34 replies to this topic

[whoosher]

nothing more satisfying than the matchmaker RNGame

Edited by whoosher, 15 February 2021 - 11:13 AM.

[Kazgruta]

Easy to blame a crushing defeat on the matchmaking rather then players.

[VonBruinwald]

Stat aggregates from Jarls:

Team 1 - 3 players without a rank

Rank 16782
Percentile 49.33%
Games Played 8660
W/L Ratio 1.02
Survive Rate 30%
K/D Ratio 0.89
Average Score 225
Adjusted Score 195.5

Team 2 - 5 players without a rank

Rank 16731
Percentile 49.33%
Games Played 3841
W/L Ratio 0.93
Survive Rate 32%
K/D Ratio 0.86
Average Score 197
Adjusted Score 180.7

Edited by VonBruinwald, 15 February 2021 - 12:22 PM.

[Wolfos31]

It could be matchmaker, but I notice that only one of the losing team's players broke 300 damage (barely) and the winning team doesn't have any real standouts with damage reasonably spread across the team.

It looks to me like this was a heavy NASCAR stomp. Probably no good communication on the losing team and folks just kept trying to run from their death instead of turning and dishing out damage.

[Heavy Money]

The primary cause for losses of all sorts, but especially stomps, is people not firing their weapons enough.

No, really. Next time you die early, flip through everyone. You'll see tons of people that, upon sighting an enemy mech, immediately reposition or take cover or something and don't even attempt to shoot. Or who are taking fire and backing away and don't shoot back. Or who just constantly leapfrog behind their team mates to avoid taking damage.

[VonBruinwald]

Wolfos31, on 15 February 2021 - 12:17 PM, said:

It looks to me like this was a heavy NASCAR stomp. Probably no good communication on the losing team and folks just kept trying to run from their death instead of turning and dishing out damage.

This. A little coordination goes a long way.

If you look at matches played between the teams, the winning team has twice as many games under their belt, it's a safe bet they had a better idea of what to do. The losing team had an entire lance of cadets they would have needed to carry.

[TheCaptainJZ]

It happens, and the nature of this game is that once one team gains the momentum, say, after 3 kills, the rest often fall in place like dominos. Those Jarl's stats are interesting. Not overly unbalanced except by number of matches, but the MM doesn't look at number of matches (and that's not calculated in Tier anymore). Clearly, there was more experience on the winning side so they were probably communicating in match. If not, they at least had a better feel for how to play this map and mode.

[ThreeStooges]

GG you died well. Try again.

[Wolfos31]

Heavy Money, on 15 February 2021 - 12:33 PM, said:

The primary cause for losses of all sorts, but especially stomps, is people not firing their weapons enough.

No, really. Next time you die early, flip through everyone. You'll see tons of people that, upon sighting an enemy mech, immediately reposition or take cover or something and don't even attempt to shoot. Or who are taking fire and backing away and don't shoot back. Or who just constantly leapfrog behind their team mates to avoid taking damage.

This is a lesson I feel like I've recently learned. See an enemy? Within range? Fire!

"But I can't isolate a component at this range." "But they're not facing me, I'll hit an arm." "But I might miss."

Are enemies right in your face? No? Then it doesn't matter. Take the shot. You'll get better at hitting with experience but especially for the opening shots of a match before you close range and have reliable targeting data damage on target is better than a shot "saved for a better moment".

To the OP: I've had losses like that too. Sucks, but the good news is that you can queue up again and try again. My only goal in every match is to give more damage than I take and have fun. If I see my team is going to lose (3-4 mechs down) my goal stays the same, maybe with a bit of an added drive to drop at least one mech before I'm wasted.

[SmokeGuar]

New account or alt?

PSR reset -> T5 match with varying levels of experience, trial mech, founder, cadets... nothing MM can really do to balance teams.
Basically same happened with PSR reset to T3, but that was even worse. Cadets ended with T1 matches and became cannon fodder.

[YueFei]

Hell, sometimes the final score doesn't really show how close a match was.

Ever been in a match where you're on the winning team, you're the only guy that's dead on your team, and as your team is up 11-1 and hunting the last guy, you jump around to each teammate's PoV... and you can see that most of your teammates have half their mech missing or are one-touch from dying?

From the losing team's PoV it might feel like a stomp, but in reality they might've been just a handful of shots away from turning it into a nail-biter. Instead of being able to secure kills, near-dead mechs on your team survived and were able to continue pumping out damage that tipped the scales.

[PocketYoda]

whoosher, on 15 February 2021 - 11:13 AM, said:

nothing more satisfying than the matchmaker RNGame

We had three of those last night in a row.. Seriously in a row.. the matchmaker is really off.. and how is it allowed to start a match with one team having no lights at all while the other team had at least 4... We got to the pee stain with an arctic wolf with 17 secs remaining on polar.. as that was the fastest mech in our team O.O

That is not accurate drops.. something is broken.

VonBruinwald, on 15 February 2021 - 12:17 PM, said:

Stat aggregates from Jarls:

Team 1 - 3 players without a rank

Rank 16782
Percentile 49.33%
Games Played 8660
W/L Ratio 1.02
Survive Rate 30%
K/D Ratio 0.89
Average Score 225
Adjusted Score 195.5

Team 2 - 5 players without a rank

Rank 16731
Percentile 49.33%
Games Played 3841
W/L Ratio 0.93
Survive Rate 32%
K/D Ratio 0.86
Average Score 197
Adjusted Score 180.7

Look at those games played, also w/l differences.. No wonder it was a stomp.. Its definitely the matchmaker.

Edited by Samial, 15 February 2021 - 04:52 PM.

[RickySpanish]

Alt account dies in a t5 game with not even 100 damage to their name, then comes to the forum to complain. Ouch.

[Saved By The Bell]

Sometimes it happens.

Yesterday, all team except me just run to enemy base, full of defending mecs. "Dont go, you ll die", I cried, firing ERPPC from 1-2 km. Team all died quickly, then last was me. I did more than 700 damage with my ERPPC. They did 0-300 damage.

Next time game crushed. When I came back, they were severely beaten, I died quickly with 300 damage. We all did 0-300 damage.

But I love this game. So challenging. Need to use brain.

You are alone against all enemy team, comrades already dead... So cool, so much endorphine.

Edited by Saved By The Bell, 16 February 2021 - 05:50 AM.

[TercieI]

All I see when I look at that is the winning team should be saying "nice damage spread." One team probably talked and the other didn't. MM can't solve that.

[The pessimistic optimist]

whoosher, on 15 February 2021 - 11:13 AM, said:

nothing more satisfying than the matchmaker RNGame

Yup matchmaker can be a cruel mistress

[Nightbird]

W/L difference should be taken to the 12th power and compared

1.02^12=1.26
0.92^12=0.42

1.26 vs 0.42, the first team should win more than 90% of the time, GJ match maker

[Wolfos31]

Nightbird, on 16 February 2021 - 08:25 AM, said:

W/L difference should be taken to the 12th power and compared

1.02^12=1.26
0.92^12=0.42

1.26 vs 0.42, the first team should win more than 90% of the time, GJ match maker

Why should it be taken to the 12th power? The W/L ratio was an average. And some players weren't ranked in Jarl's which is going to throw off the average anyway.

[Nightbird]

Wolfos31, on 16 February 2021 - 08:28 AM, said:

Why should it be taken to the 12th power? The W/L ratio was an average. And some players weren't ranked in Jarl's which is going to throw off the average anyway.

Can't do anything about data that is not there. Multiplication is used because it more visibly represents the difference in team skill. Addition and average doesn't work because the average person doesn't understand standard deviation and that 0.1 is a huge gap.

[Grus]

Hold up... this used to be reported as name and shame. Are they allowing this now?