Jump to content

Statistical Analysis Of The 12-0


187 replies to this topic

#101 Appogee

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 10,966 posts
  • LocationOn planet Tukayyid, celebrating victory

Posted 31 January 2017 - 11:43 AM

Here's a lateral thought:

Is it possible that "Players who tend to win more will more often be on the winning side of a stomp" because - in the matches that were analysed to arrive at this conclusion - they actually did win more often, which inflated their win/loss stat?

Ie. this is a form of statistical circular reasoning?

Edited by Appogee, 31 January 2017 - 11:44 AM.


#102 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 31 January 2017 - 11:48 AM

View PostAppogee, on 31 January 2017 - 11:43 AM, said:

Here's a lateral thought:

Is it possible that "Players who tend to win more will more often be on the winning side of a stomp" because - in the matches that were analysed to arrive at this conclusion - they actually did win more often, which inflated their win/loss stat?

Ie. this is a form of statistical circular reasoning?
No I think the win/loss ratio is being derived from their total wins and losses, which includes non-stomps as well.

#103 Bloody

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 569 posts

Posted 31 January 2017 - 11:57 AM

given PGI track record on doing things, i am surprised anyone believes that PGI does anything other than minimal viable release.. as expected they took the lazy way of balancing

#104 C E Dwyer

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,274 posts
  • LocationHiding in the periphery, from Bounty Hunters

Posted 31 January 2017 - 12:15 PM

If I'm reading these charts and graphs correctly, what your basically saying is P.G.I 'balance' by weight class and teir, and don't make any effort to balance by skill level within those tiers.

We all know they don't try and match up tonnage by weight class

#105 a gaijin

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,003 posts
  • LocationUS Naval Base, Yokosuka, Japan

Posted 31 January 2017 - 03:18 PM

View PostDimento Graven, on 30 January 2017 - 10:06 PM, said:

Not to side track ANYTHING here, BUT, I just can't help but post something that struck me out right odd...

Horse, y'know what intrigues me most about that PGI job posting you've quoted?

This right here:

Now, if the game had as little cheating, and PGI knew exactly as much as Russ has continued to claim over the years, I find it absolutely odd that experience with cheating/anti-cheating would even be bothered to be listed as any sort of requirement.

Me thinks there's quite a bit of... obfuscation, coming from Russ on the subject.

I'm just glad that it indicates PGI is at least being somewhat proactive on the matter, even if doing it somewhat covertly :)

#106 MadIrish

    Member

  • PipPipPipPipPip
  • The 1 Percent
  • 152 posts

Posted 31 January 2017 - 04:52 PM

View PostTarogato, on 31 January 2017 - 09:31 AM, said:

Unfortunately, premade sizes is one of the biggest factors in the outcome of FP matches. Also, since there is no matchmaker there at all, the skill disparity is even wider. And we already have a bit of a metric for overall Clan vs IS performance in that realm - it's the battle log or whatever it's called, as well as the CW map itself. Which to be honest, I never pay attention to anymore because I don't play CW. Posted Image


Lets see the straight up IS vs Clan Win Loss stat not the sugar coated fuzzy math stats provided in the in-game leaderboard. You all seem to think that losing everything is attributed to premade and the great skills of the Clan pilots Posted Image. That's a pretty arrogant assumption especially when everyone is constantly squawking over team chat about how crazy difficult it is to take out the clan mechs and your IS mechs fall like they are made out of glass.

#107 Carl Vickers

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Covert
  • The Covert
  • 2,649 posts
  • LocationPerth

Posted 31 January 2017 - 04:55 PM

View PostMadIrish, on 31 January 2017 - 04:52 PM, said:


Lets see the straight up IS vs Clan Win Loss stat not the sugar coated fuzzy math stats provided in the in-game leaderboard. You all seem to think that losing everything is attributed to premade and the great skills of the Clan pilots Posted Image. That's a pretty arrogant assumption especially when everyone is constantly squawking over team chat about how crazy difficult it is to take out the clan mechs and your IS mechs fall like they are made out of glass.


When you are grouped with potato's all you will hear is potato comments like your last sentence. Clans = easier mechs to pilot and range advantage makes it better for newer pilots. IS = better pilots as it is hard mode compared to clans.

Edited by Carl Vickers, 31 January 2017 - 04:55 PM.


#108 MacClearly

    Member

  • PipPipPipPipPipPipPip
  • The Butcher
  • The Butcher
  • 908 posts
  • LocationCanada

Posted 31 January 2017 - 04:57 PM

I would love to see an analysis of how it is possible to be in 10-15 straight losses, especially when they are stomps a great deal of the time.

Also am curious if your finding point to one or two elite players that create the avg. score bump or is the average distributed equally over the winning team most of the time?

#109 Carl Vickers

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Covert
  • The Covert
  • 2,649 posts
  • LocationPerth

Posted 31 January 2017 - 04:59 PM

Depends on the match, if everyone on one team is of equal skill then everyone will be doing between 1100 to 1400 damage. Over 1500 and you are starting to get into carry land damage wise.

Then there are some people who are just beasts and will do high damage no matter who else is on the team.

#110 Karl Streiger

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Wrath
  • Wrath
  • 20,369 posts
  • LocationBlack Dot in a Sea of Blue

Posted 01 February 2017 - 12:32 AM

View PostCarl Vickers, on 31 January 2017 - 04:59 PM, said:

Depends on the match, if everyone on one team is of equal skill then everyone will be doing between 1100 to 1400 damage. Over 1500 and you are starting to get into carry land damage wise.

Then there are some people who are just beasts and will do high damage no matter who else is on the team.

I think you are wrong here

damage is not a key performance indicator its highly dynamic because it depends on the performance of 24 guys. And last not least the sum of maximum damage is final.
when i do 20dmg to a target you can't do the same 20damage to this target.

when i have the seldom pleasure to gut a "new" mech from the rear iit s ~60-80dmg and a kill
but this depends on the guy that is killed - if he has a clue he would not be gutted from behind. So you can score more damage.

guys that deal damage no matter who is in their team have a different playstyle - and i hardly believe that you will see them often in the front lines.

#111 Tarogato

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Civil Servant
  • Civil Servant
  • 6,558 posts
  • LocationUSA

Posted 01 February 2017 - 01:46 AM

View PostKarl Streiger, on 01 February 2017 - 12:32 AM, said:

guys that deal damage no matter who is in their team have a different playstyle - and i hardly believe that you will see them often in the front lines.


Some of the best players in the game are notoriously aggressive. They're always the first to the front line, vying for the best position and trying to control the engagement by asserting dominance over the opponent. And they always score high because they know what they are doing. Even despite when they clean headshots and backshots.

That's not to say there aren't players that exist who like to hang back and abuse pug armour while plucking damage away from the safety of the back line... there are certainly those people. But those people are relying on a crutch - opportunistic vultures rather than deliberate and aggressive hunters.

#112 The6thMessenger

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Nova Captain
  • Nova Captain
  • 8,104 posts
  • LocationFrom a distance in an Urbie with a HAG, delivering righteous fury to heretics.

Posted 01 February 2017 - 01:52 AM

View PostTarogato, on 01 February 2017 - 01:46 AM, said:

Some of the best players in the game are notoriously aggressive. They're always the first to the front line, vying for the best position and trying to control the engagement by asserting dominance over the opponent. And they always score high because they know what they are doing. Even despite when they clean headshots and backshots.

That's not to say there aren't players that exist who like to hang back and abuse pug armour while plucking damage away from the safety of the back line... there are certainly those people. But those people are relying on a crutch - opportunistic vultures rather than deliberate and aggressive hunters.


I kind of notice this myself now that im kind of "seasoned". Now as an Urbie, i'm in front of the line -- paltry armor compared to heavier mechs. I flank, i don't hang back. Damn, sometimes i even get killed first for having the initiative to do something when the rest of my team is unwilling.

#113 Snazzy Dragon

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Defiant
  • The Defiant
  • 2,912 posts
  • LocationRUNNING FAST AND TURNING LEFT

Posted 01 February 2017 - 01:55 AM

View PostTarogato, on 01 February 2017 - 01:46 AM, said:

Some of the best players in the game are notoriously aggressive. They're always the first to the front line, vying for the best position and trying to control the engagement by asserting dominance over the opponent. And they always score high because they know what they are doing. Even despite when they clean headshots and backshots.

That's not to say there aren't players that exist who like to hang back and abuse pug armour while plucking damage away from the safety of the back line... there are certainly those people. But those people are relying on a crutch - opportunistic vultures rather than deliberate and aggressive hunters.


If I might add, they're usually also the ones yelling WHERE to push instead of just yelling "PUSH! PUSH! PUSH!"

#114 BluefireMW

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 238 posts

Posted 01 February 2017 - 02:17 AM

0/12 and 12/0 is the best result in this game and shows the quality of the game very precise

#115 Lily from animove

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • The Devoted
  • The Devoted
  • 13,891 posts
  • LocationOn a dropship to Terra

Posted 01 February 2017 - 03:24 AM

I have some massive issues with your new idea of how pilots should be distributed, because your example actually takes an ideal skill distribution as a base for the queue. Sure PGi could EASILY do that but then waiting times may be horrible. The queue of people wating for a game usually does not look even close to the one you took as a base. An idealised example like this is why a implemented idea fails, because the conditions for the idea aren't met ingame very often, especially not when people don't wanna wait serveral minutes without starting to complain.

what you proposed is probably what PGI implemented initially, but simply fails by the reality of available players and weight class distribution.

Edited by Lily from animove, 01 February 2017 - 03:25 AM.


#116 Karl Streiger

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Wrath
  • Wrath
  • 20,369 posts
  • LocationBlack Dot in a Sea of Blue

Posted 01 February 2017 - 03:50 AM

View PostLily from animove, on 01 February 2017 - 03:24 AM, said:

I have some massive issues with your new idea of how pilots should be distributed, because your example actually takes an ideal skill distribution as a base for the queue. Sure PGi could EASILY do that but then waiting times may be horrible. The queue of people wating for a game usually does not look even close to the one you took as a base. An idealised example like this is why a implemented idea fails, because the conditions for the idea aren't met ingame very often, especially not when people don't wanna wait serveral minutes without starting to complain.

what you proposed is probably what PGI implemented initially, but simply fails by the reality of available players and weight class distribution.

Well you could add some more functionality / checkbox options to the "Play" button.
the more you activate the more precise the MM might work - maybe there could also be a simple traffic light that consider the number of players currently online - for each criteria there should be a minimum number of players needed to work properly

Or you keep this system full automated and add a message how the bracket and seed is before the MM even start working.
For example when there are only 100 players online - the MM will tell your right from the start that he will not have any criteria.

It could also help when PGI finally realize that player numbers as well as player peak numbers is nothing they should keep hiding.
That there are issues with "players" might be the reason why there are daily events finally

#117 Lily from animove

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • The Devoted
  • The Devoted
  • 13,891 posts
  • LocationOn a dropship to Terra

Posted 01 February 2017 - 04:53 AM

View PostKarl Streiger, on 01 February 2017 - 03:50 AM, said:

Well you could add some more functionality / checkbox options to the "Play" button.
the more you activate the more precise the MM might work - maybe there could also be a simple traffic light that consider the number of players currently online - for each criteria there should be a minimum number of players needed to work properly

Or you keep this system full automated and add a message how the bracket and seed is before the MM even start working.
For example when there are only 100 players online - the MM will tell your right from the start that he will not have any criteria.

It could also help when PGI finally realize that player numbers as well as player peak numbers is nothing they should keep hiding.
That there are issues with "players" might be the reason why there are daily events finally


nice would also be "dropdeck" in which i cna put serveral mechs, and then the Mm would choose any of those to drop me in. I wouldn't mind dropping in a MDD NVA or ADR in QP. so the MM could choose the "missing" mech it may require. of course if someone doesn't wants that he doesn't have to use that "dropdeck".

#118 Verkhne

    Member

  • PipPipPipPipPipPip
  • Mercenary Rank 2
  • 299 posts

Posted 01 February 2017 - 06:03 AM

are there pilots who are statistical outliers in these 12-0,0-12 matches? Are there factors where pilots with good or bad match score averages obtain unexpected results from the MM? (ie Taragato was in more 12-0 matches than expected..why?, is there a corresponding player who was in more 0-12 matches than expected?) this sort of thing would then crank your average matchscore kills and damage up or down.

#119 Khereg

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 919 posts
  • LocationDenver, CO

Posted 01 February 2017 - 06:07 AM

View PostDimento Graven, on 31 January 2017 - 11:48 AM, said:

No I think the win/loss ratio is being derived from their total wins and losses, which includes non-stomps as well.



Right. Stomps are actually pretty rare. I was one of the folks submitting matches to Taro and I think over the course of two months I submitted maybe 5 - 10. I play a good bit (averaging about 10 matches a calendar day, most of which is solo queue), but I probably only submitted about half the stomps I was involved in.

Crunching that out, let's say I was involved in 20 stomps (on either side) out of about 600 matches (3.3% of matches played). Not nearly enough to move the needle on W/L ratio for the players involved. That's the conservative estimate, mind you.

Edited by Khereg, 01 February 2017 - 06:08 AM.


#120 xTrident

    Member

  • PipPipPipPipPipPipPip
  • Bad Company
  • 655 posts
  • LocationWork or Home

Posted 01 February 2017 - 06:21 AM

View PostKarl Streiger, on 01 February 2017 - 12:32 AM, said:

I think you are wrong here

damage is not a key performance indicator its highly dynamic because it depends on the performance of 24 guys. And last not least the sum of maximum damage is final.
when i do 20dmg to a target you can't do the same 20damage to this target.

when i have the seldom pleasure to gut a "new" mech from the rear iit s ~60-80dmg and a kill
but this depends on the guy that is killed - if he has a clue he would not be gutted from behind. So you can score more damage.

guys that deal damage no matter who is in their team have a different playstyle - and i hardly believe that you will see them often in the front lines.


My play style, say... From a year ago or so has drastically changed. I used to be one of those players that wouldn't be on the front lines but more of a support type. Now I'm a lot more apt to be in the front helping with the push, or if the team isn't pushing I now have no problem peeking out and trying to do my part in doing damage/taking it. I definitely feel like l'm on the front lines taking my fair of damage with the others. That being said, a lot of the stomps I see in which five or seven players (not kidding) don't even break 100 damage I'm still in the 250 damage area. And yes, I was right up there taking shots and doing damage as well. This has been a normal thing whenever that many players on my team don't break 100 points worth of damage and we get steam rolled.

Point is I disagree to a point. Some players can indeed, at least compared to the rest of the team, do decent damage even through a stomp and majority of your teammates do nothing. But frankly I don't believe it has as much to do with being on the front-lines as it does playing poorly. Because in no way do I believe it to be a good tactic to be up front and exposed for all the enemy to see, which is a lot of what I see out there. I'm up front quite often anymore but I use cover all the time. It helps... A lot.

Edited by xTrident, 01 February 2017 - 06:25 AM.






2 user(s) are reading this topic

0 members, 2 guests, 0 anonymous users