), the same concept applies here. A guy like Proton is probably 6x or 5x the value of the average T1. The typical T1 match features a lot of T1s, T2s and T3s who need to be carried, a handful of nominal T1s who can hold their own, and a few face-melters who do inordinate amounts of effective damage and largely determine the outcome of the match. T1 is full of "XP-bar T1s" who are only there because they've played a lot; these guys still bring terribuilds and get dismantled by truly effective players.

[Wil McCullough]

Azhrael, on 08 July 2019 - 04:46 PM, said:

I would say "In very limited " cases, really. When you talk about High level playing, individual overall skill gives way to individual skill in a role. A simple example would be you are a very good light pilot, but a terrible shooter. If you are playing in your role, you will do fine, if for some reason you end up as a Dakka Assault, you ll do poorly.

In fact, the ability to "single-handle carry " a game diminishes greatly the higher the overall quality of both teams.

Put in another way : in a low tier game you may have a match where your 2 best pilots survive to the end, vs the worst 5 on the other team, making possible the " carry "

Not so in a high tier game, where all things being equal, your 2 survivors are as good as the enemy's 5.


In short, in a top Tier match, every player contributes an equal part to the 100% efficiency of the team, assuming each one is playing to the best of its abilities. A single error is not devastating, a High Tier team is able to adapt its tactics to compensate for it.

In a low Tier match, the oppossite occurs, where a couple players could be much better than the rest, or another couple brought ineffective builds, or chassis they are not familiar with, etc etc. Every team, in average, will have individual players contributing MORE to the 100% effectiveness not just because they are better, but if only for bringing an usable mech when someone else dont, knowing the best map approaches, using minimap intel, etc.

This is untrue in a lot of ways. There are no "roles" in the game. Everything boils down to how good you are at killing enemy mechs. It doesn't matter if you're awesome at "scouting" or "harassing" or whatnot. If you're not killing enemy mechs as quickly and efficiently as possible, your contribution decreases significantly. That's why many players frown at lurmers. They cause padded damage stats but very little killing ability due to the spread damage and inaccuracy.

Matches are decided by which team has the better ability to kill quickly. The best players are the ones who can do that one thing extremely well and to keep doing it. Most of the time in pubs, this results in a high damage score, which is a byproduct of killing enemies.

Considering rank uses match score to determine your tier, and match score is heavily biased towards accumulated damage rather than wlr and kdr, you get a lot of variety of skill in every game. I'm t1 and in the top 10% of pilots. I'll be dropped with fellow t1s who can be at the 70th percentile or even t1s who are in the top 3% like lizzee, proton, magic pain glove etc.

I will win like 3 out of 4 games if there's such a player on my team and lose 3 out of 4 times if they're on the opppsite team. You can't say "everyone contributed equally" when there are players that can put out 2.0++ wlr playing solo without breaking a sweat while there are players who cant get over 1.0 wlr even at t1.

[Azhrael]

It appears i should have been more specific. That is why i never used this game TIER SYSTEM on my previous post. In fact my statement is general, for any kind of game.

I understand now that the term " High/low Tier " is deeply encroached in MWO conversation as a reference to "Tier 1/ Everyone else " but it was never my intention to use it as such. I was simply referring to " pro level " and " casual level "

My mistake.

Editing my previous post would be silly, so i hope this clarification is enough:

@Kubernetes

Your points are mostly valid when talking of MWO specifically. That the MWO Tier system is " inappropiate" is broadly and extensively agreed upon by the majority of the community.

As you point out, someone like this Proton guy can pull those numbers IN THIS GAME TIER 1 environment. But if this Tier 1 level was actually working as intended ( A Pro Tier level ), he would be facing another " Proton level players " almost all the time, which would impact his numbers accordingly and he, individually, wouldnt be able to carry the match, i pointed at this on my " high Tier " example.

Ofc there would exceptional players that even in these conditions would surpass his peers, the so called 1% , best of the best, maybe this Proton is one of them, idk. But this minuscule impact is of no consecuence on the overall issue.

@ Will

I...really have to disagree on that one. No roles on this game?...i mean, you realize that you even have classes, restrictions, skills, exclusive modules, etc etc. right?.

Then, you seem to contradict yourself with the next paragraph " Matches are decided by which team has the better ability to kill quickly... []"

Well yeah. you said it yourself, if you happen to have lurms in your team BUT coupled to a good scout, they both will do exponentially better than alone. Now if the light just go hide to try and backstab a lonely assault that got lost...your lurms wouldnt do so well. Or the typical Nascar where your assaults are left behind unescorted....etc etc.

Still i believe you refer to the fact that only pro level players would make effective use of this, and the vast majority of players just ignore it. That would be different but also true, and i agree on that.

Tnx for reading.

[Wil McCullough]

@azhrael

You seem to be saying that as long as a player does something, it counts as contribution. Like as long as a lurmer lobs a few missles in the direction of the enemy, he is contributing to his team's victory.

I don't agree.

[Azhrael]

Well, that example is a bit extreme..but we should be able to agree that a lurmer actually landing most of his lurms in a match is the best possible outcome that a lurmer could hope to achieve.

Now a " superb " lurmer would also find time and position to...idk, maybe support a team mate in a 1v1, or saturate an enemy assault when your team charges into his position in order to reduce his effectiveness, Even draw away enemy skirmishers from an impending showdown by positioning far out of the way and appear like an " Easy kill ".The examples are myriad.

I know most chassis can do this, but not indirectly and from any position and ( arguably ) any distance. That would be the Role of the Lurme AND for this role to be really effective, the Lurmer needs constant EYES on the target, that would be the role of the Scout: Keep survelliance for as long as possible while not being discovered/ dying. Be it by speed, stealth or ( in the less efficient cases ) sheer resilience.

Still far from me to try and change your opinion, i totally understand it based on my own experience on this game. I certainly understand that these concepts, common in other games, are kind of alien on this player base and i find it sad, even infuriating some times.


Im speaking of an ideal situation and you are speaking from realities, i suspect both of us are correct.

Lets agree to disagree?



Edit : Clarifying edit.

Edited by Azhrael, 09 July 2019 - 06:48 AM.

[JediPanther]

During events I just expect mm to throw together teams fast as possible in qp. I recorded a match with a known assault pilot with questionable behavior then compared out damage and kills at the end of the match. I was using an lct-3s with srm 4 srm6 and ersml. The assault pilot an atlas-d(f) with an unknown build.

It was a poor match with neither of us getting kills but...
his damage:156
his ms: 94
mine dmg:337
my ms:194

Nearly double damage in a light with a poor build. Sure I would have done better with a different mech and load out but then when other players can't meet or beat your output even when you intentionally run a non-meta sub-par mech/build it doesn't matter what you use. You'll still get stuck with those players that just can't do as well as you even when you get lumpted into a mix team where one assault can make you look bad on the high damage end of match screen.

[Kubernetes]

Azhrael, on 08 July 2019 - 10:30 PM, said:

@Kubernetes

Your points are mostly valid when talking of MWO specifically.

What else are we talking about?

Quote

As you point out, someone like this Proton guy can pull those numbers IN THIS GAME TIER 1 environment. But if this Tier 1 level was actually working as intended ( A Pro Tier level ), he would be facing another " Proton level players " almost all the time, which would impact his numbers accordingly and he, individually, wouldnt be able to carry the match, i pointed at this on my " high Tier " example.

Your hypothetical is irrelevant. PGI doesn't have a better MM. It doesn't have a way to gauge player skill past PSR. I used Proton as an example because he is recognizably one of the best of the best in high level comp. Sure, his SQ performance would probably come down some if he only played with and against other comp tier players. But then they'd probably all quit because their matches would be so infrequent. Regardless, the game we've got is the game we've got, "THIS GAME TIER 1 environment" is the only T1 environment, and T1 matches are exactly as I describe: 2-5 face-melters, 5-10 good T1s, 10-15 targets. This notion that everyone contributes equally is bs. If that were true, everyone would have a 1.0 WLR.

Edited by Kubernetes, 09 July 2019 - 05:31 PM.

[Tatula]

JediPanther, on 09 July 2019 - 04:37 PM, said:

During events I just expect mm to throw together teams fast as possible in qp. I recorded a match with a known assault pilot with questionable behavior then compared out damage and kills at the end of the match. I was using an lct-3s with srm 4 srm6 and ersml. The assault pilot an atlas-d(f) with an unknown build.

It was a poor match with neither of us getting kills but...
his damage:156
his ms: 94
mine dmg:337
my ms:194

Nearly double damage in a light with a poor build. Sure I would have done better with a different mech and load out but then when other players can't meet or beat your output even when you intentionally run a non-meta sub-par mech/build it doesn't matter what you use. You'll still get stuck with those players that just can't do as well as you even when you get lumpted into a mix team where one assault can make you look bad on the high damage end of match screen.

I'm not sure what you would consider a "known assault pilot" but I have notice that how an assault mech performs in a match depends a lot on how its team mates behave. I often see one or two assault mech getting under 200 damage on NASCAR-prone maps. What happens is lights and mediums run ahead, heavies and fast assaults follow, and the slower or more timid assaults get left behind and focused down quickly by the enemy team's NASCARing lights and mediums. If the don't have ECM, then they would draw fire from the LRM boats. They often get destroyed before they get to the main fight.

[Feral Clown]

Azhrael, on 07 July 2019 - 09:43 PM, said:

@ Prototelis.

Your valiant and unrelenting effort to avoid the point of the topic is dully noted, you can move along now. Thank you.


Now, i also noticed this weird streak for the last 3 days or so. We also had pretty high number of " Newbies " of the order of asking what the objectives meant, aimlessly walking in random directions, ignoring the overall direction of the enemy spawn, ridiculous builds, etc etc.

Just for reference, im T2, coming back after a very long hiatus, but still consistently pulling around 300 average score....i was of the idea i shouldnt be matched with new people...?

I mean, of course there could be the odd UBER NOOB that reachs T3 and STILL cant remember the maps i guess....but c'mon.

Still, so many people not landing a single volley, dying while running away from the enemy, half the team consistently pulling 2 digits scores......i dont know, its really strange.

That last one brings me to the infamous " Its ppl farming the Event " argument.:

Hm...So you are trying to tell me that people purposedly make an effort to do the minimal score possible in order to get Bags more quickly. Then quit the match, grab another chassis and repeat? That's the argument? Ok here's whats painfully ****** with that :

1) You want to win things for your game quickly, but you evidently dont want to play it . (¿?¿?¿)

2) You have what? a week? 2 weeks to finish a regular challenge? whats the hurry?

3) By purposedly throwing the game, you lose potential money.

4) " " " ", you lose potential xp.

5) " " " ", you destroy your own ranking, something that you will sorely regret when/if you want play seriously later.

In short, when you sum up all of the above and others i may miss, you find yourself saving some time to gain the bags, but losing an awful lot in the mid/long term, This is true for ANY FPS game, like ever.

So, unless most of the current playerbase is totally ****** ( In the scope of me and several other echoing these concerns ) , there seems to be something really weird going on with Match Making.

Tnx for reading.

Um...you are actually not pulling a consistent 300 match score. Currently in season 36 you at 226 which is about the same as you have done historically. Why say stuff like that when everybody can look up the current leader board on the web or in game?

[Prototelis]

Had a dude in GQ last night tell us he had a 25:0 KDR lol.

[Azhrael]

@ Feral clown.

Because never tried to look up myself historically, was irrelevant to the point. Im T2, and im asking if it is correct for me to be matched with new people.

I guess i mentioned it then because i wasnt sure if it could affect my personal matchmaking, but further investigations now tell me that its not the case.

Now that my e-p3n1s size has been properly measured and the troll fed, do you have an answer for that question?


@ Kubernetes

I joined the discussion with the idea of sharing my experience of " how should it be " or IS in other successful implementations. We all already know " how it is " and most of us already agree is not working right.

But i cant help but try to find alternatives or suggestions, instead of explaining WHY the system is not working over and over again.

Maybe i misunderstood the idea behind the Thread? I tell you the same i said to Will : I suspect both of us are right,

Edited by Azhrael, 10 July 2019 - 06:30 PM.