How Is The Pilot Skill Rating (Psr) And Tier Level Determined?
#101
Posted 06 January 2016 - 10:55 PM
#102
Posted 09 January 2016 - 01:46 AM
#103
Posted 13 January 2016 - 05:31 PM
PhoenixHawk OSF, on 09 October 2015 - 11:59 PM, said:
This. I feel like I have absolutely no control over this. If the light lance runs off like a bunch of idiots and we lose the slow 8 v full 12 battle which ensues that is not my fault but win/loss has the largest impact of all. It is extremely difficult to get even an = if you don't win.
#104
Posted 14 January 2016 - 11:41 AM
This is a team game and the way it is worked out is thus team dependant. Now as PGI made the game you would have thought that they could work that simple fact out?
Add to that the multitude of mech classes and possible builds , actual player skill would only be part of this game anyway , as ability to make a good build also plays a large part.
There were threads about ELO stating the whole multitude of flaws in any rating system , But as usual PGI took no notice... after all they know best!
If they were realistic they would realise that each mech would need it's own rating (but the mech would also need a power rating based on how it was set up, like a handicap), but again it being a team game all results ultimately come down to the team!
If you really wanted realistic ratings you would put everyone in identical mechs and have a 1 vs 1 ladder... And that would take a very long time!
#105
Posted 14 January 2016 - 01:16 PM
USMVmc, on 09 January 2016 - 01:46 AM, said:
Then carry your team. I'm a tier 1 light pilot stuck in tier 4 with no boost since I had a CB account and just came back. I win 90%+ of my solo queues, consider 4 kills a bad game, average 7+ kills, and get 700+ damage every game. At the lower tiers it is fully possible to basically win the game by yourself, I should know it's what I do almost every game.
If you truely are a higher tier than what you say you are then enjoy being in the crapshoot tier. I love getting 500,000+ cbills every game. I'm eventually going to get to tier 1, but you can be damn sure I'm going to take as much time as I can down here at tier 4 clubbing baby seals.
#106
Posted 14 January 2016 - 02:28 PM
Win 0-100 - 100-250 = 250-400 + 400+ ++ Loss 0-100 -- 100-250 - 250-400 = 400+ +
The current problem is two-fold:
1) you can't lose PSR in a win
2) it's not zero sum, which makes it more of an XP rating than a skill rating
Edit: fixed loss/win typo. Thanks Bilbo!
Edited by Roadkill, 21 January 2016 - 12:45 PM.
#107
Posted 14 January 2016 - 02:50 PM
Roadkill, on 14 January 2016 - 02:28 PM, said:
Win 0-100 - 100-250 = 250-400 + 400+ ++ Loss 0-100 -- 100-250 - 250-400 = 400+ +
The current problem is two-fold:
1) you can't lose PSR in a loss
2) it's not zero sum, which makes it more of an XP rating than a skill rating
I think you mean you can't lose PSR on a win.
#108
Posted 14 January 2016 - 03:09 PM
#109
Posted 14 January 2016 - 05:07 PM
#110
Posted 14 January 2016 - 08:08 PM
Schmalis McAdder, on 14 January 2016 - 05:07 PM, said:
Show me the screenshot of your last 20 games and I will tell you just what is going on. You are probably just having selective memory.
#111
Posted 16 January 2016 - 03:06 AM
Toxic like WoT...they will come outta the woodwork.
Edited by USMVmc, 16 January 2016 - 03:13 AM.
#112
Posted 16 January 2016 - 03:55 AM
Blue Pheonix, on 22 September 2015 - 02:54 PM, said:
I have been playing for aprox about 6 months now, ballpark. Maybe more, maybe less.
My PSR says I am a tier 5 so I must really suck or the system/rating is really messed up.
I actually consider myself a good player. I mostly run lights (Jenners) though and have since I started. While I died a lot in the beginning while I was still learning the game, I usually:
- score around 300-500 on my match scores.
- I target and hold targets for LRMs
- I get points for scouting multiple targets
- I usually last the whole round or am among the last 3-4 to die
(which is easy to do if your team is being "spanked" and everyone starts targeting the last who are left).
- I usually do around 300-700 damage per round (so I am not just hanging back to survive). I could do more in my Oxide if it had more ammo.
With these things being true, I am confident I am not a tier 5 quality player if indeed tier 5 is the worst and tier 1 is the best.
I honestly play a lot now and would have thought I am more in line with high tier 1 through low tier 3. Somewhere in there. I believe I am top 1/3 and am really confident I am at least top half. This is why I ask the question because this is so shocking to me. I know I am truly not among the group of the worst players in the game. The ones who don't really know what they are doing yet. I am so confident of this that I would be willing to bet money on this.
If you were actually getting "300-500" match score then you would still be Tier 5 until you break into Tier 4, because Tiers function similarly to Experience Bars. If you are scoring "300-500" per match then it will still take a while for you to progress from the bottom of the Tier 5 bar into Tier 4, and then proceed to score "300-500" through the Tier 4 bar into Tier 3, etc. Keep in mind it is far, far easier to get a Match Score of "300-500" in Tier 5 than it is Tier 1.
#113
Posted 17 January 2016 - 08:14 AM
Blue Pheonix, on 22 September 2015 - 02:54 PM, said:
I have been playing for aprox about 6 months now, ballpark. Maybe more, maybe less.
My PSR says I am a tier 5 so I must really suck or the system/rating is really messed up.
I actually consider myself a good player. I mostly run lights (Jenners) though and have since I started. While I died a lot in the beginning while I was still learning the game, I usually:
- score around 300-500 on my match scores.
- I target and hold targets for LRMs
- I get points for scouting multiple targets
- I usually last the whole round or am among the last 3-4 to die
(which is easy to do if your team is being "spanked" and everyone starts targeting the last who are left).
- I usually do around 300-700 damage per round (so I am not just hanging back to survive). I could do more in my Oxide if it had more ammo.
With these things being true, I am confident I am not a tier 5 quality player if indeed tier 5 is the worst and tier 1 is the best.
I honestly play a lot now and would have thought I am more in line with high tier 1 through low tier 3. Somewhere in there. I believe I am top 1/3 and am really confident I am at least top half. This is why I ask the question because this is so shocking to me. I know I am truly not among the group of the worst players in the game. The ones who don't really know what they are doing yet. I am so confident of this that I would be willing to bet money on this.
Rest assured that the tier metric is “really messed up” as you put it. No doubt it also alienates many of the new players we just gained with the Steam launch. For what it's worth, many readers here know that and likely also understand and appreciate your skills as a light pilot. Although I've only logged about 2700+ matches so far (all PUG), I know a good pilot when I see one. Most of the best pilots I've seen are light pilots. I respect them most (I'm not a light pilot). In fact, I can't bring myself to “execute” a smoking legged Urbie (reserved for the bravest light pilots of all).
You remind me of an excellent light pilot friend. He went from tier 4 to 5 to 4 to 3 to 4 and now 3. Why? Because he likes to also buy and experiment with heavier mechs. He cares about fun, not tiers. Another friend, a tier 2, pointed out that tier 4 and 5 are actually fun tiers because you get “breathing room”, more space to test other mechs, builds and tactics.
That's where the real problem with that nice bragging point metric which forum-numptees will tout as some miracle-stat . It costs PGI money! Obviously, a pilot will do best in the mech that he or she is comfortable with. However, experimenting with new mechs, loadouts etc not only costs money (let's exclude most of the amusing trial mechs for now), it also affects piloting performance in those mechs for a while. There is a very good chance that KDA win/loss etc will suffer, at least temporarily, right along with the tier bar. For some (maybe many?), this risk invites mech-polarization, staying with “what works” and stifles diversification. Diversification is not only fun, it also puts plenty money in PGI's pocket.
#114
Posted 17 January 2016 - 04:05 PM
#115
Posted 17 January 2016 - 04:36 PM
Blue Pheonix, on 22 September 2015 - 02:25 PM, said:
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!
PSR is an XP bar...it has been mathematically proven, if you play long enough, you will eventually get to tier 1.
This is because the operations and math are not zero sum.
#116
Posted 17 January 2016 - 04:40 PM
Gyrok, on 17 January 2016 - 04:36 PM, said:
PSR is an XP bar...it has been mathematically proven, if you play long enough, you will eventually get to tier 1.
This is because the operations and math are not zero sum.
The tier system is a story told by an idiot, full of noise and emotional disturbance but devoid of meaning. - Macbeth
#117
Posted 17 January 2016 - 04:49 PM
Gyrok, on 17 January 2016 - 04:36 PM, said:
PSR is an XP bar...it has been mathematically proven, if you play long enough, you will eventually get to tier 1.
This is because the operations and math are not zero sum.
yup, this exactly.
when it was first being discussed, ELO was bagged out and binned as a solution, so PGI knowing better made PSR.
a system that is more about volume of matches, rather than a players individual skills. Your skills will just limit your negatives when your side loses, but its essentially an EXP bar.
#118
Posted 21 January 2016 - 10:47 AM
#119
Posted 21 January 2016 - 12:35 PM
Playing so much now I have figured out that a players tier level is largely based on 3 things
1. Winning the match - this is simply the biggest factor. You can do poor damage and die but if your team wins then you will most likely receive a yellow equal sign (no change to your player rating) and possibly a green arrow (increase in player rating). If you do decent damage and still die and your team wins, you will mostly likely receive a green arrow. If you do good damage and your team loses, you still will likely receive a red arrow (decrease in player rating). If you do the most (or close to) damage on the team and your team loses, you will likely receive a yellow equal sign. If the team doesnt play as a team, people run off solo and get blasted so soon leaving you with 3 on 1 and you are great player who really carried the team... tough cookies. At least, this is what the current rating metric says. This brings me to my next factor:
2. Damage done in a match - while your team winning is the single most important factor in improving your team rating, damage done is second. While "scouting","lance in formation", "brawling", "lock damage" (and other team skills etc.) all play a roll in match score, no statistic seems to be as important as damage in raising your tier rating. Trust me on this. The rest of the stats seem underscored comparatively. Ive focused on all of them at various times. As discussed above, if your team wins, you are most guaranteed at least a yellow equal sign and possibly a green arrow. If your team loses, if you do enough damage, you can at best avoid a decrease in player rating by getting a yellow equal sign.
2. Number of matches played. As stated in the two previous points, if your team wins enough, you do enough damage each match and you play enough matches, you will move up from the higher tiers (3-5). Your progress bar moves up very slowly so keep grinding, winning as a team and doing damage. Unfortunately for scouts and lights with very little armor and weapons, nothing else really matters then these things.
I say this to encourage those who are discouraged with their tier level. You may indeed actually be better then your tier level. By switching to other mech types and mastering them, it can make the trip in improved tier rating possible. Widen out and get good in multiple mech types. Different mechs are good for different tiers. There are only a handful of mechs that are good at all tier levels. "Master your mechs and keep grinding the until you get to the tier you desire.
Edited by Blue Pheonix, 21 January 2016 - 08:39 PM.
#120
Posted 21 January 2016 - 08:44 PM
It seems that it's still possible to raise your PSR even after a loss. However, 500 damage and three kills doesn't seem to be enough to cut it, although it's good enough to prevent you from going down.
I'm almost tier 4, so numbers are much easier to pull off. I reckon you'd have a very difficult time getting these numbers on higher tiers
Perhaps if I had more assists in the third game, It would have gone up.
Edited by Gar Kerensky, 21 January 2016 - 08:47 PM.
3 user(s) are reading this topic
0 members, 3 guests, 0 anonymous users