51 replies to this topic

[Vasili Kerensky]

I was here when the game began, had played the Cadet matches. I returned about one month ago and have played a lot of games. I'm Tier 5. I'm losing 80% or more of the QM games. My brother has the same experience over the same timeframe and same tier.

Obviously, if the matchmaking was building a game with T5 people on both sides, we'd have a 50% ratio of defeat-victory. Why isn't this so? Also, I've noticed players on the other side with the 'Legendary Founder' title not a few times. Why are there players who have been playing this game since the beginning in a T5 QM?

How exactly does the matchmaking work at Tier 5 and I have to see defeat after defeat? If your answer is the usual get better, it's what your team deserves, etc. crap, don't bother to answer.

Edited by Warsong, 07 August 2016 - 12:00 PM.

[Roughneck45]

It groups players within 3 tiers of each other. So tier 5, 4, and 3 can be grouped in the same game.

Also reports of the MM widening that gap when it cant make a balanced match quickly enough.

If you are grouping with your brother it will put you in the group que which is only groups and generally has more experienced players.

Edited by Roughneck45, 08 August 2016 - 11:10 AM.

[Hunka Junk]

I'm not going to ask you to screenshot your .20 W/L ratio, but that seems incredibly low.

Are you sure it's not your K/D ratio?

In any event, as a fairly new person I felt more comfortable running lights because I had a certain sense that by playing a heavy or assault when I didn't know what I was doing, I was depriving my team of some serious tonnage. Maybe that's just me, though.

Stock builds also killed my performance. One thing is that new mechs seem to come with 5-6 weapons groups, which is a real pain if you're not even too sure about the different kinds of weapons. Even now, I rarely run a setup with ore than 3 groups. Then, there's the rude awakening of all your new IS mechs coming with lawn mower engines where the 440 big block is supposed to go...

Then there was me. I started by dying early almost always. Then I moved on to just seeing how long I could stay alive. THEN I was comfortable enough to start trying to help win. And that really started with me staying at impossibly long ranges hoping that I'd hit something but not really. And then once I had range figured out, I picked up on chain firing for heat management. Once I was comfy chainfiring and not constantly overheating, I realized how much more damage I could do by not chainfiring and instead riding the heat limit. All of that happened while I was thought certain weapons were crap and other weapons better, only to then find the other weapons were actually crap and the crappy ones good, only to discover I was right the first time around but wasn't using them right.

I guess what I'm saying is that in many cases I got poor results because I was trying to make the game play the way I wanted it to work instead of changing the way I was playing. I still do that.

But really, there are so many aspects of my performance and understanding of the game that need improvement, that I am far far off blaming the tier system for my frustration.

It really comes down to that. If I die, I try to figure out why I died. The two most common causes are probably impatience and poor situational awareness. Likewise, if we lose, I try to figure out why we lost. The two most common cause are probably people venturing too far out at the beginning by themselves and people just sitting in the back avoiding contact. The third is ppl afk-ing or disconnects. That's 80% of my losses right there before I'd even consider complaining that people on the opposing team are uber-experienced vets preying on noobs in tier 5 for the fun of it. Not that that doesn't happen. I just don't see it as terribly common compared to the theater of bad play that I am sometimes contributing to myself.

[Nightbird]

As a new player, there are many things to check to make sure you're not playing at a handicap. Are you playing at a good frame rate? Is your ping low? (Push F9 during a match to see this in the top left of your screen) Is your mouse sensitivity set to allow you to turn quickly but still aim accurately?

If all of this is ok, then we get to: which mechs and loadouts are you using? Have you watched any video tutorials on playing this game?

[mad kat]

Ive been playing for years but ever since the PSR was introduced matches have been crap.
I even logged some matches and wasn't surprised just how bad the ratio is.
http://mwomercs.com/...ost__p__5329677
i'd like to say it will get better. But unless PGI throw that PSR system in the bin and restore the old match maker it will get worse.

[Spheroid]

What mech are you using and what is the build?

[TooDumbToQuit]

RN45 is correct and that might be a big part of your problem. I'm at Tier 5.4. But when I group with another player who is a simple 5 Tier, we are put with other "groups" and we are facing much better players.

Your odds are low. I really suck and my team can carry me for some wins....

[Vasili Kerensky]

I appreciate all the responses. You focused on my performance which is understandable but beside the point. In a balanced matchmaking system, with a large enough sample - I have over 100 matches - probability says that I'll win half and lose half, no matter if I suck or not. That's assuming that the matchmaking system is doing its job. So, you didn't really address the question since it can't be my fault. Also, I've grouped with my brother for about 10 of those matches, so it's not that either. Plus, he's been having the same bad experience of too many defeats.

Mad Kat what's this PSR and when was it implemented?

Edited by Warsong, 07 August 2016 - 03:23 PM.

[S 0 L E N Y A]

First things first

Pilot Skill Rating is not an accurate description of skill for most.
The bigger issue is the player population is rather low, so it would be practically impossible to really divide the ques into truly accurate evenly matched games, especially group que.
Lastly -and the most important part- how much are you a contributing factor to those losses?

P.s. 100 matches is nothing. Keep at it

Edited by Boogie138, 07 August 2016 - 03:42 PM.

[The Basilisk]

Warsong, on 07 August 2016 - 03:22 PM, said:

I appreciate all the responses. You focused on my performance which is understandable but beside the point. In a balanced matchmaking system, with a large enough sample - I have over 100 matches - probability says that I'll win half and lose half, no matter if I suck or not. That's assuming that the matchmaking system is doing its job. So, you didn't really address the question since it can't be my fault. Also, I've grouped with my brother for about 10 of those matches, so it's not that either. Plus, he's been having the same bad experience of too many defeats.

Mad Kat what's this PSR and when was it implemented?

Sorry friend, but your question has been answered by at least two ppl.
Your issue is mainly caused because you are grouping with your brother.
There is a QP single player queue that will funktion pretty much like you would expect it save for some occasions when the QP player count is extremely low (FW events, holydays like right now, and timezones away from us and europe prime times)
The QP group queue has much looser matchmaker restriktions due to the lower player count and higher possible variations a team can force upon the MM.
Therefore you will naturely be forced to play against much more advanced players in QP group queue.
Also Founders tags don't say much.
I know some of them and a good portion are in T4/3.
Having a catastrophic streak of ten or more matches isn't uncommon in this game and the oposite isn't a rarity too.

PSR = Player skill rating
Its a counter fueled by the games you play. Every QP game you play either raises, does not effect or lowers your PSR. FW does not influence your PSR and so you may indeed meet high end players that have a low psr due to the fact they seldomly play qp.
And before you ask: PSR

Edited by The Basilisk, 07 August 2016 - 04:01 PM.

[Lukoi Banacek]

Group can absolutely be more unforgiving than solo queue, especially for someone new to the game. You group up, you face anyone basically. Yes, there is a MM, but it's more concerned about getting a match going eventually, over trying to match extremely closely and that's largely due to population -- the group queue is a small segment of an already small playerbase, hence the much looser restrictions on matchmaking.

That being said, you are decidedly unrelenting in your lack of willingness to accept any sort of responsibility for the losses. Because in fact, you can be a huge part of the problem, and I don't mean that as a negative commment. This game does not have the simplest of learning curves bar none, and if you aren't looking at what you're doing at the conclusion of a match to learn from it, it's on you.


Btw, your WLR is 0.6 not an 80% loss ratio as you opine initially. That's your KDR.

What I'd recommend is trying some of the things mentioned above:

- get some advice on builds; very often matches are decided in the mechlab first
- work on surviving (without running and hiding mind you), but simply fighting and lasting as long as you can
- if you want to group, consider grouping with more than just two of you and focus on staying within mutually supporting distances
- focus fire what others are shooting at, don't have your own side-bar fights
- continue to ask for help from the community and utilize the new player stuff in that respective forum, plenty of it is helpful.
- consider signing on with a unit, especially ones that are newbie friendly (but not necessarily open invitation "zerg" units); ones that embrace the new player experience, have lore-aspects etc, are also likely to be willing to train and assist personally with newer players.

Good hunting!

[Ibrandul Mike]

Your sample size (with 100) is quite low. But yes you should probably have more like 70/30 to 50/50.
We all know the matchmaker is far from perfect. To be really good you would need a much bigger player base around the clock. Then you would need a more refined system to get players ranked. I am not a good player in Tier 3. I would put my skills more in the Tier 4 range (in a good working 5 rank system), so below average.
But back to your question: You have a really low win ratio. People assume first that it is your skills because there is only one common factor in all these matches. And that is you. Therefore they tried to answer how you could improve. For your sake and that of your team. You said 10% of your sample size was group queue. So these 10% are not really interesting. The other 90% are not enough matches to say it has to be 50% win/loss. If you had 1000 matches with that ratio I would agree. But you seem to think that all maps are totally balanced. That every player with Tier 5 has exactly the same skill level. That every team composition has exactly the same chances to win. Which is not correct. So to be totally frank... your core assumptions are not correct.
As for the matchmaker... basically it tries to get you in a team with people of your rank. And it tries to find another team with the same ranking. I have no clue which step domes first. The basic range for matches is your tier. If that takes too long it gets the next tiers (1 up and down). After that the next. So you have a max tier range of 5 as a rank 3 player and a max range of 3 as a tier 1 or 5 player. There might be circumstances where the range is wider but that isn't proven. So yeah in your sample size it could just be bad luck. We can't say for sure.
Take the help from people trying to help you. Don't waste that opportunity.
Perhaps it is your fault. Most likely it is better team play on the other side. We don't have the needed information to say for sure what the problem is.

Edited by Ibrandul Mike, 07 August 2016 - 05:36 PM.
corrected the psr error!

[Lukoi Banacek]

Ibrandul Mike, on 07 August 2016 - 04:10 PM, said:

PSR was the system used before the Tier system. And your sample size (with 100) is quite low. But yes you should probably have more like 70/30 to 50/50.

He is 60/40 so he's right in line with PSR MM currently.

[SlightlyMobileTurret]

Ibrandul, PSR IS tier system.

ELO was used before PSR.

[Nightbird]

Match maker isn't really responsible for creating 50:50 win loss ratio anyways, in solo queue it tries to create balanced teams, it doesn't put more good/bad players on one team to try to force a win/loss. Ultimately your performance will affect your win/loss chances a lot, at least in solo queue.

[Aleksandr Sergeyevich Kerensky]

ironnightbird, on 07 August 2016 - 02:06 PM, said:

Is your mouse sensitivity set to allow you to turn quickly but still aim accurately?

^a million times this!

This is the usual offender that affects most players struggling to get out of tier 5. Many of these players use the default in game mouse sensitivity (which is set to maximum...). While I dont know your current situation, but this advice is the best advice anyone can give without knowing exactly whats wrong.

If you fine tune this, I am sure you can raise in tiers extremelly fast! (I personally use .16~.18 which is far left, other people use different numbers, but I dont know anyone who uses the default extremity).

[Tarl Cabot]

Quote

Btw, your WLR is 0.6 not an 80% loss ratio as you opine initially. That's your KDR.

Just want to repeat this. The other is 50/50 item is not true, it is trying to create even teams, and that this is a GROUP endeavor, not a SOLO endeavor. You are not fighting just one other player, you and 11 others are fighting 12 other players. In the solo queue, the side that communicates the most, sticks together and focus fire will generally win more often than not, more so in the Tier 4/5. Communication is something the PSR/mm can calculate.

In the group queue, which takes the average of each group, regardless of weight class being used, can be very harsh, more so for new players cause you have stepped up from the little kids yard/channel to the *******/teenager yard/channel. Two, due to the numbers, the valves are likely released to fill up the group game. Three, there is usually weight discrepancies between teams, which is not used in the MM for the group game, iirc. This is where if it is just you and your brother in one group and you are taking only lights.meds, you are doing yourself and your team a disservice because you have left behind armor and firepower.

I would suggest staying in the solo queue until you and your brother improve, or hook up with other people/unit for the group queue. Or, get a 1 Day time of premium for both of you and head into the private matches against each other for a bit.

*************

As previously noted, system setup. Mouse Sensitivity, turn it down. Setup Torso lock to a toggle. That locks arms/torso weapons together until you need to unlock the arms. Just to repeat, Mouse Sensitivity, turn it down. When the game first came out, players edit had to manually edit user.cfg file or jack with their DPI mouse settings.

Editing.. one would think something is auto-correcting things but it is just my fingers typing things that are different from my thought process /shudders...

Edited by Tarl Cabot, 07 August 2016 - 06:08 PM.

[Ibrandul Mike]

Keshav I corrected the mistake. Thanks for pointing it out :-)

[11 lazrs errwhere 11]

Looked up your stats on the leaderboard. Your win rate is, as of exactly this moment, 0.63 Win/Loss, and 0.17 Kill/Death, with 83 Average Match Score.

The fact that you mixed up these numbers shows me that you aren't actually staying until the match ends to see how it turns out. If you want to improve your score, try watching what other players are doing after you die instead of immediately leaving. You'd be surprised how much you can learn simply by being observant of others.

Of course, as someone who spams drops too I don't have much right to say that, but I try to do my fair share of work before I get myself blown up (I have slowly improved myself to reach an average of 200+ match score). Other than (maybe) getting a computer upgrade, or simply working on improving your match score, there's not many more solutions I can offer. I know losing can be frustrating, but I would highly suggest reviewing your own gameplay. The most common beginner mistake is to go off on your own and make yourself the only available to the enemy team (which ends up in you getting blown up in about 3 seconds flat).

Edited by 11 lazrs errwhere 11, 07 August 2016 - 08:06 PM.

[762 NATO]

2 man group queue is a death sentence until you hit 4000+ matches, IMO. By then you should be in tier 2 or high 3 and have a very deep understanding of the meta and why the meta (and I still run builds that aren't most of the time.). Also by then you know what mechs are good for you and which ones you can't drive (me its assaults). Hellbringers, warhammers, centurions, jenners (not so much since rescale), blackjacks and shadowhawks hit all my happy spots. They ain't tier 1 meta but if you question it, I got lots of premium time and we can go into the 1v1 test map.

Find what works for you. Read mechspecs.com and metamechs.com. Use them for inspiration. Gman has a wonderful insightfulness to tier 1 comp/meta play. Other things work much better in lower tiers where you cannot rely on you team and have to pack in 7 tons of ac20 ammo so you can carry harder. Enjoy! And note there will be a big shift in the meta when Ghost Heat 2.0 comes out and everyone has to change how they play. It is a very exciting time for new players to get involved! My condolences to Gman129 in having to rewrite (most of) his entire site on that day.

Cheers!

Edited by 762 NATO, 07 August 2016 - 09:31 PM.