Jump to content


winrate is not a reflection of player skill


  • Please log in to reply
801 replies to this topic

Valkyr #201 Posted Apr 18 2019 - 22:34

    Sergeant

  • Players
  • 69479 battles
  • 232
  • Member since:
    04-18-2011

View Postcheapbooks, on Apr 18 2019 - 03:54, said:

A bunch of nerds did an official study that proves that people playing games with matchmakers like World of Tanks should revert to a 50% winrate. The reason this does not happen in WOT is because some players are stat padding.

Then explain why I, who has researched and fought in almost all the tanks (only 10 left to go) on the tech tree, good and bad, have a 54+% win rate. Some tanks are bad, some are good...after 65,000 battles if your math was correct I would be down to 50% by now.  That is NOT the case. Your argument is logically flawed.



dunniteowl #202 Posted Apr 18 2019 - 22:43

    Major

  • -Players-
  • 31387 battles
  • 8,080
  • Member since:
    09-01-2014

View PostHansReichardt, on Apr 18 2019 - 15:09, said:

 

the guy that doesn't have the stones to play a tank over tier 6 is going to say I am making excuses? 

 

your 50% doesn't mean anything bro, since you stat-padded to that level and simply play to maintain it. I never said I was good (still earning every day), but I am quite a bit better than you.

 

all I am saying is that client-side mods should be removed, and then we can talk about stats. That isn't an excuse, it is an assertion. 

 

Oh, yes, Boy, I don't play because I'm afraid to go higher.  Yes boy, you got me.

 

Maybe you should know this, even though I've posted this many times before.

 

 

I have a laptop made in 2010, meaning it's OLDER than WoT is.  It has 6G RAM which it shares with the video chip.  I have a wireless rural internet with high ping most days.  Since 1.0 dropped, I get around 11 fps.

 

So, tell me:  Would you consider going higher up tier with such a fancy rig as mine?  Would you purposely gimp yourself and your team in those higher tiers with this great machine?

 

Now that you know, maybe you can stop making excuses and then hurling insults.  If my ping was steadily in the, say, 80 -- 120 ms, I would totally go up tier.

 

Even so, playing mid tiers is not a crime, nor is it a reason to belittle someone's 'courage' in a video game.

 

So my takeaway from your response is that you will resort to just being purely insulting and belittling as a way to "win" your argument.  Check. 

 

Then you call me a stat padder.  I challenge you to play with the rig I have and do as well.  Go ahead.  You can find something comparable at pretty much any garage sale or Goodwill.  When you do that, then you can talk smack, m'kay pumpkin?  My stats I own and earned.  With a crap machine and crap connection.

 

I didn't go higher than this because that's where I was when 1.0 dropped and I have 26 empty garage slots just waiting for more tier VIs and VIIs.  I also have not spent ONE PENNY on this game and have been a FTPP the whole time, so I didn't get to just drop cash on the game to fail my way up tier faster like many who come here and complain about rigged MM, stat padders and the like.

 

You talk like the cheating and hacks are so prolific that you couldn't possibly swim in these waters and be okay.

 

So, let's compare.  What kind of rig do you have?  I'll settle for the basics: Processor, age of computer, laptop or desktop, video chip or dedicated card, RAM, screen resolution, etc.

 

I play on lowest possible settings, as I said, at around 9-11 fps with a ping running at around 175 - 250 or so before the lag and packet loss make it unplayable at my slideshow speeds.

 

Even so, somehow, I have managed to do what you do not -- pass the 50% mark.  Now mind you, I deal with all the same issues you do, real or imagined.  If you think those things are going on, then I must be swimming in the same waters, right?

 

So, yes.  You are making excuses.  And when those fail you, you resort to insulting the folks who point it out.

 

 

Good Job.  Keep up the Good Works.  Maybe someday you'll make it past that bar.

 

 

GL, HF & HSYBF!
OvO



_Tsavo_ #203 Posted Apr 18 2019 - 22:44

    Major

  • Beta Testers
  • 46222 battles
  • 19,710
  • [BRVE] BRVE
  • Member since:
    02-16-2011
I love you, Dunnite. 

NeatoMan #204 Posted Apr 18 2019 - 23:17

    Major

  • Players
  • 28180 battles
  • 20,556
  • Member since:
    06-28-2011

View PostHansReichardt, on Apr 18 2019 - 16:09, said:

the guy that doesn't have the stones to play a tank over tier 6 is going to say I am making excuses? 

 

your 50% doesn't mean anything bro, since you stat-padded to that level and simply play to maintain it. I never said I was good (still earning every day), but I am quite a bit better than you.

 

all I am saying is that client-side mods should be removed, and then we can talk about stats. That isn't an excuse, it is an assertion. 

 

hmmm....  but if you play higher tier tanks badly, is that indicative of "having stones" or a sign of stupidity?

grandthefttankV #205 Posted Apr 18 2019 - 23:49

    Major

  • Players
  • 35830 battles
  • 2,610
  • [_EOS_] _EOS_
  • Member since:
    09-20-2013

View PostValkyr, on Apr 18 2019 - 21:30, said:

 

Saying Win Rate doesn't matter is like saying the winners of the pennant didn't deserve it.

 

Completely different from the bench team claiming credit



Flarvin #206 Posted Apr 18 2019 - 23:55

    Major

  • Players
  • 54584 battles
  • 16,893
  • Member since:
    03-29-2013

View PostgrandthefttankV, on Apr 18 2019 - 17:49, said:

 

Completely different from the bench team claiming credit

 

https://www.offthebenchbaseball.com/2018/03/21/off-the-rankings-ranking-the-best-utility-men/

“THE TREND

One thing this exercise illuminated is that good teams tend to collect the best utility guys. Maybe it’s just that these teams can afford the luxuries of life, such as a quality 25th man. Or maybe the teams are good because of the quality glue guy filling in as needed. Rather than worry about causation, let’s take a quick look at the correlation. Below is the most recent power rankings trended against the points that each team’s utility guy earned in our process.”



churchill50 #207 Posted Apr 19 2019 - 00:02

    Major

  • -Players-
  • 20689 battles
  • 2,145
  • [HIME] HIME
  • Member since:
    05-17-2014

Holy [edited], the amount of ignorance in this OP is absolutely mind-boggling.

 

OP, you might try growing a brain, or, alternatively, follow the old adage "Better to keep your mouth shut and have people think you're an idiot, than to open it and remove all doubt."



Tolos #208 Posted Apr 19 2019 - 00:41

    Major

  • Beta Testers
  • 25997 battles
  • 18,536
  • [GDF] GDF
  • Member since:
    09-22-2010

View PostHansReichardt, on Apr 18 2019 - 20:57, said:

 

I'm not making any excuses for anything, and I do respect some players I know are clean. 

 

What are you afraid of? 

 

Interesting. So because I call you out on excuses, you instantly imply I am also cheating.

 

Tell you what, let's give both our python logs to WG, and see what THEY say, as I'm willing to bet you will be closer to using illegal mods than I am.



Boghie #209 Posted Apr 19 2019 - 01:19

    Captain

  • -Players-
  • 8501 battles
  • 1,148
  • [WCTNT] WCTNT
  • Member since:
    04-10-2016

View PostHansReichardt, on Apr 18 2019 - 15:36, said:

All the fancy calculations above do not take into factor key elements that completely alter both win rate and win8

 

1. Warpack usage (or any other illegal mods--I'm sure there are a bunch out there. A google search pulls up like 3-4 sites that within a few clicks will get your account modded)

2. Percentage of players who are platooning, and how their win rates change vs. solo play. 

 

As long as client-side mods exist, all statistics are suspect, and no one's win rate means a thing. It is like conducting an Olympic 1500 meter race where 30% of the runners are juiced up on HGH, steroids, and blood-doping, and then trying to claim that the final times are somehow valid, or reflect real skill. 

 

I have personally encountered dozens of players in this game who openly admit to using illegal mods. Some old clan mates (none in my current), guys I legionairred with, etc. I know players that have been banned for using the broken objects mod, red ball, aimbot and other stuff. One guy that had an insane 80% overall win rate, huge damage per game, etc. was apparently banned, because after a lot of user complaints, he vanished from the game. I saw this same guy involved in collusion in a couple game.

 

I find that players who have unusually high hit ratios in tanks with poor gun handling are highly suspect. Not proof of cheating, but possible aimbot for sure (and yes, I know how that mod works--it greatly improves your hit ratio and penetration chance)

 

so we can argue all day about win rate, win8, etc., when at the end of the day, it is all meaningless. No one takes Florence Griffith Joyner's 100 meter record seriously, because the whole sport was riddled with doping and steroids when she set it. 

 

When WG removes ALL client-side mods, I will start respecting player statistics. 

 

 

 

Wow Hans,

 

You do KNOW that some of these folks with da' big ol' WR run live on Twitch fairly frequently.  You can actually see which illegal mods - or, maybe even WarPack itself - they are using.  It is right there on the screen - live and in color.  I hate to let you in on this, but they largely play vanilla.  Maybe a mark display mod, maybe a carousel mod, maybe XVM color coding on a lark (they are actually good enough they really don't care who they are competing against).  Zeven got a huge advantage by having color coded (SPG, TD, LT, MT, HT) tank icons.  QB is the only one I have watched that really has all the stats - and, my money is that he is doing that because it entertains the viewers.  I (ME) like to know when they are competing against the best.  Most just play vanilla.

 

You can even watch as some of the good players compete quite well even with crap crews.  Yup, they don't gold retrain to some Tier V or VI crap tank on 'chat - pick a tank day'.  They just cobble up a crap crew and use the skill between the ears to their advantage.

 

What folks who are having troubles in this game have to do is to work on one thing at a time.  And, then realize that you might max out at 51% - and that is ok.

 

One other thing to think about.  If you are odd enough to go against the meta positioning and you do not do well then expect some aggro.  Some dumb [edited]49%er like me taking a top tier HT into the Valley of Death should get full aggro if I stall and die in place in that dumb location.  I should get Battle Chat slammed, and maybe even garage chat slammed.  And, if I am dumb enough to not know why I got the aggro then perhaps that is the one thing I should work on.



I_QQ_4_U #210 Posted Apr 19 2019 - 05:05

    Major

  • -Players-
  • 22892 battles
  • 7,026
  • Member since:
    10-17-2016

View Postcheapbooks, on Apr 18 2019 - 15:23, said:

You have a higher winrate because you are a better skilled player than the players on the other team.

 

Therefor, you are getting preferential matchmaking. WOT is only matching you with other players who have less skill than you, and the matchmaking is rigged. That is the only way to explain your higher skill-based win rate.

 

Is there some contest for dumbest poster of the day or something? What's the code?

Tolos #211 Posted Apr 21 2019 - 08:44

    Major

  • Beta Testers
  • 25997 battles
  • 18,536
  • [GDF] GDF
  • Member since:
    09-22-2010
Isn't it interesting that EVERYTIME I try to get them to make a team of low WR player to take on a team of high WR players, suddenly, they go silent...

mackinawjim45 #212 Posted Apr 21 2019 - 14:59

    Staff sergeant

  • Players
  • 43553 battles
  • 429
  • Member since:
    11-04-2011
This game has  a lot of stats, none more important to me than WR. I'm pathetic...

Canadian_Mano #213 Posted Apr 21 2019 - 15:00

    First lieutenant

  • -Players-
  • 35992 battles
  • 513
  • Member since:
    01-29-2012

View PostTolos, on Apr 21 2019 - 03:44, said:

Isn't it interesting that EVERYTIME I try to get them to make a team of low WR player to take on a team of high WR players, suddenly, they go silent...

 

Almost as if you can't really argue with math

Firemoth #214 Posted Apr 21 2019 - 15:45

    Major

  • Players
  • 38080 battles
  • 4,392
  • Member since:
    05-21-2011

View PostCanadian_Mano, on Apr 22 2019 - 00:00, said:

 

Almost as if you can't really argue with math

 

reason and logic have no place in this thread

Exiledcrow #215 Posted Apr 21 2019 - 17:31

    Major

  • Players
  • 50111 battles
  • 2,506
  • [GURUS] GURUS
  • Member since:
    11-26-2011

View PostFiremoth, on Apr 21 2019 - 15:45, said:

 

reason and logic have no place in this thread

 

Reason and logic have no place in the weak minds of bad players who constantly need excuses to blame their failures on lest they start coming to grips with reality and trying to improve. It's like Dunning-Krueger is using this playerbase as a experiment sometimes ...

toesave #216 Posted Apr 21 2019 - 17:43

    First lieutenant

  • Players
  • 23952 battles
  • 575
  • [DIXIE] DIXIE
  • Member since:
    01-25-2012

you are wrong this game is all about the most useless of stats in a game like this... win/loss is all the matters...  and it is driving players away in tractor trailer trucks full of them...

 

it is great to be in on the ground floor to watch this game fade away to there  use to be a game called....



Rocketmandb #217 Posted Apr 21 2019 - 19:45

    Major

  • Players
  • 53779 battles
  • 2,140
  • [STRNG] STRNG
  • Member since:
    12-29-2011

The amount of failed logic in this thread is mind boggling.

 

Let's talk some basics:

 

Is a higher WR player always better than a player with a lower WR?

No, absolutely not. There are all sorts of variables that enter into WR that can be controlled, or at least influenced, by the players. How much you platoon with good players, what tiers and tanks you play, etc., all influence win rate. 

 

Does that mean that WR is not at all influenced by skill?

If you believe this, you're a [insert intelligence-related insult here]. Look at the extremes. I can guarantee to you that a 70% WR player who plays primarily higher tiers is better than a 40% WR player who primarily plays in the kiddie pool. Can I guarantee that a 53% player is better than a 52% player when both have similar tanks/battles in their vehicle lists? Nope - no way. WR is a " general direction indicator" of skill, but not a definitive indicator.

 

Is trying to draw a parallel with pro sports teams a valid comparison?

I see this all the time, but in most arguments made here, no - not at all. How can that possibly be? Are the NY Giants better with Saquon Barkley or without? Clearly better with him. He is one of the most dynamic backs in the game right now. But the Giants had a crappy win rate last year, so if WR indicated skill, the Giants would have won! See? WR DOES NOT EQUAL SKILL!!! ARGUMENT WON!!! Wrong, wrong, and wrong again. Apples meet oranges. If the NFL took every player, threw them into a big mixing bowl and randomly distributed them to teams every week this would be a valid comparison - and you would see teams with RBs like Barkley, Zeke, and Gurley tending to do better than teams that have Frank Gore as their lead runner. But, again, if you take a bad team (real world NFL) with Frank Gore as the lead back and drop in Zeke Elliott to take his place, that team will definitively do better. Why? Because Zeke's skill opens up plays and opportunities that Gore's 35 year old legs could never hope to do. The exact same thing applies to tanks. Better players open up more opportunities to win more often than crappy players, and thus tend to win more as a result. It ain't f#$&ing rocket science here... and I know rocket science.

 

The Bottom Line:

There are games you play in that, no matter how well you do individually or with your platoon, you will lose because the rest of your team is just that bad, and/or the opposing team is just that good. There are also games that, no matter how poorly you play individually or with your platoon, you will win because the rest of your team is just that good, and/or the opposing team is just that bad. Then there are a whole lot of games in the middle where things like your skill in a particular tank, relative tier, platoon mates, gold usage, and a host of other things will determine whether you win or lose - and these games determine your win rate. If you argue that something like gold is a determining factor of this, but skill is not, then... well... [insert intelligence-related insult here].


Edited by Rocketmandb, Apr 21 2019 - 20:23.


Viruzzz #218 Posted Apr 21 2019 - 21:24

    First lieutenant

  • Players
  • 16792 battles
  • 900
  • Member since:
    09-26-2012

View Postcheapbooks, on Apr 18 2019 - 09:54, said:

winrate is absolutely positively not a measure of player skill:

 

"Individual performance in team-based online games"

"The longer the users play, the more the performance related to their teams reverts to the mean—which is approximately 0.5"

https://royalsocietypublishing.org/doi/pdf/10.1098/rsos.180329

 

A bunch of nerds did an official study that proves that people playing games with matchmakers like World of Tanks should revert to a 50% winrate. The reason this does not happen in WOT is because some players are stat padding.

 

Did you even read what you are referencing?

 

This paper is measuring League of Legends performance. That game uses an ELO system for matchmaking, world of tanks does not use an ELO system. An ELO system will indeed converge to ~50%, but since that system is not in use in world of tanks the paper's findings isn't relevant to world of tanks.



SquishySupreme #219 Posted Apr 21 2019 - 21:54

    First lieutenant

  • -Players-
  • 3657 battles
  • 713
  • Member since:
    10-29-2018

View PostFiremoth, on Apr 21 2019 - 15:45, said:

 

reason and logic have no place in this thread

 

Neither does statistics, apparently.  But I suspect few people here have actually taken Statistics at a college level.

No, really.  WR is missing two critical pieces of data that keep it from being usable for any sort of statistical analysis.

1: Group performance is in no way indicative of personal performance.  It is impossible to extrapolate data on what the individual players did or did not do from a single aggregate score.
eg: You take a group of dice.  2D4,(new players)  2D6,(better), 3D8 (normal), 2D10, and 1D12(elite player).  10 dice of varying values. Roll them together.   That single value cannot tell what individual die rolled what value.  Those two D4s could have rolled a 3 and a 4, that D10 could have rolled a 2 or a 10.     People also mistakenly think that the previous roll or even 100 rolls will tell you what will happen on any future single roll of the die.  

Which brings us to point #2
2:A single death per match/a single set of values added together also doesn't even generate a bell curve.  The issue is that #1 might work if it was the same people over multiple matches and enough data could be collected on that team.  But the MM makes that all random. When you take 15 random players of random skill levels and toss them in a room together, then have them all roll their dice ONCE, you in no way have a means to do any meaningful long-term analysis.  There simply is not enough data to make the math work.  Giving everyone a skill rating on what the group rolled as a total is invalid unless the exact same group of players is used several times and multiple sets of data are generated to run a comparison on. 

So the WR might be useful in clan wars or something with fixed teams, but for random battles it's useless.   

This is why all other games from the beginning of multiplayer gaming have focused on personal kills and accuracy.  That can be tracked and is a valid way to determine individual skill over time.

Edited by SquishySupreme, Apr 21 2019 - 21:57.


Exiledcrow #220 Posted Apr 21 2019 - 22:12

    Major

  • Players
  • 50111 battles
  • 2,506
  • [GURUS] GURUS
  • Member since:
    11-26-2011

View PostSquishySupreme, on Apr 21 2019 - 21:54, said:

 


This is why all other games from the beginning of multiplayer gaming have focused on personal kills and accuracy.  That can be tracked and is a valid way to determine individual skill over time.

 

Low w/r players perform poorly on both those measures too, hence the low w/r. Kills are not particularly valid though, not compared to damage done vs damage taken. Accuracy isn't all that useful a yardstick either. If I'm driving a Matilda (which I've NEVER run out of ammo with), I can, and will, blow through an entire village of houses to kill a hiding tank. Effective use of my ammo, but doesn't do much for accuracy.


Edited by Exiledcrow, Apr 21 2019 - 22:13.





1 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users