18 replies to this topic

[Craig Steele]

So I have been banging on about matchmaking and thought I would throw my snapshot of grinding games just now and you can draw your own conclusions.

First 3 games were played with griffons config'd for mobile indirect support of friendly lights (fast and LRM's)

Second 3 were all Wolverines with various Light HK loadouts (mainly streaks and lasers)

Result K/D Our Assaults Their Assaults Comment
Win 12/4 6 1 Stomp
Win 12/4 1 2 Stomp
Loss 3/12 3 4 Stomp
Loss 7/3 4 7 Stomp (Spider capped us though)
Win 12/7 2 6 Entertained, good scrap
Loss 5/12 0 3 Stomp

Where there were kills in stomps for the losing side it was usually towards the end when pilots fell chasing the KS with their beat up mechs, the kills never influenced the game result. In the last match it was 3 / 5 early on and then turned into a stomp when their assaults arrived (as if losing before they arrived was not stompy enough)

I don't feel like these stats are much different from my overall MWO experience lately, which I would summarise as;

Winning as many as I am losing, but it really is a lottery
Usually stomping one way or the other, lack of entertainment
Team compsitions are often (as in half the time or more) lousy

Whether it MM or Meta, something is out of whack.

[Davers]

It is always going to be a lottery until you can 1. Have better communication or 2. Control who you drop with or against.

[Sandpit]

I played for several hours yesterday evening and night.

I didn't track the actual stats but I had a mixed bag of results.
Got stomped a few times
Stomped a few times (getting stomped and stomping were about even)
Most of our games at least half our force was down
had a few where we should have won but they were smart and capped us

We did have one game that was a bit one-sided (that we won actually)
I was the only assault on our team (My trusty crusty Stalker)
The other team had a full lance of Atlas-DDC and 2 other assaults (highlander and I want to another atlas? but they had 6)
That one kinda sucked but they weren't very good and we wound up winning (I took down 2 Atlases before I went down in a blaze of glory )

[Artgathan]

I think it's time for PGI to add in the % Health remaining to each mech in the End of Round screen. I think this would lead to fewer players claiming that "stomps" are as prevalent as they are. Consider that when you win (or lose) 12 - 4 the remaining 8 mechs might just be an MG hit away from death. The final death tally doesn't necessarily represent the actual damage on the field, and I've noticed that many people tend to assume that all surviving enemy mechs are in relatively good condition. I've finished my fair share of matches with 6+ mechs alive on my side, all critically damaged, and seen someone claim a "stomp" occured (when really it was a good fight).

I don't mean to suggest that stomps never happen - they surely do. I just don't think they happen as often as we believe they do.

[Mechteric]

I don't think 12/4 or 12/5 matches always constitutes as stomps. Unless the only 4 or 5 mechs that died were locusts and stock Atlases or something.

[Craig Steele]

Hmm, sorry Capper I may not have explained well.

In the first 4 "stomps" the majority of winning side deaths came late in the game when the result had already been decided. In the last one we got a few kills early on (but lost more) when their Assaults were not present and then the rout occurred.

The game that I felt was close we had a tonnage disadvantage yes but the game traded evenly until it got to 8 / 7. Their last four mech were 2 Atlas, a Stalker and a Pult. We had a Jag, a Jenner, my Wolverine, a SHD and something else (sorry I remember the ones I was fighting next to) The jag and the SHD were smoking hard so I knew they were not healthy.

Game wasn't good just because kills were close. Both teams used cover and moved, I suspect we used most of the map by games end. They had a good Spider pilot that was making a pain of himself but our Jenner pilot was on to him and the rest of our team didn't take the bait and stayed in the fire fight. There was a cicada sniping that we had to pin with fire until a SHD could flank him and we smashed him up as he made a run for it.

Lots of little things in the match that engaged me, I felt like I was doing something towards a result. If we had of lost, I still would have thought that was an entertaining game.

The stomps are variations of "walk to make contact with enemy, blink, casualty count now 4-0, look at mini map, casualty count now 6-1, round up stragglers / lights"

Hope that makes it clearer.

[Mudhutwarrior]

I wont call it a stomp unless its less than two kills and even then to me a stomp is 12/0 most often. As far as bad matches go I altered my sked so I could play more late at night EST. Best chances of decent matches I found. The west coast guys seem to get the better deal there.

[Auzen]

CapperDeluxe, on 07 January 2014 - 06:47 AM, said:

I don't think 12/4 or 12/5 matches always constitutes as stomps.

This. I've been keeping a log of my games since mid-November, and as of right now, the average result (win or lose) is 11.77 kills on the winning side, and 5.00 on the losing side. So a 12-5 game is typical. As such, I don't get bent out of shape for a 12-4 or even 12-3, and I don't consider it a "stomp" until it's 12-2 or worse. And those have been fairly rare.

Recently I've had several 12-11 matches that were suspenseful right to the end and could've gone either way. Those were a lot of fun, but expecting that to happen frequently just isn't realistic, since losing a couple of Mechs can have a big impact on the team's chances, reducing the overall available firepower on the team, and it can easily snowball from there.

My advice would be... take the losses in stride and don't feel bad if you just lost in a 2-12. It happens, and it's going to continue to happen, because I don't think any matchmaker, real or imagined, can be expected to give us 12-9 typical matches. Sometimes you just can't recover from losing a couple of key Mechs.

Also, in case it matters, I solo PUG and the above reflects my experiences with that. Sometimes folks work together, sometimes they don't.

EDIT: A note about tonnage... I was in a match recently where we had a Highlander and Catapult, and the rest were Mediums and Lights. Enemy team had a tonnage advantage of something like 145 tons, and I saw some chatter about "Oh, nice job, matchmaker!" - and then we won it 12-5. So pure tonnage doesn't always mean much.

Good gaming!

Edited by Auzen, 07 January 2014 - 07:36 AM.

[Mudhutwarrior]

Auzen, on 07 January 2014 - 07:33 AM, said:

This. I've been keeping a log of my games since mid-November, and as of right now, the average result (win or lose) is 11.77 kills on the winning side, and 5.00 on the losing side. So a 12-5 game is typical. As such, I don't get bent out of shape for a 12-4 or even 12-3, and I don't consider it a "stomp" until it's 12-2 or worse. And those have been fairly rare.

Recently I've had several 12-11 matches that were suspenseful right to the end and could've gone either way. Those were a lot of fun, but expecting that to happen frequently just isn't realistic, since losing a couple of Mechs can have a big impact on the team's chances, reducing the overall available firepower on the team, and it can easily snowball from there.

My advice would be... take the losses in stride and don't feel bad if you just lost in a 2-12. It happens, and it's going to continue to happen, because I don't think any matchmaker, real or imagined, can be expected to give us 12-9 typical matches. Sometimes you just can't recover from losing a couple of key Mechs.

Also, in case it matters, I solo PUG and the above reflects my experiences with that. Sometimes folks work together, sometimes they don't.

EDIT: A note about tonnage... I was in a match recently where we had a Highlander and Catapult, and the rest were Mediums and Lights. Enemy team had a tonnage advantage of something like 145 tons, and I saw some chatter about "Oh, nice job, matchmaker!" - and then we won it 12-5. So pure tonnage doesn't always mean much.

Good gaming!

I think from my perspective 12/0 happen a lot. Easily 2 out of 10 matches on a good day. Someday's depending on the time it gets much worse. I have been stomped 12/0 many times in a row. Just a matter of where your sitting. Maybe elo has something to do with that but my alternate account showed the same names as my 8000 match one after noob time passed.

[Auzen]

Without elaborating too much, I looked at the stats I'd been keeping:

Since Nov 15th:
293 matches, 165W, 128L; out of those, 5 had a result of 12-0 (I lost four of them).

Here are my full results:

Losing team score / number of occurrences
0 / 5
1 / 13
2 / 38
3 / 35
4 / 41
5 / 53
6 / 25
7 / 23
8 / 29
9 / 19
10 / 5
11 / 7

So in my recent experience (since starting to record my results), I've had just as many 12-10 matches as I've had 12-0, and actually a couple more 12-11 results. I am encouraged by that, and I'm glad I took the time to look up the numbers.

Anyway, based on my own definition of a "stomp" (12-2 or worse), I've been involved in 56 of them, which is 19.1% of my matches. Hmm... looking closer, it turns out that the average result in my numbers (12-5) actually winds up being that or worse in 63.1% of matches. Half or more of the losing team actually survives only about 1/3 of the time. And 12-9 or better only happens 10.6% of the time.

I understand these numbers only apply to me, and can't necessarily be a broad representation of everyone's experience. But I think it can clearly illustrate that people shouldn't be discouraged that they aren't seeing most of their matches being close, nail-biting affairs that come down to a chess match between 2-3 remaining Mechs.

Another quick bit of anecdotal info: I played five on Christmas. 12-2 win, then a 10-10 tie (heh), a 12-11 win, a 11-12 loss, and then a 12-6 win. Those three really close matches were pretty cool, and I decided to call it quits for the day with a good feeling rather than continuing and chancing a 1-12 stomp putting me in a mood.

Good gaming!

[Joseph Mallan]

Mudhutwarrior, on 07 January 2014 - 07:32 AM, said:

I wont call it a stomp unless its less than two kills and even then to me a stomp is 12/0 most often. As far as bad matches go I altered my sked so I could play more late at night EST. Best chances of decent matches I found. The west coast guys seem to get the better deal there.

That is an interesting point.
Serious!

[Sandpit]

Mud, I congratulate you! You're actually having some good constructive posts lately!

I agree with a lot that's already been said. a 10-2 "stomp" and things of that nature aren't always as clear cut as they seem. They might have 10 mechs left on the map but they aren't strolling around talking about kicking my team's ***. They might be gimping around thankful the last few mechs that were alive on my team didn't sneeze on them though.

Don't get me wrong, I've been on both ends of stomps and they're boring. Both as the winner and the loser. But sometimes it's simply because the other team was just really good.

I have found that when the MM goes joe derp rambo on me with weight matching that the other team USUALLY isn't very good skill wise. Hence my earlier example of the 4 Atlases with me as the only Assault in a Stalker. If they had been on par with my team's skill level (ELO wise I suppose) not only would we not have won, they'd have stomped us before we got past our base.

[wanderer]

Craig Steele, on 07 January 2014 - 07:18 AM, said:

The stomps are variations of "walk to make contact with enemy, blink, casualty count now 4-0, look at mini map, casualty count now 6-1, round up stragglers / lights"

Hope that makes it clearer.

Basically, a stomp is what happens when a disorganized mass impacts with an even slightly better organized (even in partial degrees) group. This isn't a sign of dysfunctional game, it's a sign of dysfunctional teamwork.

Teamwork is -always- OP. I've watched lances of stock 'Mechs mow down twice their weight/number in opponents because the other guys were a mindless blob and the stock lance simply called targets and focus fired. Focus leads to numerical advantage. Numerical advantage leads to victory.

[Thimble]

wanderer, on 07 January 2014 - 01:43 PM, said:

Basically, a stomp is what happens when a disorganized mass impacts with an even slightly better organized (even in partial degrees) group. This isn't a sign of dysfunctional game, it's a sign of dysfunctional teamwork.

Teamwork is -always- OP. I've watched lances of stock 'Mechs mow down twice their weight/number in opponents because the other guys were a mindless blob and the stock lance simply called targets and focus fired. Focus leads to numerical advantage. Numerical advantage leads to victory.

I think the OP's point is that matchmaking is severely dysfunctional.

That is, players that communicate well and thus kill quickly SHOULD NOT be placed into games against trial mech noobs (grouped or not) that stand still and spray medium laser fire at a target 600M out. I've been seeing a lot games lately where Alpha/Beta lances (most often grouped players) were of COMPLETELY different skill levels on the two teams.

This points to a matchmaker failure, not simply a lack of teamwork.

FYI I PUG, so I pay a lot of attention to who is grouped on each side.

[SuomiWarder]

Bring greater force against the enemy has always been a prime "winning" strategy. If one side loses two or three Rambo pilots and is spread out a little, the other side can bring more guns on target and flank to reduce cover. The snowball goes down hill fairly fast.

It is fairly rare that I see a game with one side down 3 mechs to the other come back and win. Oh - it does happen. Usually when a com ussing group of bigger Mechs is against a mob of solos and stick to gether while duos some filtering in to slaughter. But it seems fairly rare to me.

[Joseph Mallan]

Thimble, on 08 January 2014 - 05:14 PM, said:

I think the OP's point is that matchmaking is severely dysfunctional.

That is, players that communicate well and thus kill quickly SHOULD NOT be placed into games against trial mech noobs (grouped or not) that stand still and spray medium laser fire at a target 600M out. I've been seeing a lot games lately where Alpha/Beta lances (most often grouped players) were of COMPLETELY different skill levels on the two teams.

This points to a matchmaker failure, not simply a lack of teamwork.

FYI I PUG, so I pay a lot of attention to who is grouped on each side.

I don't think this is true... meaning, If you wanna play a combat game you should be ready to face combat! I came in with that mind set and I was not disappointed. I got handed to for weeks! Learned something new every day. Then I learned how to start killing the enemy and helping the team win a match or three.

Match Maker is "random", and for me... random is less boring than an even fight (which is a misnomer).

Edited by Joseph Mallan, 09 January 2014 - 05:59 AM.

[Mudhutwarrior]

Joseph Mallan, on 09 January 2014 - 05:59 AM, said:

I don't think this is true... meaning, If you wanna play a combat game you should be ready to face combat! I came in with that mind set and I was not disappointed. I got handed to for weeks! Learned something new every day. Then I learned how to start killing the enemy and helping the team win a match or three.

Match Maker is "random", and for me... random is less boring than an even fight (which is a misnomer).

15000 posts of the same paragraphs over and over.

People know its broken. I just had two matches. 1st all noobs and I mean all noobs. I took more fire from my own team in the back and the kill shot was a gauss from my teams atlas.
Next match was full premade drops with very experienced players and a full hard work battle. We lost but not by much.

The reason is not enough players because its been getting steadily worse. Yesterday half my matches were short fill usually one or two but one had four slots empty. An entire lance. You know how that match went.

So please Joe. We know your in love. Just stop with the same tired excuses over and over and over. 15500 worth,

[Joseph Mallan]

Mudhutwarrior, on 09 January 2014 - 06:20 AM, said:

15000 posts of the same paragraphs over and over.

People know its broken. I just had two matches. 1st all noobs and I mean all noobs. I took more fire from my own team in the back and the kill shot was a gauss from my teams atlas.
Next match was full premade drops with very experienced players and a full hard work battle. We lost but not by much.

The reason is not enough players because its been getting steadily worse. Yesterday half my matches were short fill usually one or two but one had four slots empty. An entire lance. You know how that match went.

So please Joe. We know your in love. Just stop with the same tired excuses over and over and over. 15500 worth,

1,043 of the same posts as well Mud. So turn in your pot if you wish me to relinquish my kettle!

[wanderer]

Thimble, on 08 January 2014 - 05:14 PM, said:

I think the OP's point is that matchmaking is severely dysfunctional.

That is, players that communicate well and thus kill quickly SHOULD NOT be placed into games against trial mech noobs (grouped or not) that stand still and spray medium laser fire at a target 600M out. I've been seeing a lot games lately where Alpha/Beta lances (most often grouped players) were of COMPLETELY different skill levels on the two teams.

When matchmaker can take into account 3rd party VOIP, you'd have better odds. I mean, I play PUG, and I play on Teamspeak.

The same players with or without voice comm can be like night and day. I've seen them miss a team-mate being ganked by an entire lance 300m behind them without comms, while they wiped up a similar lot without losing a 'Mech because a little chatter gets everyone on the same page.

Even a scrub player with comm access improves greatly over a silent one who flails quietly and dies.