Jump to content


Why Try? Visualization.


  • Please log in to reply
41 replies to this topic

LesterQuaestor #1 Posted Feb 22 2016 - 23:36

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

 

  We often see people complaining about poor MM, lack of balance, bad teams, and how it is useless anymore to play in Lights, Mediums, Heavies, TDs, or arty. Many people say you should uninstall XVM because there is no point to trying if you have less than 50% chance to win.

 

  But what is it that differentiates a Very Bad player from a Great Player? According to the WoTLabs definitions, a VBP has a win rate below 46%. A GP has one of 56% or higher. What does this look like, and what does it mean?

 

The figure below is a graphical representation of the average 100 games by the average player. As you can see they have 49 wins, 50 losses and one draw (which counts as a loss). This is the overall performance of all players combined, on the NA servers. It is also the performance of a hypothetical Bone Nuts Average player.

 

 

If you are a Great Player, your 100 games look like this:

 

Through your own personal effort and skill, you convert 7 games from losses to wins. That's all it takes to be a great player. You need to have a decisive effect on only 7 games out of every 100. Needless to say, this is not a very large effect.

 

 

If you are a really bad player your hundred games looks like this:

 

Here you have an effect on only 3 battles per one hundred, but it is in a negative way. This is harder to do, hence the lower number. It means that but for your lack of skill or inaction, a game that would have been won was lost.

 

The bottom line message is that the difference between the best players and the worst players is not very large. In many battles, even most battles, you do not have a decisive role. But your win rate depends solely on those games where you have the opportunity to make a difference. Your results are determined by those few games where your personal action can convert a loss into a win. Or else the other way round.

 

You must try your best all the time, because you can never know when you might be able to convert a game. If you give up, suicide, or otherwise let your team down, you will be creating losses. Not always, of course, but inevitably.

 

TLDR: Always Try.

 

 

 



Hurk #2 Posted Feb 23 2016 - 00:02

    Major

  • Players
  • 59979 battles
  • 17,409
  • [KGR] KGR
  • Member since:
    09-30-2012

your chart makes linear what is not. 

getting a 60+ win rate isnt simply doing better 11 battles out of 100. 

being able to influence the outcome of 11% of your battles, when you are only 3.33% of the total input, is doing the average work of over 2 players.



65Camino #3 Posted Feb 23 2016 - 00:02

    Captain

  • Players
  • 42847 battles
  • 1,299
  • [2CMBG] 2CMBG
  • Member since:
    09-18-2013
+1 Lester, nice little summary, but a huge challenge to convert those 7 deciding/all-important/the whole she-bang games. :amazed:

LesterQuaestor #4 Posted Feb 23 2016 - 00:10

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

Block Quote

 your chart makes linear what is not. 

 

I suspect that the skill needed to excel is not linear. But the _results_ are linear. They must be considering that W+L=100.

 

The bigger point is that since you don't know when that critical game will happen, you must try all the time.

 

Do you stop playing when your team is down? Well, that might be the 1 game out of the next 30 where you can pull back a victory.

 

Suicide because you are low tier? Well, you may have just created your 4th or 5th unnecessary loss out of the current 100 battles.

 



LesterQuaestor #5 Posted Feb 23 2016 - 00:13

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

Block Quote

 but a huge challenge to convert those 7 deciding/all-important/the whole she-bang games.

 

  One key point, that is impossible to bring out, is that we cannot know when those critical moments occur. Getting the last 5 kills in a blowout is not what decided the battle. At the exact moment when a balance tips it still appears to be balanced.

 



Qinu #6 Posted Feb 23 2016 - 00:23

    Sergeant

  • Players
  • 22643 battles
  • 245
  • Member since:
    02-28-2014
I agree with what you say, but sometimes, within any 100 games, RNGesus could say "Not today!", making one look like a bad player

TheLoveHitman #7 Posted Feb 23 2016 - 00:28

    First lieutenant

  • Players
  • 13218 battles
  • 812
  • [CRAKD] CRAKD
  • Member since:
    06-04-2012
A GREAT players will influence all 100% of their battles, they will only lose the battles where their team was just uncarriable. I'd say 30% of games you'll probably win without doing much, 30% of games you'll lose without doing much. The other 40% comes down to how much you can do. That's why you don't see many people above 70% winrate and below 30% winrate. (If you manage below 30% winrate you need to uninstall, this is not a place for you)

Hurk #8 Posted Feb 23 2016 - 00:49

    Major

  • Players
  • 59979 battles
  • 17,409
  • [KGR] KGR
  • Member since:
    09-30-2012

View PostTheLoveHitman, on Feb 22 2016 - 15:28, said:

A GREAT players will influence all 100% of their battles, they will only lose the battles where their team was just uncarriable. I'd say 30% of games you'll probably win without doing much, 30% of games you'll lose without doing much. The other 40% comes down to how much you can do. That's why you don't see many people above 70% winrate and below 30% winrate. (If you manage below 30% winrate you need to uninstall, this is not a place for you)

 

barring platooning and cherry picking which tanks you play, almost all players are below a 65% win rate, which is a 30% window. 

 

 



Overdile_Spittle #9 Posted Feb 23 2016 - 00:54

    First lieutenant

  • Players
  • 63925 battles
  • 993
  • Member since:
    07-29-2011

I do run xvm and will not turn it off.  I do not despair if I get a game that is sub 40% to win.

  • These are the battles that make the difference IMO. 
  •  these are the ones that I get the most satisfaction out of winning.
  •  These are the ones that I usually get my 'Aces' in.

 

 

I lie,  I do sometimes despair if I get endless sub 40% games.  I not however give up or suicide.



LesterQuaestor #10 Posted Feb 23 2016 - 01:15

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

Block Quote

 I do not despair if I get a game that is sub 40% to win.

 

An attitude which does you credit.

 



LesterQuaestor #11 Posted Feb 23 2016 - 01:30

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

Block Quote

 they will only lose the battles where their team was just uncarriable.

 

   We might need to discuss what "carry" means. A monster game is probably easier to get in a blowout. But a blowout isn't a carry. Games where the unicorn actually makes the difference might seem to be pretty poor battles. I believe every game is winnable (and losable) until the result is known.

 



TheLightKnight #12 Posted Feb 23 2016 - 01:59

    Captain

  • Players
  • 19158 battles
  • 1,371
  • [X-COM] X-COM
  • Member since:
    06-09-2014
I'm telling you man, win percent denyers are like a broken record.  "But...but...there's always games you can't win when RNG and mm decide you die, what about those!!"  :facepalm:  Obviously no one in this game will have a 100% win rate, because you can't have complete control over everything.  If you put me on an NFL team as a quarterback, we're going to lose more games than if say Aaron Rodgers was quarter backing.  Of course there's going to be times I would win just because the team could overcome my terribleness.  The point is, you need to be the one overcoming your teams badness, not the other way around.  

yewchung #13 Posted Feb 23 2016 - 02:17

    Major

  • Players
  • 4438 battles
  • 4,915
  • Member since:
    05-03-2011
If you damage your enemy 10 times EVERY match you have a lot higher chance of winning regardless of what tank you are on..

LesterQuaestor #14 Posted Feb 23 2016 - 03:29

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

Block Quote

 Obviously no one in this game will have a 100% win rate, because you can't have complete control over everything.

 

 One of the points I want to bring forward in the future is that WoT is much more balanced than real life. If you got 9 random people together at a gym to play basketball and added one top quality NCAA player, you could never arrange teams that had even a 90-10 balance. Even if you left out the good player and got a random 10 people to play, you would still be unable to make balanced teams with any sort of regularity. WoT is quite exceptionally balanced.

ISNomads #15 Posted Feb 23 2016 - 18:16

    Major

  • Players
  • 43265 battles
  • 2,495
  • [WHAMO] WHAMO
  • Member since:
    11-30-2013

The only part I agree with is that, in a very small sample, it is easier to influence the win rate and that players should not give up. But once again you have oversimplified everything resulting in a meaningless post. Your example depends on exactly even teams or ignoring all other variables. Platooning with a good player and working together can have a much greater influence that a solo player with no support. Sometimes you play like crap and the team carries you and it has nothing to do with your influence. And your conclusion of 'if you try and don't give up you are more likely to win' isn't really rocket science.



ISNomads #16 Posted Feb 23 2016 - 18:35

    Major

  • Players
  • 43265 battles
  • 2,495
  • [WHAMO] WHAMO
  • Member since:
    11-30-2013

View PostLesterQuaestor, on Feb 22 2016 - 18:29, said:

 

 One of the points I want to bring forward in the future is that WoT is much more balanced than real life. If you got 9 random people together at a gym to play basketball and added one top quality NCAA player, you could never arrange teams that had even a 90-10 balance. Even if you left out the good player and got a random 10 people to play, you would still be unable to make balanced teams with any sort of regularity. WoT is quite exceptionally balanced.

 

Those are false and misleading statements.

 

The balance in win rate over a significant sample size is due to there being essentially only 2 outcomes, win our lose. Sometimes you will be on the winning team, sometimes on the losing team. In individual battles there is often imbalance, particularly when you have a platoon of good players on one side (i.e. your 3 NCAA players).

 

For your basketball example, you can never create teams that have an exact percentage balance whatever it may be, if you assume static teams then it cannot be correlated with WoT, and over a more significant sample 10,000 games and randomly picked teams you would probably approach a 50/50 balance in win rate as would be expected due to the (essentially) 2 possible outcomes. Why provide an example about 'arranging' teams to achieve a specific skewed percentage in order to illustrate a point about a game where the teams are not specifically arranged to achieve a given percentage? In WoT, individual teams are just as imbalanced as your basketball example, you are just ignoring sample size for both.



LesterQuaestor #17 Posted Feb 23 2016 - 19:47

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

Block Quote

 The only part I agree with is that, in a very small sample, it is easier to influence the win rate and that players should not give up

 

Remarkably, the only part you agree with is completely backwards. For a small sample size it is more _difficult_ to influence the win rate. You might get 5 bad teams in a row. You might get 5 good teams in a row. In fact, such runs should be expected to occur every few hundred games. But for a large sample size, everything cancels out and what you are left with is _your_ effort.

 

Block Quote

  Your example depends on exactly even teams or ignoring all other variables.

 

  No. My example depends only on random teams and a sufficiently large sample. Then all variable cancel out leaving just your performance.

 

Block Quote

  And your conclusion of 'if you try and don't give up you are more likely to win' isn't really rocket science.

 

  No, it isn't rocket science. But you'd be surprised how many people want to argue about it.

 

 



LesterQuaestor #18 Posted Feb 23 2016 - 20:00

    Captain

  • Players
  • 12944 battles
  • 1,313
  • Member since:
    03-26-2015

Block Quote

  In WoT, individual teams are just as imbalanced as your basketball example, you are just ignoring sample size for both.

 

No. You haven't thought it through carefully enough.

 

Take 9 random people and one NCAA star. The win rate will be 99%+ to the team with the star. It will be so imbalanced that it wouldn't even be worth playing.

 

Take 29 random players and add one super unicum. In this case the win rate will be only about 55-65% to the strong side.

 

WoT is much more balanced than real life. If you want balance for a real sport or game, you need to divide people up by ability, by age, by a draft-and-trade system, or else use a handicap or ladder format. In WoT, you can mix purples and reds in the same player pool and still almost all games are with the 40-60% range.

 



Jyarbro28 #19 Posted Feb 23 2016 - 21:27

    Sergeant

  • Players
  • 26083 battles
  • 211
  • [VULCN] VULCN
  • Member since:
    08-07-2014
As Stephen Hawking himself notes in his first book's acknowledgements that he was warned that for every equation in the book, the readership would be halved. Even with visualizations people see math and refuse to understand. The concepts are simple, the conclusion accurate, but, like the need to spam gold, and cast salty commentary at "lesser" players, some people just don't get it. If it isn't easy, it's not worth doing.

Edited by Jyarbro28, Feb 23 2016 - 21:28.


UR_TANK_SPLODED #20 Posted Feb 23 2016 - 21:37

    Captain

  • Players
  • 7865 battles
  • 1,358
  • Member since:
    02-26-2012
When I see a 45% chance, as a below average player, I KNOW it means me.  I have to step up and go to a key location and get some early damage, hopefully 2 kills.  When you are below average, XVM should help give motivation.  When I am bottom tier in a 60 percenter, I know I can take a little longer and explore some new locations and practice spotting, kemp bush, hull down, and such.  I quite like to know how much is riding on me, and this info has helped my WR increase and XVM actually has accounted fir this as my average winning chances SEEM to be higher, more 50ish than 40ish.




1 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users