46 replies to this topic

[Roadkill]

Ashvins, on 08 February 2016 - 08:27 PM, said:

For the most part this is true. But you sometimes get "stuck" with a "terribad" team.

Conversely, sometimes you're on the other side of that equation and get "gifted" a "superteam."

Since that determination is essentially a random coin flip, it evens out in the long run. But since PSR is biased toward rising, that coin flip ends up increasing your PSR over time. Not by a lot, but it is an overall increase on average.

[jss78]

Roadkill, on 09 February 2016 - 09:59 AM, said:

Conversely, sometimes you're on the other side of that equation and get "gifted" a "superteam."

Since that determination is essentially a random coin flip, it evens out in the long run.

Agreed, uneven games happen, but they shouldn't affect anyone's PSR over the long haul.

Roadkill, on 09 February 2016 - 09:59 AM, said:

But since PSR is biased toward rising, that coin flip ends up increasing your PSR over time. Not by a lot, but it is an overall increase on average.

It's tough to really get a handle on how much the PSR system is skewed towards everyone rising. The problem is that people's skill level does actually rise as they play more, so a gradual rise for most people should be expected and is not an indication of the system not working.

Looking at the tiers of forum people who keep them public, my hunch is that we do in fact have people who are in lower tiers despite having played for a long time, so not everyone is bubbling up. But it's tough to know how many of these people we have.

My own anecdotal experience is that it's a little skewed though. In my own Tier 2 games I'd estimate I'm very nearly average -- typically right in the middle if I rank my team's players by match score. This suggests that my current tier is about right. Yet the e-peen bar keeps growing relentlessly.

[Roadkill]

jss78, on 09 February 2016 - 10:15 AM, said:

My own anecdotal experience is that it's a little skewed though. In my own Tier 2 games I'd estimate I'm very nearly average -- typically right in the middle if I rank my team's players by match score. This suggests that my current tier is about right. Yet the e-peen bar keeps growing relentlessly.

I think you're right (at least I certainly hope so) in that there are people who will "max out" at very tier. The problem that I think most people are talking about is that the threshold to get into Tier 1 is really pretty low.

Any given person's typical match score will go down as their tier goes up, or at least it should. That's because the competition in matches becomes more fierce in theory. But because the threshold for your PSR increasing over time is so low, people will continue to climb anyway.

The 250 and 100 thresholds really aren't that hard to beat. It's a bit of a challenge to reach 250 in a loss, but at least in my experience it's no more difficult than reaching 400 in a win. And in that situation, your PSR is still going up over time.

Note that this has nothing to do with your relative ranking within a match. You could be dead average compared to other players, but if your match score is 250+ (win or lose) you're still "under ranked" according to PSR. Conversely it's also possible to finish in the top third according to relative match score, but if you did that with a match score of 200 (read: less than 250) then you're "over ranked" according to PSR.

[jss78]

Roadkill, on 09 February 2016 - 10:36 AM, said:

Yeah, the reason I mention the relative ranking within your team is that to me that's the "real" metric of whether you're in the right tier. If you're continuously in top 3 in your team (better than your team mates), then you're arguably playing with the wrong people (too low tier). This is why I like to pay attention to it, it's kind of the bottom line of how I'm doing among the people I play with. Any match score thresholds such as 100 or 250 are arbitrary criteria.

Anyway, I agree the system is probably skewed towards too many people bubbling up. But my hunch is that it's not quite as bad as some people think.

Edit: I'd suggest that this is a useful metric to look at if PGI's doing any sort of data mining about how the PSR system's working, i.e., people's average ranking within their team, and the rate of change (and direction) in their PSR. Are the average performers staying level or not? I presume PGI's already doing something like this though.

Edited by jss78, 09 February 2016 - 11:27 AM.

[jonfett]

Thank you everyone who has contributed meaningful info on this topic. I'm picking myself up and dusting off every match and am determined to move forward. I've since had some really good matches and am definitely noticing more often than not true teamwork in Puglandia. There are still a few salty players, but it's truly awesome to see other players go out of their way to help you out during a match.

...and a shout-out to NARC BAIT!! There were so many times I had no idea how I kept getting narced in matches. He is so stealthy and never gives away his position unless warranted.

[Helsbane]

Short answer to the original question of this thread = Flamers. Just outfit a light mech with a pile of flamers and charge the enemy team. Lather, rinse, repeat. Hell, keep a few of them kitted out so you can drop out of the match after you die and repeat the entire process. As a fun mini game, see how many you can have stuck in various games at the same time!

Just follow that simple process and you can be back at the lower end of tier 4 in no time.

[MrMadguy]

That's, what usually happens, when dump valves to Tiers 5 open - you solely kill 1/3 of enemy team. And it's good for them, if they have good players too - in this case it usually feels like your team immediately collapses, when you die. I.e. it's 1:1, 2:2, 3:3...6:6...you die...12:6 loss. But not in this case.

Another one, lol:

Edited by MrMadguy, 10 February 2016 - 02:19 AM.