Jump to content


How to Control your Win Rate


  • Please log in to reply
229 replies to this topic

B4lduR_28 #41 Posted 05 April 2016 - 12:58 AM

    Master Chief Petty Officer

  • Members

  • 273
  • Member since:
    12-25-2015

another great writeup LWM

 

+1:medal:



Slick201 #42 Posted 05 April 2016 - 05:23 AM

    Master Chief Petty Officer

  • Members

  • 371
  • Member since:
    02-19-2016

I was in a match with LWM tonight,  Though LWM was on the "other team"; it was neat to actually be in a match with someone as respected here in the forums as LWM is.

 

If I didn't say it in the chatbox; "GG, and TY for the match".

 

o7


StLouis - Karlsruhe - Königsberg - Nürnberg - Yorck - Admiral Hipper - nig - Bayern - Gneisenau - Bismarck - Roon

 

http://www.taskforce16.us/home


huevos22 #43 Posted 05 April 2016 - 07:28 AM

    Petty Officer

  • Members

  • 80
  • Member since:
    08-09-2012
Individual performance does have an effect on win rate, but the absolute fastest way to get w/r up is to division. 60% solo, 72% in division for me. That's like an added 12% without changing how I play at all.

Lonewolfpj #44 Posted 05 April 2016 - 03:36 PM

    Lieutenant

  • Members

  • 1,991
  • Member since:
    08-09-2012
Great read and great points I will take your advice on a few items

poeticmotion #45 Posted 05 April 2016 - 04:33 PM

    Lieutenant Commander

  • Members

  • 2,336
  • Member since:
    12-25-2015
I can attest to a lot of this that I'd worked out on my own...had a 43% win rate and a couple months ago decided to actually study the game and figure out how to maximize my wins. You can't blame MM and RNG as all players should statistically run into a similar # of total failteams, faildivs,bad matchups...one can mitigate a bit through divisioning but one can't blame their low wr entirely on factors beyond their own self and be intellectually honest. 

I'm up to 46.5% wr now overall, with 51% for last 60 days, 53% for last 30 days, and 55% this week. And I've done it through focusing on fewer ships and getting better with those, learning mechanics, positioning, and maps better, and studying better players than myself. I wish I'd seen this writeup months ago, as it encapsulates so much of what I had to teach myself, and adds some stuff that I hadn't thought of which will help me climb even higher. Thanks as always for an awesome guide, LWM. This is amazing, and I can attest that it works even though it's a new post since I've been using a lot of these techniques to improve my own rate.

Tier 10: Khabarovsk, Kurfurst, Minotaur, Tier 9: Fletcher, Friedrich, Udaloi,, Z-46 Tier 8: North Carolina, Bismarck, Z-23, Edinburgh ARP Takao, Akizuki, Kiev, Ognevoi Tier 7: all ARP Myokos, Gneisenau, Leningrad, ShiratsuyuFiji, Tier 6: Budyonny, FarragutShinonome, Fuso, Leander, Bayern, Graf Spee  Tier 5: Kamikaze-R, all ARP Kongos Tier 4: Clemson, IshizuchiV170, Kaiser Tier 2: Tachibana, Mikasa

My guides: Poetic's Guide To Getting Gud Or At Least Less Bad: Tips From A Former Potato     Poetic's post-0.6.0 guide to DD skills and capt builds

The URL Guy is OP, but please don't nerf him. 


Nachoo31 #46 Posted 05 April 2016 - 04:39 PM

    Lieutenant Commander

  • Beta Testers

  • 2,313
  • Member since:
    08-09-2012
Dang it. I was up to 55% till the awful games on Saturday.

Looking for Clan?  TheKriegsmarines [KMS]  

Have you clicked on the picture in my profile?  Enjoy!

http://na.warshipsto...836912/dark.png


KromaBaSyl #47 Posted 05 April 2016 - 06:44 PM

    Master Chief Petty Officer

  • Members

  • 259
  • Member since:
    09-12-2012

View PostXannari, on 04 April 2016 - 03:53 PM, said:

 

You, nor I, nor anyone gets to decide that. You can try your hardest to influence it, but at the end of the day, you aren't the deciding factor in how often you are on the winning team.

 

Not "deciding" but "influencing" when one is talking about improving win rate over long haul. While improving your individual play won't guaranty a win on any given match it will improve your win rate over long haul by influencing the total number of wins over time, which was the point of the OP.

KromaBaSyl #48 Posted 05 April 2016 - 06:46 PM

    Master Chief Petty Officer

  • Members

  • 259
  • Member since:
    09-12-2012

I will also note from personal experience that sailing the same ship consistently was the best way to improve my individual performance and in turn my win rate got better in those ships I specialized in. 

 



Gaishu_Isshoku #49 Posted 05 April 2016 - 07:26 PM

    Ensign

  • Beta Testers
  • In AlfaTesters

  • 1,019
  • Member since:
    08-09-2012

One of, if not the best, write ups for improving play that I've seen so far. Should be required reading c:

 

I have only one thing to add, which would be to play all the ship classes (even CV) after you master your one ship. Nothing like playing a ship class to understand it, and nothing like understanding your enemy to winning against them.


"There are eight sail of the line, Sir John"
"Very well, sir"
"There are twenty sail of the line, Sir John"
"Very well, sir"
"There are twenty five sail of the line, Sir John"
"Very well, sir"
"There are twenty seven sail of the line, Sir John"

"Enough, sir, no more of that; the die is cast, and if there are fifty sail I will go through them".

 

your friendly neighbourhood high tier cruiser player


LittleWhiteMouse #50 Posted 05 April 2016 - 11:07 PM

    Captain

  • WoWS Community Contributors
  • Members
    Beta Testers

  • 5,797
  • Member since:
    01-04-2013

One of the misconceptions is that of semantics.  How can one "control" something when skill isn't the final (or in some people's opinions) or deciding factor?  For me, it's a matter of ownership and personal integrity. 

 

  • I cannot control RNG.  My choice of captain skills, modules and signals can change the parameters and minimize their influence in certain aspects, but I cannot control them.
  • I cannot control my team mates.  I am capable of influencing them, both through actions and through chat.  But I cannot control or depend upon them (including Division mates -- love you guys, but you're not my personal slaves.  Wait ... maybe you should be.  Note to self:  Explore feasibility of making mind-slaves of Division mates.).
  • I cannot control which map I will get.  The matchmaking spread of certain ships will include or exclude certain maps (for example, I must expect that I will end up on Big Race occasionally when I play ships up to tier 5), but beyond that, I have no influence on which map pops up.
  • I cannot control who I get for team mates or who I get for opponents.
  • I cannot control where I end up on the team rosters, be it top tier or bottom tier.
  • I cannot control my opponents.  I am capable of influencing them, both through action and through chat.  But I cannot directly control them short of making them explode into burny wrecks.

 

That's a lot I'm leaving up to chance.  How much of my win rate percentage does that surrender?  I don't think we have the proper data on that yet.  Lemme check warships.today/na and see what the highest and lowest win rates are for 1000 solo games at this point.  To do this, I am setting the filters to consider all battles solo played (so since the start of the system's tracking) and first sorting by highest win rate.  Unfortunately, I don't have a way of simply filtering for the lowest win rate players.  So to find the closest approximation of the bottom tier, I'll filter for most battles played and look for the lowest win rates among these players.  The reason for using this second filter is that players using bot programs are more likely to have a higher number of games played than other players. This is not conclusive, of course, but I am more likely to see low-end performance this way. 

 

Here's the results. 

  • The top 25 players (filtered by solo win rate) have win rates between 64% and 69%, all with over 1000 solo games played.
  • The bottom 25 players (filtered by most games played) have win rates between 39% and 47%, all with over 4500 solo games played. 

 

This isn't a bad result.  I would expect to see some lower stats than 39% (in Ranked Battles, we saw players with over 100 games played as low as 35%), but we'll use the 39% to 69% (a thirty point difference) as our margin of how much we can directly influence on our own.  What do these numbers mean?

 

For a start, this means that 31% of your games cannot be won.  Flat out.  The best players recorded on North America are incapable of solo-carrying 30% of their games.  So that means in about 1/3 of all of the game you play, you are going to lose no matter how well you do or how skilled you are.  This is the current ceiling limit for North American random battle solo performance.  This also means that no matter how terrible you are, you're going to win 39% of your games.  You can be downright awful, a detriment to your team and dead weight.  But you'll still win 39% of your games no matter how badly you play.  You could be all but AFK, sitting in your deployment area, twirling your turrets (if you have them) and doing nothing else.  BAM.  39% win rate. Congratulations.

 

This means all of that junk I listed at the top of the page -- all of those things I said I cannot directly determine, those values say that I will always have at least a 39% win rate.  All of the elements of RNG, all of the demons of bad teams, great teams, bad matchmaker, great matchmaker, RNG detonations, OP ships and the like, are all contained in those values.  If I do nothing, I have a 39% win rate.  That's our starting point.  That's our null-value.  That's where the counting actually starts.  I am not getting lucky and ending up with a 40%, 41% or 42% (or higher) win rate.  Luck has already been factored into the equation and the sum is 39%.  That's what luck will get you, full stop.  Nothing higher than that.

 

Skill now accounts for the rest.  There are 30 percentage points up for grabs before you reach the ceiling of the maximum solo value we've determined possible on the server.  How good you are at all of the little things that makes a brilliant player resides in those thirty percentage points.  That's the area you control.  That's the area that is solely in your hands.  When I look at a player that boasts a 53% win rate, I am looking at someone that can grab 14 out of the 30 points available.  That's how skilled they are.  That's how much they can carry.  That's how much their presence influences the course of the games they play and the fortunes of their team.

 

And it's those 30 points that you control.  It's those 30 points that you have to stand up, take ownership for.  How many do you currently claim?  How many do you want to claim and will work towards?


Q♥  Most Recent Review: HMS Hood
  


Nhi_Vanye #51 Posted 05 April 2016 - 11:20 PM

    Warrant Officer

  • Members

  • 590
  • Member since:
    12-29-2015
Brilliantly stated! :medal: As I, and I think many others, have observed.  If you work at getting better, you make steady repeatable progress in that 30 point range. :)  With your advice, I expect to keep improving and move towards the top of that range over time. :honoring:  Thanks for another great post!

"There seems to be something wrong with our bloody ships today" - Vice Admiral Sir David Beatty


Hopelesshobo #52 Posted 06 April 2016 - 05:54 AM

    Chief Petty Officer

  • Beta Testers

  • 143
  • Member since:
    05-21-2014
  • Always play the objective.
  • Know when you need to tank and know when you need to DPS.  For instance, if you are taking fire from several enemy BBs, don't worry about applying all your guns.  Just mitigate as much damage as possible while your team dishes out the pain.
  • Work on shooting accurately
  • Put yourself in the enemy's shoes, and play accordingly.  On the loading screen see what their fleet comp is and predict where they will go and what you will probably have to face on your side of the map
  • If you are in a USN BB before the North Carolina, don't try to flex to the other side of the map (A few exceptions apply).  This applies for any ship that goes 20 kts, most of the time the fight will be over by the time you get there
  • If you very low on health (I was sub 100 health the other day in my Pensacola), know when to hide from the enemy.  This can deny the enemy the points, and you turning and running behind and island to live can win it for the team instead of trying to put another salvo into a ship that won't die, which can put the enemy team up on points.
  • If you do not have a base that you need to defend.  Lemming trains are good.  It allows you to overmatch 1 side and it allows you to cover each other with your AA.  Also it makes it easier on your CVs to cover you, because it's easier to intercept bombers when they are all going to 1 side of the map.


MeKanism #53 Posted 06 April 2016 - 06:32 AM

    Warrant Officer

  • Members

  • 733
  • Member since:
    09-25-2015
Knowledge is power +1

Edited by MeKanism, 06 April 2016 - 06:33 AM.

Complete Super-Unicum right here

Ravioli, Ravioli, what's in the pocket-oli

 

 


Tk3997 #54 Posted 06 April 2016 - 07:52 AM

    Seaman

  • Beta Testers

  • 48
  • Member since:
    08-09-2012

Whatever, here my reply: If you're team can't find it's own [edited]with both hands you're going to lose the overpowering majority of the time no matter how well you play.

 

I say this being on a massive losing streak that has nothing to do with me. My stats aside from win rate are actually up, and are above average to start with, but in the last fifty matches I've had a 38% win rate which has tanked my previous rate down from almost 53% to slightly under 52. I'm doing nothing different; my accuracy, damage, and survival rate hasn't suddenly tanked, I haven't been playing new ships that are stock or something. The only variable is that I've been getting a highly unusually number of very crappy teams. Just tonight I played three matches in my Atago, arguably my best ship, I averaged 70,000 damage per game with multiple torp hits, plane kills, citadels, caps and defenses... and lost all three. Looking back there is nothing I could have done better (barring be able to teleport to other areas of the map), I played as well as I possibly could, but if your team just sucks you cannot save them. 

 

I can't stop the pristine cruiser sailing next to me losing 95% of his HP in an instant because he didn't bother to come out of binocs to look around even when his AA guns are firing at incoming TBs, I can't stop our CV taking a strike load out and losing his entire air group within like five minutes to monstrous strafing runs, or our two top tier BBs from sailing directly at six enemy ships T crossing them four minutes into the game all of which I've watched happen in the last day or two.

 

That's the lesson I have to give, your actions can only marginally influence the outcome, if the team sucks you're going to loose. If you keep getting strings of crappy teams your WR is going to tank regardless what other stats look like.


Edited by Tk3997, 06 April 2016 - 07:53 AM.


LittleWhiteMouse #55 Posted 06 April 2016 - 02:31 PM

    Captain

  • WoWS Community Contributors
  • Members
    Beta Testers

  • 5,797
  • Member since:
    01-04-2013

View PostTk3997, on 06 April 2016 - 02:52 AM, said:

Whatever, here my reply: If you're team can't find it's own [edited]with both hands you're going to lose the overpowering majority of the time no matter how well you play.

 

And statistically, you're going to be on such teams about 31% of the time.  And statistically, you're going to be able to win no matter what about 39% of the time.  That's the math of it.  The "marginal" influence accounts for the other 30%, more or less (and we're talking a level of error of roughly 10% as far as I can tell, looking at as many stats as I have).  Good luck, bad luck, good teams, bad teams?  They do not matter after a 39% win rate. 

 

A lot of the complaints people have about how often their stats say they win when compared to how often they're actually winning is an overestimation of how good they're actually playing.  This disparity between perceived performance and actual performance is a difficult pill to swallow.  If you feel like you're playing like someone who deserves a 60% win rate but you're only managing a 54%, why is there a disparity?   It's easy to point out a high damage total, multiple kills and a good survival rate and say, "Hey, I did my part.  It was my team that was garbage."  But did we?  Have you gone over the replays, looked for what you could have done differently?   I'm continually surprised by how much I missed whenever I watch a replay.  My grasp of what was going on around me isn't anywhere close to perfect.  My assessment and estimations on what my team mates are actually doing is usually quite poor.  I have had to come to the conclusion, both in my time here and in World of Tanks that if my team was bad, I was bad.

 

If I wanted more good teams, I had to play better -- to make up the core of a potentially good team.  I realized that the amount of effort i had to place in a given match had to be disproportionate to what I would end up having as an actual win rate.  If I thought I was playing good enough, odds are I wasn't.  If I thought I had done my part, odds are I hadn't.  If I wasn't seeing the numbers I felt I had earned, that's because i hadn't done enough to deserve said numbers.  If my team lost but I still cranked out 100,000 damage, then I should have pushed for 120,000 damage.  Or done that 100,000 sooner.  Or put that 100,000 damage on different targets.  I wasn't good enough.  Accepting that is difficult, especially when you feel like you've already put in 110%.  Analyzing what I could have done better and putting it into practice is almost simple by comparison.


Q♥  Most Recent Review: HMS Hood
  


MrDeaf #56 Posted 06 April 2016 - 03:01 PM

    Admiral

  • Members

  • 10,819
  • Member since:
    07-16-2015

Man, I need to Git Gud with Nagato, Hipper and Atago.

Especially Nagato, considering my Colorado is able to win significantly more often, despite doing less damage.


Now with more Youtube.

Subscribe to this thread for latest videos: http://forum.worldof...youtube-videos/

The Corgiolis Effect: When you notice you need more torpedoes on your ship


zeke3 #57 Posted 06 April 2016 - 03:07 PM

    Petty Officer

  • Members

  • 70
  • Member since:
    08-09-2012

Great thread Mouse!

The key to this analysis is "solo games", like solo pub in tanks.  If you play platoons or divisions or whatever

they are called, your group can influence your win % greatly.   Much harder to do solo, paying attention to all

of those variables Mouse mentioned in the original post.



USSViper1 #58 Posted 06 April 2016 - 04:02 PM

    Petty Officer

  • Members

  • 58
  • Member since:
    03-02-2016
Great write up Mouse! thanks.

Neltak #59 Posted 06 April 2016 - 07:20 PM

    Master Chief Petty Officer

  • Members

  • 345
  • Member since:
    03-22-2015
Great guide/write up! The largest hurdle for many players is accepting that their actions directly influence their winrate. If one wants to improve they should always be looking at what they might have done better in a battle to turn a loss into a win (or potential win). From what I've seen the stronger a person is as a player the easier they see where they should go to influence the battle to a win, and the easier time they have noticing where they could have done something better that might have led to a win.

zeke3 #60 Posted 06 April 2016 - 08:09 PM

    Petty Officer

  • Members

  • 70
  • Member since:
    08-09-2012

View PostNeltak, on 06 April 2016 - 07:20 PM, said:

Great guide/write up! The largest hurdle for many players is accepting that their actions directly influence their winrate. If one wants to improve they should always be looking at what they might have done better in a battle to turn a loss into a win (or potential win). From what I've seen the stronger a person is as a player the easier they see where they should go to influence the battle to a win, and the easier time they have noticing where they could have done something better that might have led to a win.

 

 

Now guys like Neltak are outside of Mouse's stats.   If 39% of your solo games are flat out un-winnable (max win rate  = 61%) no matter what you do according to Mouse, especially after 4000 games, that means if you have a 64% win rate you are doing things exceptionally well.      What sort of things do you notice that can lead to more wins?   Because you are doing things way better than most, unless you are platooning a lot which also can lead to an abnormally high win rate (and aren't solo games)


Edited by zeke3, 06 April 2016 - 08:10 PM.





2 user(s) are reading this topic

0 members, 2 guests, 0 anonymous users