57 replies to this topic

[Tarogato]

----> Link to Survey (only 7 quick questions!)


I've been mulling over posting this survey for a while, unsure that it would be worth the effort, but when Trevelyas made a post recently, I decided "screw it, why not, let's do this."


----> Link to Results

---

The results graphs and charts update (albeit laggardly) in real time as responses come in, so it will start out fairly empty. Check back after a day or so when everything is better filled out. I'll be continually cleaning up and tuning the formatting, as well as adding more charts, so if there are any additional comparisons that you'd like to see, such as "how many competitive respondents said they play Clan mechs!?", I could prioritise adding those first.


Individual player stats are automatically pulled from the Quickplay Leaderboards, and are actually the combination of all stats from Season 1 - 6. This is contingent upon people entering their in-game name correctly and also indicating if they had any name changes, so please do that properly - your response will automatically be culled if your name isn't valid. And don't worry, your super-duper secret name(s) will not show up publicly, I have them locked away in a private sheet that only I have access to.


(last time I did something like this was Sept. 2015)

Edited by Tarogato, 15 January 2017 - 11:31 AM.

[Siegegun]

Out of curiosity, won't someone playing in their non optimal mechs skew your match numbers? I know before the skill tree announcement I used to master all the mechs and their variants as a personal goal. I stopped doing that with the skill tree announcement and started playing the mechs I was decent in regularly again. Theres a huge difference in numbers between me attempting to play lights and assaults and me playing the classes and mechs I'm actually decent with (I tend to play mediums). I am sure there are others that have done something similar. I.E.- Me derping around and playing pokemech makes my average match score (and other stats) much lower than if I had not done that, the current "season" shows that if you compare it to the earlier ones.

Just curious. Or perhaps I am not understanding the reason your collecting the info?

[Tarogato]

Siegegun, on 15 January 2017 - 12:16 PM, said:

Out of curiosity, won't someone playing in their non optimal mechs skew your match numbers?

Nope. All that matters is how they actually performed (ie, their stats) versus what Tier they are in. My hypothesis is that Tier 1 and 2 are littered with players that have below-average performance, and they got there because the PSR system is positively biased.

[Bud Crue]

Tarogato, on 15 January 2017 - 12:25 PM, said:

Nope. All that matters is how they actually performed (ie, their stats) versus what Tier they are in. My hypothesis is that Tier 1 and 2 are littered with players that have below-average performance, and they got there because the PSR system is positively biased.

Tarogato, are you doing something like this on Reddit too? Seems to me the limited nature and number of the "official forum" user pool limits the usefulness of a survey like this. Just thinking that reddit or other sites (when combined) might get you a broader pool.

[Tarogato]

Bud Crue, on 15 January 2017 - 12:30 PM, said:

Reddit too?

https://www.reddit.c...hmaking_survey/

[Siegegun]

Tarogato, on 15 January 2017 - 12:25 PM, said:

Nope. All that matters is how they actually performed (ie, their stats) versus what Tier they are in. My hypothesis is that Tier 1 and 2 are littered with players that have below-average performance, and they got there because the PSR system is positively biased.

Oh, ok I see. Well considering I got to tier 2 (almost tier one) playing for the most part with mechs and chasis that I am terrible in and some of which are just bad robots, plus my personal experience, I think you are right. Makes sense now. It helps to explain Rambo Gyr that went off by himself in a skirmish match a couple days ago. I might not be supercomp level but I know enough not to be doing that kind of thing.

"Oh Rambo Gyr, where are you going? The deathball is the other direction!"

[El Bandito]

Finished the survey. Let's see if it indeed will shed light on how diluted T1 has truly become.

[Oberost]

Good job. Survey completed.

[Snazzy Dragon]

I feel like if tiers actually worked I would be a T2 pilot and most of my teammates would be T4. But funnily enough, we're all T1. Hrm.

[Mcgral18]

That's an overwhelming number of Tier 1...but might also be linked to the whole EXP bar as well


New players aren't the kind to check out forums or Reddit, so it will be primarily long term players, and if you play, you'll go up in Tier

[Tarogato]

Mcgral18, on 15 January 2017 - 03:01 PM, said:

That's an overwhelming number of Tier 1...but might also be linked to the whole EXP bar as well


New players aren't the kind to check out forums or Reddit, so it will be primarily long term players, and if you play, you'll go up in Tier

Yeah, the distribution of tiers in the sample is going to be off, but that's expected.

If somebody wants to help by posting this to Steam for me (I don't use that), maybe that would help?

[Bud Crue]

Tarogato, on 15 January 2017 - 03:07 PM, said:

Yeah, the distribution of tiers in the sample is going to be off, but that's expected.

If somebody wants to help by posting this to Steam for me (I don't use that), maybe that would help?

I don't know that it is off. I play most evenings as part of a 6-12 man of casual, mixed tier players. Most of us know the game and the better known units. I would say 80% of the matches we play there is at least a 4-6 man of known tier 1 and/or competitive players on the opposing team or joining ours. Point being: tier 1 players seem far more common in group queue at least. Is that an indicator as to the state of the game? Beats me, but a year ago it sure felt like there was a lot more variety of teams and now it is a much more limited rotation of: EMP, AS, ISEN, AWOL, MS, MJ12, etc.

[Deathlike]

Bud Crue, on 15 January 2017 - 03:41 PM, said:

I don't know that it is off. I play most evenings as part of a 6-12 man of casual, mixed tier players. Most of us know the game and the better known units. I would say 80% of the matches we play there is at least a 4-6 man of known tier 1 and/or competitive players on the opposing team or joining ours. Point being: tier 1 players seem far more common in group queue at least. Is that an indicator as to the state of the game? Beats me, but a year ago it sure felt like there was a lot more variety of teams and now it is a much more limited rotation of: EMP, AS, ISEN, AWOL, MS, MJ12, etc.

LARGE AND DIVERSE PLAYERBASE

[MacClearly]

What about my alt account? Should I put that in and let you know what it is so you can link it for the info to be more valuable or is that maybe excessive?

[NighthawK1337]

Good job, really good. Well Done.
At least the segregated match scores graph shows a significant skew to the right (high match score) as Tier increases. So we can conclude that increasing in Tier does correlate to having better average match scores.
However I do have concerns about the sample size. Not everybody goes to the forums, people who do are usually the ones passionate about the game, and there's a lot of T1 players.
Could you add another set of graphs segregating match numbers instead? like W/L for people with more than 1000 matches, between 1000 and 500, less than 500, etc. It could show also how people learn with respect to number of matches, it could add another insight how good matchmaking is.

Edited by NighthawK1337, 15 January 2017 - 04:26 PM.

[Kamikaze Viking]

Tarogato, on 15 January 2017 - 03:07 PM, said:

If somebody wants to help by posting this to Steam for me (I don't use that), maybe that would help?

Done - http://steamcommunit...13524087027152/

[Bombast]

Interesting so far. I put in that I felt like I was worse then my tier and the people I play with, but apparently I'm smack dab at average for T2, so I do belong there?

Or maybe things will change when the poll isn't so lopsided towards Tier 1.

Neat poll. Thanks.

[The Lost Boy]

Tier only reflects the ability to cause decent damage in a match more times than you dont. If you can do that top tier 1 is doable. I am living proof. My win/loss is negative, just barely even k/d. I play lots of games, often in an assault, and can get 500 points and higher damage regularly. If you can do that it will climb.

[Kasumi Sumika]

Thanks for making this. I Completed the survey.

[Tarogato]

NighthawK1337, on 15 January 2017 - 04:20 PM, said:

Could you add another set of graphs segregating match numbers instead? like W/L for people with more than 1000 matches, between 1000 and 500, less than 500, etc. It could show also how people learn with respect to number of matches, it could add another insight how good matchmaking is.

Except that assumes that all players started playing the game in June 2016. They may well be veteran high-division competitive players who just don't play as actively as anymore.

I could create such a graph though, from the global player base stats - no poll/survey necessary. I'm just lazy, so ... I'll do it tomorrow. Bug me if I don't have it here in the next 30 hours.