What Makes Psr System Unique a. The Way the Probe Is Read
#1
- Location Strana Mechty
Posted 24 September 2015 - 08:12 PM
The New PSR (Actor Skill Rating) Tier organization.
This article is meant to present data PL has helped me to deduce through analyzing postal service match scores and player stats post match.
At that place are five tiers, the lowest being 5 and the highest tier beingness tier ane.
We practise know PGI's distribution model is just stock-still PSR watermarks to meet. As the PSR acual values are hidden, PSR watermarks used are irrelevant since all one sees is a PSR Tier progress bar on one's abode folio. After each friction match, nether the player stats tab, a Greenish chevron pointed up indicates a PSR increase, a Yellow equal sign indicates no PSR modify, or a Red chevron pointed downwardly indicates a PSR decrease for that given match operation every bit measured by match score.
We have through trial and error deduced some of the watermarks for friction match scores on wins and losses that trigger a change or no alter in one's PSR.
On a Win:
PSR goes UP a lot if friction match score is >400
PSR goes Upwardly a moderate corporeality if match score is >250 but <=400
PSR goes UP a piddling if lucifer score is >= 100
PSR has NO Modify for a match score <100
Note: I do non have data on a zero match score if that would cause PSR to go downwards on a win, but I suspect PSR does not drop on a win, at the very least I accept not experienced that.
On a Loss:
PSR goes UP if friction match score is > 400
PSR has NO Alter for a match score >250 only <= 400
PSR goes DOWN a fiddling if match score is <=250
PSR goes Downwards a lot if match score is < 100
On a Tie:
Both sides, all players see a NO CHANGE to PSR regardless of friction match score.
Equally y'all tin can encounter, winning makes it easier to ameliorate PSR.
However, PSR can go upwardly steadily if you can consistently match scores above 250 and 400 on losses.
So what goes into match score?
The short answer is virtually everything from Spotting, flanking, protected formations, shooting down enemy UAV's to kills, assists, damage, etc. See more on this hither.
Currently, the single largest contributor seems to be damage dealt. Assists seem to be the second biggest contributor and then kills, the residual are very pocket-sized contributors in the nowadays state of PSR (September 2015), and may possibly change over time.
It appears to me that about half the damage dealt contributes direct abroad to lucifer score with assist, kills and all those other things making up the departure. Sometimes a PSR, no alter on a loss, tin occur without a high damage (>400, meaning obtaining well under 400 impairment) if many assists are made. But that assumes the losing team made many kills... which ways it was a close match anyway.
This is what we know at present, with more data it is discipline to alter. PGI will very probable alter it and not say how since they take not shared the higher up information watermarks.
Info on PSR from Paul @ PGI
Edited by 7ynx, x Oct 2015 - 04:47 PM.
#ii
- Location Seraphim HQ: Asuncion
Posted 24 September 2015 - 08:32 PM
Would similar to mention... I've had a lucifer score over 300 (I believe information technology was 330ish) and lost a match. PSR when downwardly. (Then I won a match with a match score of 200ish, and it went upwardly...)
Just notation I wish to add together, if it helps whatever.
#3
- Location Strana Mechty
Posted 24 September 2015 - 08:38 PM
Tesunie,
All my testing indicates that a losing friction match and lucifer score of 330 would generate a NO CHANGE to PSR as a outcome.
Your win is right in line with my testing.
Cheers for sharing.
#4
- Location Seraphim HQ: Asuncion
Posted 24 September 2015 - 08:41 PM
7ynx, on 24 September 2015 - 08:38 PM, said:
Tesunie,
All my testing indicates that a losing match and match score of 330 would generate a NO CHANGE to PSR as a event.
Your win is right in line with my testing.
Cheers for sharing.
Just saying. Don't now why then, simply I had a red down pointer when I lost. I only had the yellow equal symbol once when I had a rare (for me) lucifer score in the 400s. (Which does lucifer your information.)
Only want information to be as accurate every bit possible, even if I happen to be wrong.
I just don't feel it's right to preform better and lose and drop in PSR, and and then preform worse and win and increase PSR.
#5
Posted 24 September 2015 - 10:28 PM
Good start there, 7ynx! Some questions, if I may. And I figure fifty-fifty if they're not asking, the new folks would like to know these answers too. After all, we merely came off a match score based weekend claiming, and we're headed right into another in several hours. It would be helpful to know, then, exactly what ane must do to authorize/quantify the following:
Scouting
Flanking
Brawling
Protected (light, medium, heavy, attack, as appropriate)
Lance in Formation
Counter ECM (Specifically, how long, if NARC qualifies, if TAG qualifies, tin it be washed with PPC fire, etc)
Savior Kill
And so on. Those are the ones nosotros see most often, I'm sure. If you take some insight on the topic, it could but be helpful. Subsequently all, if EVERYONE in a given match makes a 300+ score, or if all but one or two on the winning team come in under 100, it makes no real difference. So permit'south help each other make the most of this coin-making opportunity!
#vi
- Location Searching alternate universes via temporal wormhole generator.
Posted 25 September 2015 - 04:47 AM
7ynx thanks for the link. Looks like I demand to become snipe,backstab,flank and help my team more.
#vii
- Location In your base of operations, killing your dudes
Posted 25 September 2015 - 05:59 AM
Perhaps Tier matters likewise. I had a friction match score of 450 and a loss resulting in a reduction.
#8
Posted 25 September 2015 - 07:04 AM
I hope so otherwise it would be style as well like shooting fish in a barrel to reach T1.
#9
Posted 25 September 2015 - 08:15 AM
Spungie, on 25 September 2015 - 07:04 AM, said:
I hope and then otherwise information technology would be mode also easy to accomplish T1.
Some higher tired players showed it'due south piece of cake equally information technology is, it's all about winning.
Get a group and win most of your games, as long as you lot go to do at least 200 dmg you volition go up in rating, more than so than y'all will on average lose rating on a loss.
It really is equally Paus said, it'south not showing skill it'due south showing how many games you played and, in role, how well you lot did in them, as you lot will climb college faster if you do max harm on good games and minimize the prs loss on bad games.
As the graph shows if you practice about the same in most of the games, loss or win, the gain from wins is larger so the loss from losses , pregnant that as long every bit you don't lose a lot more then you win ( without you doing enough to go the "=" prs change ) y'all will steadily get upwardly...
With things every bit they are at present this system only serves to not match players with less played games from the veteran players, at that place is no accounting for skill, as it never checks how many games you needed to get to your current prs or fifty-fifty what your average match score is...
#10
- Location Strana Mechty
Posted 25 September 2015 - 08:48 AM
From Paul:
#11
- Location Strana Mechty
Posted 25 September 2015 - 08:52 AM
Notice that Paul DID Non adhere any match score numbers to the Y-centrality, and probably for good reason, as it allows them to alter it whenever they desire to tweak things and non needing to inform us. Can be changed on the backend westward/o being tied to a patch update.
Based on the information we collected and the chart you posted, we tin friction match some match score watermarks to the Y-axis (until PGI changes the values ofc).
Winning Match Scores
Very High = in a higher place 400
Loftier = 251 to 400
Medium = 101 to 250
Low = Below 100
Losing Lucifer Scores
Very Loftier = above 400
High = 251 to 400
Medium = 101 to 250
Low = Beneath 100
This explains why consistent high friction match scores see the fastest progression at improving one's PSR.
Notation: our testing included members in Tiers 2,3,4, and 5. This testing saw consequent results. It is possible that Tier 1 operates with different watermarks, but imho I doubt that is the case. Though it would not be the first time I'd be incorrect.
Edited by 7ynx, 06 January 2016 - 10:58 PM.
#12
- Location Strana Mechty
Posted 25 September 2015 - 09:01 AM
TheRAbbi, on 24 September 2015 - 10:28 PM, said:
Good kickoff there, 7ynx! Some questions, if I may. And I figure fifty-fifty if they're not request, the new folks would like to know these answers as well. Later on all, nosotros just came off a match score based weekend claiming, and we're headed right into another in several hours. It would exist helpful to know, then, exactly what ane must do to qualify/quantify the following:
Scouting
Flanking
Brawling
Protected (light, medium, heavy, attack, equally appropriate)
Lance in Formation
Counter ECM (Specifically, how long, if NARC qualifies, if TAG qualifies, tin can it be washed with PPC burn, etc)
Savior Impale
So on. Those are the ones we meet most oft, I'thousand certain. If yous have some insight on the topic, it could simply be helpful. After all, if Everyone in a given match makes a 300+ score, or if all but i or two on the winning team come in under 100, it makes no real departure. So let's assistance each other brand the near of this coin-making opportunity!
Thank you rabbi! I am unable to quantify an reply to your question, only I similar where you lot are going with information technology. I, also, feel that the electric current match score as well heavily weight damage dealt. I have stopped running lights and medium mechs as a result of my findings in lodge to improve my PSR considering harm dealt is weighted so loftier on match score. A stealthy light pilot can pick upwardly a kill for less than 100 damage by either head shoting over heated mechs, or back stabbing LRM boats that are unaware, or just striking that cored CT section or even the side torso of a suspected XL engine. Thus making it much harder for average to to a higher place boilerplate light pilots to heighten their PSR, the elite light pilots will have no problem, I see them score 600-800 damage routinely, so it is possible, merely harder. Heavy mechs are simply easier (for me) to score that kind of damage.
Lastly, I exercise back up PGI'southward switch to the PSR arrangement. I just think it needs a bit of tuning is all. For the record, I am a tier 2 pilot at present and started PSR in the upper end of Tier three. Skilling up new mechs may temporarily cause a hit to i's PSR, simply I think this is to be expected.
cheers,
Lynx
Edited past 7ynx, 25 September 2015 - 09:06 AM.
#xiii
- Location Seraphim HQ: Asuncion
Posted 25 September 2015 - 09:09 AM
7ynx, on 25 September 2015 - 09:01 AM, said:
Lastly, I do support PGI's switch to the PSR organisation. I just recall information technology needs a bit of tuning is all. For the record, I am a tier ii pilot now and started PSR in the upper end of Tier 3. Skilling up new mechs may temporarily cause a striking to one'south PSR, just I retrieve this is to be expected.
The system is a dandy base of operations. I just feel information technology needs to be unlinked to wins and losses. So, PSR could raise and fall past bodily private performance. Match score values would need to exist adapted (such equally = results are 150-300. Going downwards would be beneath that, and up above). Then, win or lose your individual (not teams) operation would straight influence your PSR.
However, great piece of work with the current PSR system. Practiced data here.
#14
- Location Strana Mechty
Posted 28 September 2015 - eleven:06 PM
Proficient data on what goes into the MATCH score hither.
#fifteen
- Location Looking right at you, lining up my shot.
Posted 29 September 2015 - 01:24 PM
7ynx, on 24 September 2015 - 08:38 PM, said:
Tesunie,
All my testing indicates that a losing friction match and match score of 330 would generate a NO Change to PSR equally a result.
Your win is right in line with my testing.
Cheers for sharing.
Information technology'southward probably not based on lucifer score, but on a similar metric that'due south weighted slightly differently. That would explicate why Tesunie had a 300+ score and nevertheless lost some points in PSR. PGI probably wouldn't carp to keep the formula hole-and-corner if it was the aforementioned every bit the friction match score formula.
#xvi
- Location Strana Mechty
Posted 29 September 2015 - 01:51 PM
Sns,
PSR delta is based of the friction match result and player match score as documented by PGI's lead developer Paul Inuoye.
All we did here was to give those match score watermarks current values. They are subject to change at PGI's discretion at any time and I believe that is why they did not share the numbers. In fact I expect them to alter every bit the electric current system is good but not tuned optimally. It is too heavily based on harm dealt.
Edited by 7ynx, 29 September 2015 - 01:55 PM.
#17
- Location The states
Posted 29 September 2015 - 03:24 PM
I've had an increase on friction match losses with having decent damage, assists, kills, support of various kinds (for example, I carry a tag in my BK-x-KNT's head slot to assist missile boats and reduce heat load. Spotting and tag damage etc... adds up on the friction match score, I've frequently gotten ='southward on match losses and some increases. For the record though, this is unremarkably in the PUG queue, group queue I get smashed lol. I'll endeavor to tape some match scores this night and mail service back hither with scores on losses.
#xviii
Posted 29 September 2015 - 05:55 PM
Yeah, I've been keeping tabs on this. I tin can do pretty atrocious in a WIN, and go up in PSR. I can do an above-average (for me) job in a LOSS, and go downwardly. Not sure I've ever gone UP in a LOSS; maybe ONCE that's happened since we started keeping track. I haven't washed terrible enough yet in a WIN to become down in PSR.
WIN/LOSS weighs pretty heavily on PSR tier. It SHOULD weigh, only I wonder if information technology doesn't weigh Too heavily currently. Last drop I played so far today in a light, I was in a ACH-B with TAG and 4x cSPL. Did upper 400'due south in DMG, 4 K, 413-ish match score I think. LOSS. Came downward to me versus a pretty banged-upwardly Thunderbolt, and I screwed-upward trying to drop an artillery strike on him (he was camping his own cap in Assault, after nosotros caprushed it being behind by 2 kills). There wasn't much left of his mech, to be honest. And I'm shocked that I made information technology THAT long, what with an exposed body, missing both arms (I recollect), and both legs in BAD shape. That WIN should be worth more than to my fallen teammates, as whatever contribution they made left us, despite losing, leaving merely ONE enemy mech on the battlefield, and that a pretty desperately damaged one. As opposed to getting rolled 12-1 (with ME scoring the squad's only kill, and that subsequently the other eleven were down, saving SOMEONE a postal service-mortem assist at to the lowest degree), where whatever damage was washed didn't corporeality to much at all; the enemy was left with xi/12 mechs notwithstanding operating on the battlefield, almost the entire company.
So, margin and means of victory/loss SHOULD play into the PSR modify, and therefore, into the match score.
AND Over again, here I go playing the broken record. We demand to know what constitutes 'BRAWLING', 'SCOUTING', 'PROTECTED Calorie-free/MEDIUM/HEAVY/ASSAULT', and and so on.
Aye, perhaps nosotros're all PSR-obsessed right at present. Could be. Only video games have conditioned that in the states for DECADES. Who played PacMan only for the fun, and never paid attention to the Top Score or to raising their own personal best score? Huh? NO ONE. That's who. Now nosotros know if we're on the leaderboard (Tier i or non?), and we know what needs to improve to get u.s.a. there. SCORE. That is, PSR.
How do yous improve PSR? Yous play a heavy or assault, 1 of the meta-favorites, in a meta build, and you focus on DPS. Oh, you lot like to do recon? Absurd. Bask Tier 5, LOSER! You're really quite magnificent at burn support? Well, improve be able to average 800 D and four M per game, or you'll exist supporting in the basement, FREAK! No, do the meta thing that everyone else does, or suck it. In that location is no recognition for any other part, no affair how loudly PGI proclaim their support of ROLE WARFARE FTW! (That 'W' stands for LOSS, unless your role is meta-brawler.)
Merely buffing/nerfing certain mechs' sensor quirks isn't going to solve this, considering the PSR scores for anything other than MDK just aren't worth specializing for.
I've washed One driblet in a heavy in over 2 weeks. EBJ-B, cGauss, 5x cERML. Qualifying lucifer for the current claiming. I SUCK at Gauss, I SUCK at HPG Manifold, and I got > 300 with a Gauss-centric build on HPG Manifold, one that I haven't run in weeks.
Become out there with the SHCs and ACHs, and it'southward maybe every second or third match that qualifies, even though I do improve relatively speaking in those. I do MORE, anyway. I spot more enemies, I communicate more than with the team, I counter more ECM, and then on. And it's Simply Non WORTH Information technology unless I'thousand comfortable in the Underhive.
And I am.
F*** being a tryhard. That'southward about every bit dull equally this game gets. It's also what'southward rewarded in this game.
We really demand someone to get on that "Helm Tryhard and MetaGirl" comic...
#19
- Location Looking right at you, lining up my shot.
Posted 30 September 2015 - 10:34 AM
7ynx, on 29 September 2015 - 01:51 PM, said:
Sns,
PSR delta is based of the lucifer result and player match score every bit documented by PGI's lead developer Paul Inuoye.
All we did here was to give those match score watermarks current values. They are subject to change at PGI'due south discretion at whatever time and I believe that is why they did not share the numbers. In fact I expect them to change equally the current arrangement is skillful but non tuned optimally. It is as well heavily based on damage dealt.
I read that post and while the graph does say lucifer score, previous posts regarding PSR made information technology clear that the exact formula will be kept secret, which would be completely pointless if it was the match score formula as I already stated. Keeping only the benchmark numbers secret when the balance of the system is already known would be equally pointless because you've merely demonstrated how easy it would be to find those.
The fact that your caption fails to account for a detail event ways it must be flawed. This could be due to PGI tweaking values or it could be because you're making assumptions most how the system really works, but you won't know that until you've tested information technology thoroughly.
You can document diverse ways of reaching match scores around your watermarks to find out if whatever of them give unexpected results and try to come up with a more authentic system, but don't pretend you've establish the definitive respond when your enquiry is based on an unproven assumption.
#20
- Location Strana Mechty
Posted 30 September 2015 - 10:fourteen PM
No unexpected results, and I gained additional data points to narrow downward the high watermark at a lucifer score of 400.
SNS, please practice share with me your assumption on which assumption I've made regarding an unproven assumption? I am not understanding you. I am not providing annihilation definitive, just the matchscore equally they are currently for the match score h2o marks in Paul Inouye'southward graph. I have shared all my findings. Test them yourself if y'all are still a skeptic?
Source: https://mwomercs.com/forums/topic/210164-the-player-skill-rating-psr-system-explained-as-best-i-can/
0 Response to "What Makes Psr System Unique a. The Way the Probe Is Read"
Post a Comment