24 replies to this topic

[Strelok7]

How many *rounds* in your gaming session end in one team completely obliterating another? Aka steamroll. Plus/minus 3 players is still acceptable.

This is a follow-up to the feature suggestion post that I had made. https://mwomercs.com...5-chat-balance/

P.S. Thanks for all the work Devs! And thanks to all the awesome players out there! GLHF 07

[dubstep albatross]

Strelok7, on 07 February 2025 - 06:12 PM, said:

How many *rounds* in your gaming session end in one team completely obliterating another? Aka steamroll. Plus/minus 3 players is still acceptable.

What do you mean by "plus/minus 3 players is still acceptable"? I think you are making a statement about your definition of a stomp (steamroll) being a difference in surviving team members of four or more. There has been lots of discussion in this forum, over many years, about what constitutes a stomp. The high level of the subject is that kill count by team, while definitely a signal, does not always tell the whole story.

Can you give us some acceptable outcomes? I ask because I have lots of data (thousands of matches, across multiple tiers) and could provide some numbers.

Out of 3,497 matches of data I have saved:

• About 89.5% (3,133) end with the defeated team having no survivors.
  
• The most common outcome is the victory team having seven survivors (about 14.6%).
  
• Given the bullet point above, most (about 93%) of the seven survivors for the victory team outcomes are 12:5 outcomes (victory team destroyed twelve, defeated team destroyed five).
  
• A 12:0 outcome is rare at about 1.4% (52 matches).
  
• About 15.6% (548) are +/- 3 survived, interpreted as a three survivor difference between teams (irrespective of victory). Examples: victory seven, defeated four, and victory three, defeated six.

[Saved By The Bell]

12-0 definitely not 1%, 10% may be.

[Meep Meep]

12-0 to 12-4 are blowouts in my book and they happen a lot depending on the time zone I am playing in. Late night especially when the sweaty lances tend to be concentrated in t1. Not the majority of games but its not too far off. Close competitive games that go down to the wire are the rare outcome.

[Hagen Zwosta]

Yes it is terrible. I simply quit when I got steam rolled twice in a row right at the start of a (planned) gaming session.
Yeah, I am therefore basically not playing any more.
The matchmaker plain sucks.

[sycocys]

For me w/l or whether I do end up in some actual rolls doesn't matter to whether I'm playing more matches, it's more if I'm seeing the same exact maps for the 4th 5th 6th times this session or if I'm bored with the game's extremely shallow play loop - a lot of times I just don't get into it after a match or two and do something else.

[Hagen Zwosta]

Maybe just leaving the match is an option. You can see how it will end after 20 seconds. All your team runs in a circle on low ground? Just leave.

[Saved By The Bell]

Just leaving suck, hide and take close one with you.

Some dead start crying "dont waste time". F**k them, you still alive and try to kill somebody. Bad, they can tell your position to enemies.

Edited by Saved By The Bell, 08 February 2025 - 06:27 AM.

[Hagen Zwosta]

Thats what I usually do. Just wait a small little until the enemies rotate into me and do as much as possible. Def. not running in a circle all the time.

[Saved By The Bell]

When I slow I do the same! There is no way to run from lights, lol.

[Moadebe]

"Blowout" matches or "stomps" are generally 12-4,3,2,1,0 or 4,3,2,1,0-12

And generally speaking its about 50-50 in games outside that range and within.

Are there days where you can have a variance of that? Absolutely.

But the "GOOD" matches where its close are the rarity.

Its all down to matchmaker honestly, but eh. There is a wall when it comes to that discussion.

Try to enjoy it as best as can. Use comms and do your best and try to change the outcome to the best of your ability.

[LordNothing]

there is a bias, i forget the name, where by people remember the bad games more than the good ones. if a match is even its not really worth remembering. if a match is really good you are usually disapointed by any match that doesnt duplicate it. i think true stomps are < 1/3, maybe < 1/4 by spitball estimate.

also if you happen to die in a good match that ends up winning, your brain usually counts that as a loss. especially if you quit before the conclusion.

[dubstep albatross]

I see the definition of a stomp being generally 12:4 or 12:3, and lower. There is also sentiment that these outcomes indicate poor matchmaking. A few comments:

• Poor matchmaking can certainly contribute and there are so many variables that this matchmaker, and in some cases any matchmaker anywhere, cannot really account for.
  
• The nature of this game, the non-respawn and fixed team sizes, combined with loss of damage output as the team loses mechs, does create a snowball effect. For a game to go "neck and neck" means a lot of very even trading and even loss of mechs. These outcomes are rare because these situations are rare.
  
• Even a perfect matchmaker will not change singling targets out, asymmetry in applied damage output, and so on. Not unless one person from each team go into a corner, pair by pair, and duke it out and the winner sits out the rest of the match or something.
  
• I generally feel that 12:2 or less is probably a good signal that the teams were effectively unbalanced. Overall, this seems to happen about 15% of the time in my data, starting at about 10% for Tier 5 and increasing to about 18% for Tier 2. I have no Tier 1 games.

I have spent some time looking more closely at other match data. There are many ways a 12:4 match can turn out when looking at match score, damage done, sequence of events in the match (kills by team indexed by time). It has been said before: the winning team could be completely stripped and mostly cored, the outcome could be 12:4, it could have been a really close match, but the way it worked out (and mostly always does) is one team ran out of mechs first. I will concede that this is probably rare, but the point is that the pure kill count does not always tell the story.

Saved By The Bell, on 07 February 2025 - 10:51 PM, said:

12-0 definitely not 1%, 10% may be.

In full disclosure, I have no Tier 1 games as the highest Tier of any account I have is 2. That said:

• There is an increasing trend for counts of 12:0 matches with increase in tier. 0.9% in Tier 5, 1.3% in Tier 4, 1.9% in Tier 3, and 2% in Tier 2. I do not have a lot of games in Tier 2, so take that percentage with a grain of salt. Still, 10% seems wildly high, but maybe Tier 1 games see this kind of snowball effect more often.

Meep Meep, on 07 February 2025 - 11:01 PM, said:

12-0 to 12-4 are blowouts in my book
...
Not the majority of games but its not too far off.

Moadebe, on 08 February 2025 - 09:20 AM, said:

"Blowout" matches or "stomps" are generally 12-4,3,2,1,0 or 4,3,2,1,0-12

And generally speaking its about 50-50 in games outside that range and within.

Globally (Tiers 2 through 5) in my data, using the 12:4 or less definition of a stomp, I see about 38% of matches fall into this category. However:

• I see a similar split for each tier, however there is also an increasing trend similar to the 12:0 match counts. About 30% in Tier 5, about 37% in Tier 4 and Tier 3, and about 41% in Tier 2.

LordNothing, on 08 February 2025 - 11:56 AM, said:

there is a bias, i forget the name, where by people remember the bad games more than the good ones.

Agreed. Bob ate a berry once and then he died, better remember to not eat that berry.

[Nomex 99]

Almost every game ends with a clear victory for one side. There is no longer a balance, and probably no matchmaker either.

[Strelok7]

dubstep albatross, on 07 February 2025 - 10:21 PM, said:

I ask because I have lots of data (thousands of matches, across multiple tiers) and could provide some numbers.

Sure. But don't spend too much time on it... How about One Team 10 survivors vs Another 0 Survivors. Actually... I see, you are saying it is about 15% of the time.

11 to 0, 9 to 0 ?

[Glymbol]

dubstep albatross, on 07 February 2025 - 10:21 PM, said:

I ask because I have lots of data (thousands of matches, across multiple tiers) and could provide some numbers.

Out of 3,497 matches of data I have saved

Are you interested in more data? I could save and send it to you. What do you need?

[Hagen Zwosta]

First 2 games today;

2-12 with me getting an = they even had one disco.
1-12 with me getting an ^

more than half team every time below 200 damage.
I have screenshots if you wanna see.
Same as yesterday or any day really.
Unplayable. And no fun.
What kind of matchmaking is this? Put all the people who can not aim in one team?

Edited by Hagen Zwosta, 09 February 2025 - 02:23 AM.

[dubstep albatross]

Glymbol, on 09 February 2025 - 01:46 AM, said:

Are you interested in more data? I could save and send it to you. What do you need?

Thank you for the offer. I will keep it in mind. I built a custom OCR pipeline. It requires a specific resolution because of how I built it. I have been considering reaching out to collect data from the community, but it may not necessarily be a simple task in the end. Plus, if I am being honest, I am a bit lazy.

[dubstep albatross]

Hagen Zwosta, on 09 February 2025 - 02:20 AM, said:

First 2 games today;

2-12 with me getting an = they even had one disco.
1-12 with me getting an ^

more than half team every time below 200 damage.
I have screenshots if you wanna see.
Same as yesterday or any day really.
Unplayable. And no fun.
What kind of matchmaking is this? Put all the people who can not aim in one team?

If I may ask, what tier are you in?

[Saved By The Bell]

today stomp defeats (0-12 to 3-12) is near 90%

so good luck for today

(most people is off so mm doing job)

Edited by Saved By The Bell, 14 February 2025 - 09:24 AM.