99 replies to this topic

[Jag Hiroshi]

EDIT: Logged in today - won most of my matches. I suspect the feeling that I'm being dunked on is just subjective / in the moment. As Curccu states - the stats seem roughly in the right place. It's weird how it seems sometimes. Must work on that KD!

- - -

ORIGINAL MESSAGE:

I thought I'd post this as a follow up to a previous thread I made on this matter a couple of months ago.

I'm not claiming any skill in this game, but I would expect matchmaking to help spread the load a little. I should win some games and lose others. That doesn't mean I expect a 50/50 split, but I should be able to reflect on an evening of combat and come away with enough wins or good matches to feel like it's been worth playing.

However...

I log on in an evening and all I seem to get are a steady stream of losses, peppered with the occasional win. Before I started playing this evening, I thought I'd tally winds losses (and ties) just to ensure that I'm not experiencing some kind of perception bias. Here they are:

L L W L L L L T W L L

That's:

• Wins (W): 20%
  
• Losses (L): 70%
  
• Ties (T): 10%

20% win rate. 20% !!!






This seems quite a typical experience for me - and it feels like MM is pitching well organised teams against solo fodder (I take it that <ENEMY_NAME (A)> alpha tag at the end denotes someone in a squad?

I'd be keen to hear the experience of other solo players? I'll be honest, I love the game - well over 1k hours in it, but recently this kind of experience is all that I seem to get and I think I've hit my threshold.
</p>

Edited by Jag Hiroshi, 19 December 2023 - 01:16 PM.

[Curccu]

https://leaderboard....h?u=Jag+Hiroshi shows you are doing bit less than 50/50 in your history.

https://www.mwomercs...ser=Jag+Hiroshi
This month so far 84 wins, 87 losses.

Doesn't look like 20%

[Tarl Cabot]

The <xxxx> does indicate someone is unit member but it does not indicate they are in a high level unit, or if the individual is a high Tier member of said unit. And you will have streaks of wins or losses. What we do not see are thread from players who have are on a winning streak , only when it is a losing streak, just saying... One always need to look at their overall numbers, then per season (month).

Understand though, in each game you are really the only constant. Will you discuss what tier you are in? It would look like Tier 3? Close to tier 4 or tier 2? ie, ya get where you may bounce back/forth ?

And when you say evenings, whose evening? Time zone/location? At the time of my post it is 4:39pm CST in Texas, or 10:39pm UTC.

[Jag Hiroshi]

Tarl Cabot, on 18 December 2023 - 02:38 PM, said:

The <xxxx> does indicate someone is unit member but it does not indicate they are in a high level unit, or if the individual is a high Tier member of said unit. And you will have streaks of wins or losses. What we do not see are thread from players who have are on a winning streak , only when it is a losing streak, just saying... One always need to look at their overall numbers, then per season (month).

Understand though, in each game you are really the only constant. Will you discuss what tier you are in? It would look like Tier 3? Close to tier 4 or tier 2? ie, ya get where you may bounce back/forth ?

And when you say evenings, whose evening? Time zone/location? At the time of my post it is 4:39pm CST in Texas, or 10:39pm UTC.

I was mid tier 3 but the second half of this year it feels like I'm constantly getting match streaks as above - so even if I scrape 500 damage and a kill, I can still go down. Now I'm bouncing around the bottom of tier 3. This is most evenings EU (and I tend to play EU servers only because of hit-reg issues).

[Jag Hiroshi]

Curccu, on 18 December 2023 - 02:04 PM, said:

https://leaderboard....h?u=Jag+Hiroshi shows you are doing bit less than 50/50 in your history.

https://www.mwomercs...ser=Jag+Hiroshi
This month so far 84 wins, 87 losses.

Doesn't look like 20%

I'm guessing the stats say that it's selection bias then. I mean, it certainly doesn't feel like it but hard to argue with the figures.

[Mobster]

My irk about this is even at 50/50 win or loss, the PSR goes down because losses appear to drain more than a win would gain.

I've watched over time, and even though I don't "suck" (I can do 2 or 3 kills in most games, I have a fair grasp of mechanics and tactics and tend to a 0.9x WL ratio so I'm not godawful) but even with my 'nearly' balanced WL I loose more PSR than I gain over time.

[RickySpanish]

If your win/loss is less than 1, then yes your PSR should fall as you are being matched above your skill.

[crazytimes]

Mobster, on 18 December 2023 - 04:26 PM, said:

My irk about this is even at 50/50 win or loss, the PSR goes down because losses appear to drain more than a win would gain.

I've watched over time, and even though I don't &quot;suck&quot; (I can do 2 or 3 kills in most games, I have a fair grasp of mechanics and tactics and tend to a 0.9x WL ratio so I'm not godawful) but even with my 'nearly' balanced WL I loose more PSR than I gain over time.

Losses are capped at -24 due to the formula, and to get that you need a zero MS. Wins are notionally uncapped- I think match score stops at 999, which gives a PSR of ~+50. You can offset 3 or 4 losses where you were median scoring with one great game.

You are either tiered above your actual skill level, or are suffering selection bias as well. 0.9 winrate isn't nearly balanced too- that's a solid standard deviation or more below average. https://leaderboard....ats#WLRPopchart

This isn't criticism of you- it's just noting you aren't being punished by the PSR system for a "nearly balanced" win rate

[Tarl Cabot]

Mobster, on 18 December 2023 - 04:26 PM, said:

My irk about this is even at 50/50 win or loss, the PSR goes down because losses appear to drain more than a win would gain.

I've watched over time, and even though I don't "suck" (I can do 2 or 3 kills in most games, I have a fair grasp of mechanics and tactics and tend to a 0.9x WL ratio so I'm not godawful) but even with my 'nearly' balanced WL I loose more PSR than I gain over time.

Being almost equal on MS points, exception of slightly higher PTs for a win vs lose, the piece is a slight percentage modifier for the winning team. And damage contributes approx 46% of its points into Matchscore.

https://mwomercs.com...ost__p__6523065

And damage converts approx 46% of its points (less than half) to the Matchscore. No including any team damage, 300dmg = 138 MS. 400dmg = 184 MS So your 334 dmg = 153 MS, slightly less with the team damage. Damage is the larger contributor to the Matchscore, but not the only trigger. And kills is just that kills. Yes, you removed the mech/player from the field by dealing the last point of damage to destroy that mech. If you do the most damage then you also earn KMDD (did the most damage before mech was killed by someone else). Then if you receive the Solo Kill, that is for doing the KMDD and getting the Kill Shot.

Sticky, Formula and examples.
https://mwomercs.com...ity-version-10/

Criteria what triggers are used for generate the matchscore. The only item we are aware that has been changed, or I should say, are points generated by AMS destroying incoming missiles.
https://mwomercs.com...ost__p__6337261

[Tarteso]

Solo player here.

I don't care if W/L. Problem is that 70-80% matches are stompy or one-sided as hell.
MM is trash.

[torsie]

These are all my games I played this morning: https://imgur.com/a/ro4wXwr

Skirmish, Assault, Domination, Conquest
W is win, L is loss, + is green up arrow, - is red down arrow and = is yellow =
And some other notes, if something different happened and which map it was, for example when we had win without killing everything.

First 10 games 6 wins, 4 loses. 2 yellow and 1 red, 7 green and only 2 games were super fast.

Second 10 games 3 wins, 7 loses. 1 yellow, 5 red and 4 green and 1 super fast game.

I think that makes it 9 wins and 11 loses, which is actually pretty shocking to me .
I was expecting much worse.

Edited by torsie, 19 December 2023 - 05:26 AM.

[Horseman]

crazytimes, on 18 December 2023 - 07:06 PM, said:

I think match score stops at 999

It doesn't, but breaking 1K score requires somewhere around 2K damage which under most realistic scenarios simply does not happen.

[Saved By The Bell]

In first minute after start you must determine, what kind of team you are.

If they fault the test you must act alone and never wait-call any assistance.

Edited by Saved By The Bell, 19 December 2023 - 06:17 AM.

[crazytimes]

Horseman, on 19 December 2023 - 05:35 AM, said:

It doesn't, but breaking 1K score requires somewhere around 2K damage which under most realistic scenarios simply does not happen.

Wasn't sure, I got a 999MS with ~1700 damage this year and tried to work out if I was unlucky or it was actually capped. I couldn't find any QP >999 screenshots from after 2019, which was about when they did some shuffling in match results display.. but maybe I was just in denial for missing 1000 by one.

[pbiggz]

Tarteso, on 18 December 2023 - 11:40 PM, said:

Solo player here.

I don't care if W/L. Problem is that 70-80% matches are stompy or one-sided as hell.
MM is trash.

I have said this before, and will say it again now. I don't think most stomps are actually stomps.

If the post-match screen showed each team's total armor, and showed how much of each team's total armor was left, I think it would indicate how close a match was much better. Kills are the criteria for a win, but they aren't a comprehensive indicator of performance. Both teams could do very close to even damage, but one team might spread just a little more damage to arms or legs, leading to that same damage not killing a target, while the other team might land their shots on CT a few more times, and get the kills.

As I have also said before, this game was evidently made with respawns in mind, and we've never had them outside of event queue and faction play, simply because we are still playing the beta game modes, each of which is basically skirmish, or a different flavour of skirmish, that PGI never bothered to comprehensively update. Stomps would go away if respawns existed and game modes were updated to reflect that. It would take work, there are complexities, nobody wants 40 minute spawn farms, that would be disastrously bad, but a lot of peoples perception of stomps, and bad quality matches is the fault of PGI's dated game modes and incomplete information being presented to players, so people can't actually evaluate their own performance, relative to their team's performance. It is very easy to blame the matchmaker, but I don't believe that's truly the source of your woes.

As for OP, he was just having a bad day. I, a mediocre player, have many bad days. Try a different game for a bit, or a different build. Freshen things up, take a walk, do something other than stay in the game losing over and over because its really easy to get into a nasty funk and start hating MWO when you keep slamming your head into loss after loss.

[Sjorpha]

You can't take 10 games and claim anything, even just flipping a coin for a while will generate streaks of heads or tail.

[Sjorpha]

Mobster, on 18 December 2023 - 04:26 PM, said:

I can do 2 or 3 kills in most games

Your k/d ratio is 0.67, so you are averaging well below 1 kill per game. Your stats suggest you should lose psr over time since you are clearly being overwatch.

[Ignatius Audene]

Mobster, on 18 December 2023 - 04:26 PM, said:

My irk about this is even at 50/50 win or loss, the PSR goes down because losses appear to drain more than a win would gain.

I've watched over time, and even though I don't &quot;suck&quot; (I can do 2 or 3 kills in most games, I have a fair grasp of mechanics and tactics and tend to a 0.9x WL ratio so I'm not godawful) but even with my 'nearly' balanced WL I loose more PSR than I gain over time.

Psr is a 0 zum calculation around all players (with the small problem of maxed tier 1 and empty tier bar). Some rise, some fall, some stay depending on the performance of all other players.

Your W/l is under 1.0. your K/d is lower and most important your AVG match score is quite low. Shoot more and should be fine or drop in psr and hopefully get better match ups (tier is no achievement!)

[Horseman]

the check engine light, on 20 December 2023 - 05:28 AM, said:

The maxed T1 bar is a serious problem and shouldn't be discounted. The presence of that issue makes T2 suck.

The problem is, in my opinion, overscoring raw damage numbers above the precision of dealing said damage. There are players like myself who end up with the bar maxed not because they're *good* as such but because they spam a lot of bullets.

[An6ryMan69]

QP matches are fairly stupid right now because the generous events are bringing in all manner of players trying to farm rewards and not caring about good team play or winning matches. And the worse the quality of matches are, the faster some players just want to farm rewards and get out as general enjoyment seems pretty low, driving match quality down even more.

It's kind of a vicious cycle.

Then, for players who always drop solo, you also have the added but significant factor of organized group of reward farmers popping in (often in "paid real money" Legendary or other similar mechs) and making life even harder for solo players. And, when things are generally chaotic in QP, the impacts of an organized group with comms seems amplified on the battlefield even more than usual.

While it might be a good time to build up an MWO account, but it may very well be coming at the expense of a solo player feeling like they're just unpleasantly grinding out matches.

Edited by An6ryMan69, 22 December 2023 - 10:28 AM.