35 replies to this topic

[DaZur]

ImperialKnight, on 17 January 2021 - 01:54 AM, said:

MWO is a snowball game. It comes down how much damage is traded at the correct spots and when. early critical hits/kills matter.

Too few players truly understand this... It's basic math and laws of averages.

My findings are pretty straight forward and boils down to compound team strength. In short, once a team loses 3 to 4 teammates (Kind'a depends on what those mechs were, what their combined shared armor and firepower were)... The opposing teams composite strength is too much to overcome and is essentially overwhelmed.

No match-maker boogieman... No solo or premade boogieman... No [insert excuse here] boogieman.

With few exception, once a team is reduced by a third, it's extremely hard to climb out of that hole. Not impossible and I've seen and been party to it... But it's a hard row to hoe.

[General Solo]

DaZur, on 19 January 2021 - 08:16 PM, said:

Too few players truly understand this...

My findings are pretty straight forward and boils down to compound team strength. In short, once a team loses 3 to 4 teammates (Kind'a depends on what those mechs were, what their combined shared armor and firepower were)... The opposing teams composite strength is too much to overcome and is essentially overwhelmed.

No match-maker boogieman... No solo or premade boogieman... No [insert excuse here] boogieman.

With few exception, once a team is reduced by a third, it's extremely hard to climb out of that hole. Not impossible and I've seen and been party to it... But it's a hard row to hoe.

Why does the enemy allow this

Could it be skill gap

and Thus matchmaking

[Elizander]

General Solo, on 20 January 2021 - 03:15 AM, said:

Why does the enemy allow this

Could it be skill gap

and Thus matchmaking

Matchmaking won't ever take into account what type of mech a player brings or if they are leveling a new mech or trying something out. A player's performance can change drastically depending on what they are playing or what they want to do so the more inconsistent ones like me are kept down which is good.

Of course since the MM won't know if I'm leveling a 0 skill mech I'm unfamiliar with, trying out a new build, playing a weight class I'm no good at, or ready to tear things up with my Scorch or some other favorite mech... it won't really have any idea at any given time if I'm going to perform like a T5 cadet going down with under 100 damage or a T3 player that's capable of getting 3-4 KMDDs and 800+ damage in one game.

It also can't determine if a player is aggressive because let's face it, it's no top-of-the-line AI so it can't just put enough aggressive players on both teams to even out. It can't even match up LRM boats and I've had games where one team gets 6+ of them.

Of course, the more things they require of the MM, the longer it is to find matches because instead of generalizing players by tiers it'll factor in a lot of other things that can eliminate players from being qualified for a specific match.

We can't talk Match Maker without diving into the details of what data is required from each player to determine a good match, but that would require a whole other thread and we don't even have enough info on population and player data to determine if anything we come up with would be feasible.

Edited by Elizander, 20 January 2021 - 09:20 AM.

[Gagis]

The snowball effect happens in competitive play too. If one player makes a mistake and gets taken out early, the advantage given by fighting 8 to 7 is substantial and is likely to cascade into something like a 8-1 end result even if both teams started out on an even level.

What matchmaker affects is just how likely each team is to be the one to get the snowball rolling. If it is working well, both teams should be equally likely to win, even if that win is far more likely to be overwhelming than a close call.

Both teams failure cascading simultaneously at even rates until there are just a few mechs left on both teams is far less likely to happen, even with balanced teams. Those matches can be frustrating too, as you have to endure two teams racing to the bottom making mistakes and failing to take advantage of their opponent's mistakes. Games that are not stomps can be garbage too.

But if you can see from the starting screen which team it is that is more likely to take and hold momentum, then the matchmaker has failed to balance them, stomp or no stomp.

[R Valentine]

DaZur, on 19 January 2021 - 08:16 PM, said:

Too few players truly understand this... It's basic math and laws of averages.

My findings are pretty straight forward and boils down to compound team strength. In short, once a team loses 3 to 4 teammates (Kind'a depends on what those mechs were, what their combined shared armor and firepower were)... The opposing teams composite strength is too much to overcome and is essentially overwhelmed.

No match-maker boogieman... No solo or premade boogieman... No [insert excuse here] boogieman.

With few exception, once a team is reduced by a third, it's extremely hard to climb out of that hole. Not impossible and I've seen and been party to it... But it's a hard row to hoe.

Losing 3-4 players first is almost always a match making issue. The weaker, more careless players always fall to the top level players within 60 seconds of the timer starting. You have players who know the strong spots of the map, the spawns, and the most likely paths for enemies to take by memory vs. players just sort of blundering around or who only know how to play one way, and they fall in very short order. Games were both teams were able to gather strength before engaging, or at the very least having players quickly reach the stronger map points at around the same time, end up with a more balanced outcome.

[Vlad Ward]

Kiran Yagami, on 20 January 2021 - 07:07 AM, said:

Losing 3-4 players first is almost always a match making issue. The weaker, more careless players always fall to the top level players within 60 seconds of the timer starting. You have players who know the strong spots of the map, the spawns, and the most likely paths for enemies to take by memory vs. players just sort of blundering around or who only know how to play one way, and they fall in very short order. Games were both teams were able to gather strength before engaging, or at the very least having players quickly reach the stronger map points at around the same time, end up with a more balanced outcome.

Or luck? A properly functioning matchmaker should give you a 50/50 chance to win a match. That win/loss is probably going to be decided in the first 3-4 kills of the game. One misplay can screw a match, even if the player who misplayed is otherwise strong. **** happens.

Edited by Vlad Ward, 20 January 2021 - 07:17 AM.

[R Valentine]

Vlad Ward, on 20 January 2021 - 07:17 AM, said:

Or luck? A properly functioning matchmaker should give you a 50/50 chance to win a match. That win/loss is probably going to be decided in the first 3-4 kills of the game. One misplay can screw a match, even if the player who misplayed is otherwise strong. **** happens.

Then why are there so many people on the extremes of the W/L ratio? Most of the losing team's I'm on include the same players from a previous loss. People who literally learned nothing from an event that just occurred 5 minutes ago. Not only that, if the previous match was a stomp, why does the next match have 80% of the same players on the same teams as the previous match? Wouldn't a balanced match maker remix those teams in an attempt to change the outcome from the previous match? Surely that's the least it could do, but it doesn't happen. **** happens, but match maker is AOK with the same **** happening multiple matches in a row.

[Haipyng]

The Matchmaker just cannot be smart with the small pool of players we have left. They have safeties built in that keep people from waiting ages for a match. Those safeties allows lopsided matches based on skill level (highly skilled players can get lumped on one side), weight (more assaults (health and dps) on one side, or mechanics (tons of long/short, cool/hot builds on a map that favors that particular mechanic) on one side and not the other). It all allows for many opportunities for uneven matches, leading to stomps (and cranky unhappy players, which don't stick around)

Starting from the point where they started restricting player choice (removing a place for small groups, removing real map and game type choice for a perceived vote 'choice' (which killed real light viability), and reducing the viability of old mechs by introducing far better newer mechs all in the name of chasing the sale of new mech packs, has eroded the player base. They burned player goodwill and entered the coup de grâce by abandoning Steam for MW5 after preselling it on that platform.

The only fix is increasing the player base. That means fixing all of these issues and more in ways that they can't really charge for (short of rereleasing it as a new title). Solaris ignored MWO's glaring problems and introduced a new game mode that nobody asked for in the name of making it an e-sport. None of this is new. Players have been advocating these things for years with little heed paid. It's sort of like being in a plane that the pilot turned the engines off on. You know its coming down eventually and there is nothing we the passengers can do about it.

[Vlad Ward]

Kiran Yagami, on 20 January 2021 - 07:33 AM, said:

Then why are there so many people on the extremes of the W/L ratio? Most of the losing team's I'm on include the same players from a previous loss. People who literally learned nothing from an event that just occurred 5 minutes ago. Not only that, if the previous match was a stomp, why does the next match have 80% of the same players on the same teams as the previous match? Wouldn't a balanced match maker remix those teams in an attempt to change the outcome from the previous match? Surely that's the least it could do, but it doesn't happen. **** happens, but match maker is AOK with the same **** happening multiple matches in a row.

That's likely because PSR movement isn't fast. If the configuration of teams you had in Match 1 was what the MM determined to be the best fit, it's likely still going to be the best fit for several more matches. Individual players would need to be absolutely wrecked in order for their PSR to significantly drop, which means maybe 3-4 players would fluctuate over time. The bulk of the match would be expected to stay the same, though.

W/L is also not very heavily weighted in the PSR movement calc. If you lose but pull your weight, your PSR is unchanged. You can call a match a stomp, but was it really? Were people running 1-2 digit match scores on your side while the opposing team was all 300+? Kill count is a pretty poor indicator of how close a match was in a game with such high TTK.

[Gas Guzzler]

Kiran Yagami, on 20 January 2021 - 07:07 AM, said:

Losing 3-4 players first is almost always a match making issue. The weaker, more careless players always fall to the top level players within 60 seconds of the timer starting. You have players who know the strong spots of the map, the spawns, and the most likely paths for enemies to take by memory vs. players just sort of blundering around or who only know how to play one way, and they fall in very short order. Games were both teams were able to gather strength before engaging, or at the very least having players quickly reach the stronger map points at around the same time, end up with a more balanced outcome.

Um yeah that's a load of crap.

I've been the first dead before. I've been careless and got myself into bad spots early on or thought I had someone isolated only to realize that they weren't isolated and had buddies hiding around the corner. I've been a slow fatty trying to keep up with my team and just can't because of where I spawned and/or how fast they are nascaring. Or I'm slow and was just barely able to catch up to my team before they all quickly change their minds and go somewhere else leaving me tanking the other teams push.

**** happens, and it happens a lot in quick play. Especially if you aren't mobile, because you are at the mercy of your team to not leave you to die because you can't reposition quickly enough. There are strong spots in the maps, and I can't tell you how many times I knew my team was doing something stupid but since I'm 50 kph bringing up the rear, I have no choice but to try to keep up, or stand and deliver. Neither typically works. Even if I took up the "strong spot" if my team is going somewhere else its not like I'm going to last very long alone.

There are thousands of variables at play, and drawing conclusions like "the worst players on the team are the ones that die first" is nonsensical.

Also, the matchmaker can't predict whether I'm in a Mad Cat MCII or an Incubus. And I guarantee my performance will not be the same for those.

[R Valentine]

Gas Guzzler, on 20 January 2021 - 09:14 AM, said:

Um yeah that's a load of crap.

I've been the first dead before. I've been careless and got myself into bad spots early on or thought I had someone isolated only to realize that they weren't isolated and had buddies hiding around the corner. I've been a slow fatty trying to keep up with my team and just can't because of where I spawned and/or how fast they are nascaring. Or I'm slow and was just barely able to catch up to my team before they all quickly change their minds and go somewhere else leaving me tanking the other teams push.

**** happens, and it happens a lot in quick play. Especially if you aren't mobile, because you are at the mercy of your team to not leave you to die because you can't reposition quickly enough. There are strong spots in the maps, and I can't tell you how many times I knew my team was doing something stupid but since I'm 50 kph bringing up the rear, I have no choice but to try to keep up, or stand and deliver. Neither typically works. Even if I took up the "strong spot" if my team is going somewhere else its not like I'm going to last very long alone.

There are thousands of variables at play, and drawing conclusions like "the worst players on the team are the ones that die first" is nonsensical.

Also, the matchmaker can't predict whether I'm in a Mad Cat MCII or an Incubus. And I guarantee my performance will not be the same for those.

It's not non-sensical at all. "**** happens" is just as much an acknowledgement that it happening is a rarity as much as it is that it happens. Plenty of people get away with rolling slow mechs on the outskirts of the map while their team furiously spins in circles as fast as humanly possible. You roll up the rear going 50 because you choose to, not because you have no choice. I've rolled along sluggishly in a quad peep mech, brawling enemies at 250m just as much as I've chilled in the back of the map near the border shooting from 1000m. Neither situation was forced upon me. You pays your money and you takes your chances, but you have even half an inkling of what you're doing in either situation you're already leaps and bounds ahead of the majority of the player base. Weak players will fail regardless of which tactic they choose. Strong players will prevail more often than they don't. If everything really was just random chaos, like you all keep claiming, there wouldn't be so many players on the extremes of W/L and average match score as there are. Things would settle more naturally. But the numbers don't bear that out. The people who put more thought into how they play regardless of play style succeed more often and eek out high scores even in 0-12 losses while their teammates continue to middle about in mediocre to low scores regardless of match outcome. Pretending the weaker players don't die first is a denial of reality.

[Gas Guzzler]

Yeah we do prevail more often then we don't. That doesn't mean that **** doesn't happen.

[Vlad Ward]

Who/where are these extremes you're talking about? The tip top players in the game are immensely better than the other 99%. I'm ranked 96th percentile in Jarl's and I'm trash. Unless #1 and #2 are literally playing against each other every match, you're going to see some extremes at the top and bottom edges of the curve.

[DaZur]

General Solo, on 20 January 2021 - 03:15 AM, said:

Why does the enemy allow this

Could it be skill gap

and Thus matchmaking

No one "allows" it to happen.. lol. Call it bad luck, bad positioning, wrong mech for the wrong map... Sometimes even the best do stupid stuff and get insta-gibbed.

Yes, skill and knowledge does influence outcome but I've observed high skill teams get wiped just like random potatoes. The common denominator still is composite team strength and the 1/3rd tipping point.

[Wildstreak]

Gas Guzzler, on 19 January 2021 - 02:00 PM, said:

Moral of the story, stop yelling at people for the duration of the match. It doesn't help.

I don't, haven't used my mic in a while.

I do sit there either in amazement, laughing or both at things like someone wanting to push one spot not seeing half the team including all the Assaults are competing for the exact same corner spot to poke, duck back, poke, duck back, etc., etc.

Was actually amazed a LRMboater repeatedly shot me in the back one match.

I swear half the players must be doing more damage to their own team than the enemy.

[General Solo]

Gas Guzzler, on 20 January 2021 - 09:14 AM, said:

Um yeah that's a load of crap.

I've been the first dead before. I've been careless and got myself into bad spots early on or thought I had someone isolated only to realize that they weren't isolated and had buddies hiding around the corner. I've been a slow fatty trying to keep up with my team and just can't because of where I spawned and/or how fast they are nascaring. Or I'm slow and was just barely able to catch up to my team before they all quickly change their minds and go somewhere else leaving me tanking the other teams push.

**** happens, and it happens a lot in quick play. Especially if you aren't mobile, because you are at the mercy of your team to not leave you to die because you can't reposition quickly enough. There are strong spots in the maps, and I can't tell you how many times I knew my team was doing something stupid but since I'm 50 kph bringing up the rear, I have no choice but to try to keep up, or stand and deliver. Neither typically works. Even if I took up the "strong spot" if my team is going somewhere else its not like I'm going to last very long alone.

There are thousands of variables at play, and drawing conclusions like "the worst players on the team are the ones that die first" is nonsensical.

Also, the matchmaker can't predict whether I'm in a Mad Cat MCII or an Incubus. And I guarantee my performance will not be the same for those.

I think your nit picking, can't see the forest for the trees
What he said was thunder mentally true
Sure we all make mistakes, top players included

But if the clueless play the clued up that is totally a matchmaking issue

Plus soloes and groups are not exactly united, have differing motovations
Which makes it worse

When the last solo has left town, you are left with a group que

We know what happened to the last one

I didn't make it, couldn't get guid to be viable

Due to matchmaking

Vlad Ward, on 20 January 2021 - 10:48 AM, said:

Who/where are these extremes you're talking about? The tip top players in the game are immensely better than the other 99%. I'm ranked 96th percentile in Jarl's and I'm trash. Unless #1 and #2 are literally playing against each other every match, you're going to see some extremes at the top and bottom edges of the curve.

Trash is relative

to T5's

I'm sure you would be viewed as a Lord with 8 kills often

Edited by General Solo, 21 January 2021 - 03:02 AM.