219 replies to this topic

[Bilbo]

tripcat1021, on 29 March 2016 - 02:13 PM, said:

I'm only playing when I can - which may be a negative. But I"m learning, building my Mechs I now have 3, and one more on the way in May. I get in the public matches and for the most part they are scrums no team organization. The ones I do well in - are organized - most are just freaking chaos. I'm still adjusting my targeting and my sensitivity because I can't target to save my butt. I'm Tier 5. As long as that tier does not keep me from playing I'm having fun. Will I get better, well I"m not staying up from 11:00 to 2:00 AM playing with the groups out there. I would love to but I have a job and a house with all kinds of to-dos. So if playing a lot is the key to the a higher PSR. Then I will always be a 5 I guess ... Hell I still say cadet but I have probably only play in contiguous time - 2 weeks, MAYBE three weeks. That's it.

Playing a lot is not key. Doing well in more matches than you do not do well is the key to moving up. Playing a lot of matches in this manner simply gets you there faster, but there is no real reason to rush up the tiers anyway.

[General Solo]

Me thinks
If you play enough games and are lucky to be on the winning team most of the time, its possible that a lower skilled player could reach Tier 1.

However imho the rate at which you climb the Tiers does represent skill

For example

Player A takes 2 weeks to get from Tier 5 to Tier 1
Player B takes 11 months to get from Tier 5 to Tier 1

IMHO Player A is moar skilled than player B, IMHO

[Bilbo]

OZHomerOZ, on 29 March 2016 - 04:16 PM, said:

Me thinks
If you play enough games and are lucky to be on the winning team most of the time, its possible that a lower skilled player could reach Tier 1.

However imho the rate at which you climb the Tiers does represent skill

For example

Player A takes 2 weeks to get from Tier 5 to Tier 1
Player B takes 11 months to get from Tier 5 to Tier 1

IMHO Player A is moar skilled than player B, IMHO

Assuming they played the same number of games you would be wrong, in my humble opinion.

[General Solo]

Bilbo, on 29 March 2016 - 09:40 PM, said:

Assuming they played the same number of games you would be wrong, in my humble opinion.

I'm comparing apples with apples, so....

My assumption is that they play games at the same rate over time
After 2 weeks they have played the same number of games and therefore after 11 months they have both played the same number of games.

So for players which play at the same rate (for example 10 games a week), the player that climbs tiers fastest is in my opinion more skillful.

[Nine-Ball]

OZHomerOZ, on 29 March 2016 - 11:15 PM, said:

I'm comparing apples with apples, so....

My assumption is that they play games at the same rate over time
After 2 weeks they have played the same number of games and therefore after 11 months they have both played the same number of games.

So for players which play at the same rate (for example 10 games a week), the player that climbs tiers fastest is in my opinion more skillful.

Are we talking strickly PUG dropping alone here?

And are we talking people who have to grind for their mechs and skills or people who have everything mastered?

Does it matter if one is a scrub light or someone who likes to pilot jack-of-all-nothing mediums, or someone with bravado and big nuts piloting a heavy or assault?

Besides if we are talking PUG dropping here you have to take into account the nature of PUGs. Some days you'll be winning without even firing a shot, or get carried to wins all day while in others you can't even buy a win.

PS. HI BILBO! :>

[Kuritaclan]

OZHomerOZ, on 29 March 2016 - 11:15 PM, said:

I'm comparing apples with apples, so....

My assumption is that they play games at the same rate over time
After 2 weeks they have played the same number of games and therefore after 11 months they have both played the same number of games.

So for players which play at the same rate (for example 10 games a week), the player that climbs tiers fastest is in my opinion more skillful.

It doesn't account for randomness. While yes you somewhat have a point with the assumption of same rate over time, it is stil weak. No battle is like the other. And both players wouldn't have the same teammates all the time during this progress. So to speak even if both players have the same W/L Ratio one can carried by team and get on average higher matchscores what translate in higher PSR gains vs the other.

[General Solo]

Kuritaclan, on 30 March 2016 - 02:36 AM, said:

...... While yes you somewhat have a point with the assumption of same rate over time, it is stil weak .......

That is about what I am saying. Rate of ascent
in Tier level is some what related to skill, all other factors being equal.

Edited by OZHomerOZ, 30 March 2016 - 04:27 AM.

[PyckenZot]

OZHomerOZ, on 30 March 2016 - 04:27 AM, said:

That is about what I am saying. Rate of ascent
in Tier level is some what related to skill, all other factors being equal.

Rate of ascent means nothing in PSR. Rate of ascent is defined primarily by mech choice. As repeated ad nauseam, PSR is putting way to much weight on damage done. You want to go up. Choose a mech that deals 400-500 damage per drop consistently. (Lurm boats, Dakka builds,...)

[General Solo]

PyckenZot, on 30 March 2016 - 04:47 AM, said:

Rate of ascent means nothing in PSR. Rate of ascent is defined primarily by mech choice. As repeated ad nauseam, PSR is putting way to much weight on damage done. You want to go up. Choose a mech that deals 400-500 damage per drop consistently. (Lurm boats, Dakka builds,...)

I did say "all other factors being equal" which would include mech choice

[MW222]

tripcat1021, on 29 March 2016 - 02:13 PM, said:

I'm only playing when I can - which may be a negative. But I"m learning, building my Mechs I now have 3, and one more on the way in May. I get in the public matches and for the most part they are scrums no team organization. The ones I do well in - are organized - most are just freaking chaos. I'm still adjusting my targeting and my sensitivity because I can't target to save my butt. I'm Tier 5. As long as that tier does not keep me from playing I'm having fun. Will I get better, well I"m not staying up from 11:00 to 2:00 AM playing with the groups out there. I would love to but I have a job and a house with all kinds of to-dos. So if playing a lot is the key to the a higher PSR. Then I will always be a 5 I guess ... Hell I still say cadet but I have probably only play in contiguous time - 2 weeks, MAYBE three weeks. That's it.

If you haven't completed all the training go ahead and do it. It helps and you get a bit of C credits. Then the next step is tactics, position and location awareness. Shoot and move, move and shoot. Never attack head on unless you have a clear advantage, double team the red guys as much as you can, other wise go for the flank attacks or from the rear. Use your battle map and zoom but do not leave the zoom on as you will miss other information and sometimes shoot your lance mates that are close. For now aim center mass and as you get better look at the mech (s) you have locked with the "R' key aim at most damaged components. If you are not using Team Speak or in game VoIP start, communication is key, ask for help, call targets, report flanking to your team mates and offer help as you can. If you are flanking let you team mates know they may go with you or they may watch your back or tell you it's not a good idea.

Lastly have fun.

[Seraph6]

Having not played since the closed beta, I'm new to the game relatively speaking. I've been watching my rating over the past few weeks.

My findings:

5 is low (where I started) and 1 is high. I move closer to 4 if I play well (based on exp earning on the close game screen). This is regardless on winning or losing. I've not noticed an increase more when I win (though I've not been looking out for it).

When i started looking I was above the 'n' of 'minimize' that appears below the bar. I'm now passed the 'e' and moving to the finish where I think I will move to a 4.

I'm playing the HBR with CERL, so my dmg is high power game. I'll choose a lighter mech and see if dmg or exp effects this more and report back here

Edited by Seraph6, 01 April 2016 - 10:17 AM.

[SidBlue]

Ok my question is,when you group with someone with tier 1. What happens with the group tier? Do we all go to Tier 1? Does it average out or what.Seems like everytime i get a group started and a buddy of mine joins"that is tier 1. We get our butts kicked. Im in tier 5 and i get more kills now that i have found the mechs I am better at playing. So i know i dont suck that bad,i mean everyone has that bad day.

Edited by sidblue, 02 April 2016 - 07:22 AM.

[MW222]

sidblue, on 02 April 2016 - 07:19 AM, said:

Ok my question is,when you group with someone with tier 1. What happens with the group tier? Do we all go to Tier 1? Does it average out or what.Seems like everytime i get a group started and a buddy of mine joins"that is tier 1. We get our butts kicked. Im in tier 5 and i get more kills now that i have found the mechs I am better at playing. So i know i dont suck that bad,i mean everyone has that bad day.

That's a Dev question but as far as I know you already have the right of it. The tier matching is not very sophisticated nor a terribly complex "algorithm" .

[Victus B]

Look, I'm just going to say what the devs of this game don't seem to be willing to say, however I have played other 'pub battle' matchmaker games for years. Seriously thousands of lost hours on World of Tanks and War Thunder alone and they have similar systems and rating systems used by the playerbase to 'rate' players...mostly in using W/L ratio as a rough guide.

Many things have already been said in this thread that are VERY true, such as this form of rating is not very efficient and does not do much to actually show a players individual skill or tactical play. From what I can tell in my limited time here a pilot who plays mostly support lights (ecm, NARC, TAG focused) are prob going to stay in Tier 5, mabe tier 4...a good Heavy or Assault solo player will prob hang out around tier 3 at best...and here is why:

Its a 'ranking system' to segregate people who play mostly 'pubs' solo and those who usually team up. ie platoon, clan, guild whatever.

To use an example from another game, in WoT the players who had a W/L ratio over 1.5 were considered 'cream of the crop' peak players...and yet if you ever played against them for awhile you would realize they generally were not godlike or much more than 'above average' players at best, if even that...but what you NEVER saw was them playing SOLO! They were always in a platoon of 2 or 3 on teamspeak or linked up, and generally with people they played with all the time...so they could in fact sway pub battles where most everyone else is winging it...but even then only a few .1ths...in the games stat algorithms and the public eyes though that means they must be godlike skilled le3t players right?

What this tier system means is players who play solo will almost always be matched with others who do, and 'clans''guilds''companies' who play together regularly or on a 3rd party comms system will generally play against other players who play in the same manner.

I for one think its a great way to segregate the community in a way that ensures people are all 'playing the same game' and preventing constant steamrolls at the hands of 3 or 4 players on one team who are in sync and dominating every 'random pub' match they play.

Edited by Victus B, 13 April 2016 - 07:05 PM.

[Roadkill]

Seraph6, on 01 April 2016 - 10:16 AM, said:

5 is low (where I started) and 1 is high. I move closer to 4 if I play well (based on exp earning on the close game screen). This is regardless on winning or losing. I've not noticed an increase more when I win (though I've not been looking out for it).

It isn't based on XP, it's based on match score. Here's how it works:

match score  win  loss
  0 to 100     =  ---
100 to 250     +  --
250 to 400    ++  =
400 and up   +++  +

The exact values of + and - aren't known, so the amount you go up or down can only be expressed relatively as I've done in the chart.

Your intuition is correct, though. If you play well your PSR will go up, and if you play poorly it can (though often doesn't) go down.

One note on match score - the largest contributor to match score is damage done. Half of your damage done is added to match score, so if you want to ensure that you get a match score of 250 just make sure you do 500 damage. 400 damage will usually get you there, though... damage typically makes up 70-80% of your match score.

[Karl Streiger]

An average game would place you in between a MS for 150-300.
While the chances are better that your score rises during a victory is simple based on the concept of assists and the snow ball effect when the first mech is falling.

Anyhow. With 4 average games and 2 looses and 2 victorys the chances are very good that its 1x --; 1x+; 1= and 1++

just an example for some average games during the last event
/tr]
[tr]

'Mech	Match Score	Kill Most Damage	Damage Done	Kill Assists	Win	PSR
ZEU-9S	127	1	135	6	1	+
ZEU-6SR	323	1	521	7	1	++
GAR-A	339	1	512	5	1	++
GAR-A	101	0	167	0	0	--
WHK-PRIMEI	253	0	432	7	0	=
AS7-S	301	1	490	3	1	++
KGC-000B	100	0	144	4	0	--(-)
GRF-2N	273	0	339	9	1	++
GAR-C	252	2	294	6	1	++
GAR-C	137	0	242	1	0	--
NVA-S	120	0	164	2	0	--

So its a +++ and a W/L of 1.2 - the change in the tier bar - must have been something from the i in Tier towards the e|r in Tier - great measurement

[C E Dwyer]

Blue Pheonix, on 22 September 2015 - 02:25 PM, said:

Hi all,

How is the Pilot Skill Rating (PSR) and tier level determined?
Is tier 1 the best players and tier 5 the worst?
I noticed this "PSR" after I did a patch today.

I searched these forums and did a Google search to find this answer but all that I really came across is if this info should be public or not.

Thanks!

You get a donkey drunk and when it craps in a square, that's how you work out your tier.

Sooner or later that donkey is going to crap on the 1 square, that's how you get there, just play and wait, no skill needed might take a year or two but one day the donkey will drop it's load in that square

[Ex Atlas Overlord]

Blue Pheonix, on 22 September 2015 - 02:25 PM, said:

Hi all,

How is the Pilot Skill Rating (PSR) and tier level determined?
Is tier 1 the best players and tier 5 the worst?
I noticed this "PSR" after I did a patch today.

I searched these forums and did a Google search to find this answer but all that I really came across is if this info should be public or not.

Thanks!

K/D means absolutely nothing
W/L means absolutely nothing

Your Tier is determined entirely by how consistently you can earn a high "match score".

You gain "match score" for your actions during a game.... I don't know the vaules...you'd have to find these.

At the end of the game if your team won and your match score is over 250? The MM will consider you too good for your current Tier and start moving you up.

If your team won and your match score was less than 250 the MM will consider you in the correct Tier and it will not change.

If your team lost and your match score is over 450 the MM will consider you too good for your current Tier and move you up.

If your team lost and your match score is 250-449? the MM will consider you in the correct Tier and it will not change.

If your team lost and your match score is below 250 the MM will consider too bad for your current Tier and move you down.

Summary: You going up or down has everything to do with your actions in a game, and the match score you get as a result of those actions.

If you constantly have a high match score you will go up over time, b/c winning and losing won't matter.

If you constantly have a terrible match score you will go down over time, b/c winning and losing won't matter.

If you have both good and bad match scores equally...you're probably in the right Tier.

Edited by The Atlas Overlord, 14 April 2016 - 04:35 AM.

[Roadkill]

The Atlas Overlord, on 14 April 2016 - 04:32 AM, said:

K/D means absolutely nothing
W/L means absolutely nothing

Your Tier is determined entirely by how consistently you can earn a high "match score".

You gain "match score" for your actions during a game.... I don't know the vaules...you'd have to find these.

Correct so far.

Quote

At the end of the game if your team won and your match score is over 250? The MM will consider you too good for your current Tier and start moving you up.

If your team won and your match score was less than 250 the MM will consider you in the correct Tier and it will not change.

If your team lost and your match score is over 450 the MM will consider you too good for your current Tier and move you up.

If your team lost and your match score is 250-449? the MM will consider you in the correct Tier and it will not change.

If your team lost and your match score is below 250 the MM will consider too bad for your current Tier and move you down.

Mostly incorrect. Refer to the chart I posted above for the correct values.

Quote

If you constantly have a high match score you will go up over time, b/c winning and losing won't matter.

If you constantly have a terrible match score you will go down over time, b/c winning and losing won't matter.

If you have both good and bad match scores equally...you're probably in the right Tier.

Mostly correct. Win/Lose does matter, because it affects how your PSR is adjusted. Again, refer to the chart that I posted above.

[Davers]

Tiers don't mean as much as they should. We all see post about people complaining about losing streaks, but you never see posts about people losing tiers. If the system punished losing by actual fear of tier loss than people would complain that the matchmaking should be much tighter, and obviously the population is too low (at all hours and regions at least) to do that.