37 replies to this topic

[Marius Evander]

What title says. unless its a 1v1 ladder it means nothing coz people level mechs, play fun builds, get unlucky bad teams, get dc's on team, play only meta mechs and never level a mech while others do, have different quality pc's.

etc etc etc, the whole system is flawed and most of it reflects how much you play / how little rl commitments you have not how good you are at working your muscle memory.

individual Stats lives in the slowest paced FPS in existant matter roight.

Also, did that 3 page KDR thread I was trying to read get deleted ?

Edited by Cadoazreal, 02 February 2017 - 02:51 AM.

[Ghogiel]

There have been murmers of ranked queues for looooonnnnngest time.

PGI just don't seem to get it though.

As for the stats, they need to be split group/solo and show tier. Plus add KMDD stat as well. Then they start to become more meaningful. As they are right now they don't really rank anyone, all you can do is tell who the complete potatoes are from everyone else and kinda who is decent to good. But you can't really tell them apart in those groupings.

Edited by Ghogiel, 02 February 2017 - 03:12 AM.

[Marius Evander]

They should probably add a "dont use this match/games stats in my overall stats coz im testing stupid builds / leveling mechs " option. Not that it would matter.

[El Bandito]

After a thousand matches, one can easily tell who is scrub and who is not, all stats considered. And I know plenty of good players who help their teams win through good calls, but cannot do as well in 1v1. 1v1 is a much different game than 12v12, in the first place.

Also, SQ and GQ stats need to be separated, as Ghogiel said above.

Edited by El Bandito, 02 February 2017 - 06:24 AM.

[Tarogato]

Playing unleveled mechs, in my experience, doesn't necessarily degrade performance that noticeably. I've been making a habit of tracking some of my mechs as I level them, because I'm a spreadsheet nerd... and some of my mechs performed better without Basics than they performed after I finished Elites. Why? Because other factors have more impact than whether your mech has skills on it or not.


What I actually do is play all of my mechs without leveling them, I just stock pile the XP for 20 matches or so, and then I save my stats. Then I spend all the XP at once, get all the variants Elited in one go, and play 20 more matches again per variant, and compare the before and after. On average, my mechs put out about 62 extra damage after they are elited. It's not a big difference.


Similarly, what builds you play doesn't tend to matter as much as you think they might. For instance, my best performing mech at the moment is a Kitfox (lol, it's doing better than my Kodiaks atm), and you should see some of the potato builds Proton runs in his spare time, and he still tops the charts with those builds, like gauss+AC2+SRM4, and gauss-streaks.

Edited by Tarogato, 02 February 2017 - 07:55 AM.

[NRP]

LOL, should we call him Protato? Kidding (I hope he doesn't read this).

Anyway, quit worrying about stats and just play the game.

[Davegt27]

Quote

Posted Today, 08:50 AM
Playing unleveled mechs, in my experience, doesn't necessarily degrade performance that noticeably. I've been making a habit of tracking some of my mechs as I level them, because I'm a spreadsheet nerd... and some of my mechs performed better without Basics than they performed after I finished Elites. Why? Because other factors have more impact than whether your mech has skills on it or not.


What I actually do is play all of my mechs without leveling them, I just stock pile the XP for 20 matches or so, and then I save my stats. Then I spend all the XP at once, get all the variants Elited in one go, and play 20 more matches again per variant, and compare the before and after. On average, my mechs put out about 62 extra damage after they are elited. It's not a big difference.


Similarly, what builds you play doesn't tend to matter as much as you think they might. For instance, my best performing mech at the moment is a Kitfox (lol, it's doing better than my Kodiaks atm), and you should see some of the potato builds Proton runs in his spare time, and he still tops the charts with those builds, like gauss+AC2+SRM4, and gauss-streaks.

sadly it doesn't take a spread sheet to see all this

don't get me started about FW/FP

[Dimento Graven]

Tarogato, on 02 February 2017 - 07:50 AM, said:

Playing unleveled mechs, in my experience, doesn't necessarily degrade performance that noticeably.

...

Out of curiosity, did you note the same lack of performance bump prior to the great skill nerfing of 2016 (remember PGI seriously reduced the effects of the skills, the skill effects were what... double? what they are now)...

Edited by Dimento Graven, 02 February 2017 - 08:44 AM.

[Tarogato]

Dimento Graven, on 02 February 2017 - 08:44 AM, said:

Out of curiosity, did you note the same lack of performance bump prior to the great skill nerfing of 2016 (remember PGI seriously reduced the effects of the skills, the skill effects were what... double? what they are now)...

When was that change? I can't for the life of me remember the month that happened, and I can't find it in the patch archive either. But I feel like it was before I started my little experiment.

Edited by Tarogato, 02 February 2017 - 09:16 AM.

[Dimento Graven]

Tarogato, on 02 February 2017 - 09:15 AM, said:

When was that change? I can't for the life of me remember the month that happened, and I can't find it in the patch archive either. But I feel like it was before I started my little experiment.

As I recall they started this in like December 2015, and followed up with more later on, but I can't remember for sure, maybe I'll look it up when I have more time.

Pretty sure on the December 2015 patch though...

[Mystere]

Don't care.

[Jman5]

Cadoazreal, on 02 February 2017 - 02:50 AM, said:

What title says. unless its a 1v1 ladder it means nothing coz people level mechs, play fun builds, get unlucky bad teams, get dc's on team, play only meta mechs and never level a mech while others do, have different quality pc's.

That's why you don't look at individual games, but rather their aggregated seasonal stats. Variables like bad teammates, and disconnects average out over the season.

Quote

etc etc etc, the whole system is flawed and most of it reflects how much you play / how little rl commitments you have not how good you are at working your muscle memory.

You can easily disprove this assertion by looking at Season 7 data sorted by "Matches played". Many of the 100 most active players have frankly terrible stats. On the flip side if you search by average matchscore you will see many of those players have a rather modest amount of games played.

This isn't an RPG. You get better by actively trying to improve, not by grinding out a freakish amount of games. You don't have to be in a metamech or an unfun build to do that.

Edited by Jman5, 02 February 2017 - 12:44 PM.

[Dex Spero]

Tarogato, on 02 February 2017 - 07:50 AM, said:

Playing unleveled mechs, in my experience, doesn't necessarily degrade performance that noticeably. I've been making a habit of tracking some of my mechs as I level them, because I'm a spreadsheet nerd... and some of my mechs performed better without Basics than they performed after I finished Elites. Why? Because other factors have more impact than whether your mech has skills on it or not. What I actually do is play all of my mechs without leveling them, I just stock pile the XP for 20 matches or so, and then I save my stats. Then I spend all the XP at once, get all the variants Elited in one go, and play 20 more matches again per variant, and compare the before and after. On average, my mechs put out about 62 extra damage after they are elited. It's not a big difference. Similarly, what builds you play doesn't tend to matter as much as you think they might. For instance, my best performing mech at the moment is a Kitfox (lol, it's doing better than my Kodiaks atm), and you should see some of the potato builds Proton runs in his spare time, and he still tops the charts with those builds, like gauss+AC2+SRM4, and gauss-streaks.

That's an interesting idea. What is your opinion, though, of trying new mechs (rather then trying new builds)? Intuition says that if I've never piloted a Hunchback my performance in games 10-20 will be higher then in games 1-10, but if my intuition was always right, or even right half the time, I'd be a billionaire on a yacht instead of an online gamer. Even if I am right, maybe that's not enough data points to really affect your stats (what's 20 bad games out of 800 really going to change?). I'd also like your opinion regarding piloting a new class of mech (i.e. first time in an assault), or even just a class that you seldom use.

I'm asking you cuz I love the active thread you've got going on the 12-0 stats! Thanks in advance

[Tarogato]

Dex Spero, on 02 February 2017 - 01:07 PM, said:

That's an interesting idea. What is your opinion, though, of trying new mechs (rather then trying new builds)? Intuition says that if I've never piloted a Hunchback my performance in games 10-20 will be higher then in games 1-10, but if my intuition was always right, or even right half the time, I'd be a billionaire on a yacht instead of an online gamer. Even if I am right, maybe that's not enough data points to really affect your stats (what's 20 bad games out of 800 really going to change?). I'd also like your opinion regarding piloting a new class of mech (i.e. first time in an assault), or even just a class that you seldom use.

I'm asking you cuz I love the active thread you've got going on the 12-0 stats! Thanks in advance

My answer would be rather specific to me, so I'm not sure how useful.

I've always found new chassis to be an instant transition. If I'm not comfortable with a new mech in the first 1-2 matches, I probably never will be. For me... it's just simple application of pre-existing knowledge. Variables like hardpoint height and cockpit height are easy to remember, and really the only other variables are speed, jumpjets or not, and how much to torso twist according to hitboxes (for instance, you wouldn't torso twist in a Stalker, that just makes you a bigger target).

Of my first 10 or so matches in a KDK-3, most of them were near or above 1000 damage ezpz. And I almost never play assaults (they only comprise 11% of my games, and a further 30% of those are Gargoyle, which plays more like a medium anyways)

[El Bandito]

Tarogato, on 02 February 2017 - 01:24 PM, said:

(for instance, you wouldn't torso twist in a Stalker, that just makes you a bigger target).

Unless you deadside with the Misery.

[Marius Evander]

Davegt27, on 02 February 2017 - 08:04 AM, said:

Spoiler

sadly it doesn't take a spread sheet to see all this

don't get me started about FW/FP

http://makeagif.com/dtTorx
http://makeagif.com/nSyeY3
Restoslll my master I need help !!!!!!

Edited by Cadoazreal, 02 February 2017 - 05:43 PM.

[RestosIII]

Cadoazreal, on 02 February 2017 - 05:36 PM, said:

http://makeagif.com/dtTorx
http://makeagif.com/nSyeY3
Restoslll my master I need help !!!!!!

Copy the actual gif address by right clicking it, and clicking on "copy image address". Then, put this on your post:

[ img] "copy pasted link here" [/ img]

Without the spaces and " marks obviously.





Now, if I actually knew what you were trying to get across with these gifs (I don't), I'd go and make less terrible FPS versions for you, with decent resolution.

Edited by RestosIII, 02 February 2017 - 05:54 PM.

[Scout Derek]

El Bandito, on 02 February 2017 - 05:31 PM, said:

Unless you deadside with the Misery.

mmmm. deadside!

[Marius Evander]

thankyou, theyre supposed to relate to the spock youtube link above that i spoler tag imbedded where hes going on about believeing the bullets aren't real.

[Snazzy Dragon]

A ranking system doesnt really work in the long run though

The more players of high skill you have on the same team the lower their average match scores will be because they will be positioning better, focusing targets, rotating armor, and doing less damage to kill