Jump to content
Flandre Bug Read more... ×
You need to play a total of 50 battles to post in this section.
qdfl

Win Rate, what is that mean?

124 comments in this topic

Recommended Posts

[EX-TH]
Players
57 posts
13,971 battles

Hi,

I really wondering why some players has busy team mates or enemies win rate. When I suggest some thing while I am in game, after couple of min or after game players send a message to me that my win rate is low therefore I am noob I should not give suggestion etc.

 

Why players dont understand that win rate is not a PERSONAL RATE, it is numbers how many times YOUR TEAM WIN, not you alone win!

 

Suppose you start today for playing and your team generally has %70 win rate I assume your team win and your win rate %100 for first game. And basically it is possible, if you play 1000 games with high win rate team mates, your win rate will be also up to %70 win rate after 1000 games even if you didnt shoot or kill any enemies.

 

So, what is the points for this statistic? This is all depend match making system...

  • Cool 5
  • Funny 8
  • Boring 2
  • Bad 7

Share this post


Link to post
Share on other sites
[SHAD]
Players
3,438 posts
14,153 battles

It's to do with large sample sizes of random quantities (other players) with one constant (you).

In a large enough sample it works well enough but It's not perfect.

 

I think the reason we use it is because if we don't use that, what do we use?

  • Cool 6

Share this post


Link to post
Share on other sites
Beta Tester
2,189 posts
6,128 battles

Over a few matches, it's true it could be down to luck. I've had it where after 11 matches I lost 9 (always in top three with up to 5 kills), due to team mates performing poorly as you suggest. But that's very excessive and had never occured to me before. However, over a large amount of matches, your personal contribution will show because you're either a constant liability or a constant contributor to the team, or anywhere in between. That same ship now has a winrate of ~60% at a significantly larger amount of matches.

 

The reason being that if you perform well every match or nearly every match, the chances of your team winning increase. If you always perform poor, your team will stand to lose a player or have lower damage per second than they need to have. Thus they will be more likely to lose.

 

 

Hence WR says something about your performance by indicating how well your teams do on average.

 

 

  • Cool 4

Share this post


Link to post
Share on other sites
[EX-TH]
Players
57 posts
13,971 battles
3 minutes ago, Figment said:

Over a few matches, it's true it could be down to luck. I've had it where after 11 matches I lost 9 (always in top three with up to 5 kills), due to team mates performing poorly as you suggest. But that's very excessive and had never occured to me before. However, over a large amount of matches, your personal contribution will show because you're either a constant liability or a constant contributor to the team, or anywhere in between. That same ship now has a winrate of ~60% at a significantly larger amount of matches.

 

The reason being that if you perform well every match or nearly every match, the chances of your team winning increase. If you always perform poor, your team will stand to lose a player or have lower damage per second than they need to have. Thus they will be more likely to lose.

 

 

Hence WR says something about your performance by indicating how well your teams do on average.

 

 

If I change your above word to ; If some players always perform poor, my team will stand to lose a player or have lower damage per second than they need to have. Thus we will be more likely to lose. And my win rate going to down because of other player/'s.

  • Funny 1

Share this post


Link to post
Share on other sites
Players
828 posts
3,739 battles

Most likely you are not getting flamed for your WR only but also your PR and / or actions in that certain game. Your influence on a match is higher than you would think. 30-40% of the games are either lost or won without your contribution, but in 60-70% of the matches you can directly influence the outcome.

 

I am in no way stat-shaming you, but with a PR of 840 after almost 8.000 games over all classes and ships it is very safe to say that you do not really master the ships you have yet sailed. Your average damage is also very low (e.g. 44.000 in tier X ). In the majority of games you are more of a burden for your team instead of a help and you have died yourself up the tiers too quickly before actually learning the mechanics and improvig your gameplay.

 

This contributes to your bad renown, because within the community those types of players are hated the most. In the majority of cases such players are rather ignorant towards feedback and tips and know it better all the time and they also usually insult their team and blame it for their fail. I am not saying this is justified, but many players are having fun winning and there is nothing worse than being let down by a team member that clearly refuses to do his homework.

 

The worst part is that many players like you actually improved and want to learn, but their stats are ruined for a long time into the future because of their approach in the past.

  • Cool 8
  • Funny 1

Share this post


Link to post
Share on other sites
[EX-TH]
Players
57 posts
13,971 battles
10 minutes ago, gopher31 said:

It's to do with large sample sizes of random quantities (other players) with one constant (you).

In a large enough sample it works well enough but It's not perfect.

 

I think the reason we use it is because if we don't use that, what do we use?

Use statistic is ok, but using against to other player is strange idea (while I know I can make private)

Share this post


Link to post
Share on other sites
[SM0KE]
Players
6,576 posts
15,046 battles
Just now, qdfl said:

If some players always perform poor, my team will stand to lose a player or have lower damage per second than they need to have. Thus we will be more likely to lose. And my win rate going to down because of other player/'s.

I suspect you may not be distinguishing between small and large data-sets: in an individual game (or even a dozen or so), the individual players on your team will have a drastic impact on your WR (although you should recall that it can easily be a positive impact, if you happen to get a unicum on your team, rather than someone tuberous); once your data-set starts getting into the hundreds, then the effect of individual other players becomes largely irrelevant, and the only constant is you.

 

The bulk of Random games are going to have their result regardless of what you do; however, it's the edge cases that matter if you are concerned about WR - these are the games where your good performance will clinch the win, and/or where doing less well will clinch the defeat, and over a large enough sample, tell you - in general terms - how good the player is.

 

For example, lets look at this hapless berk:

 

Clearly a potato?

image.png.0c82596d4efa8af0edcc5acc3c15e67c.png

 

Maybe a unicum?

image.thumb.png.4b32e824b5e30be272dbdc3f15449180.png

 

Nope, just a slightly sub-average player:

 

image.png.5ebd606c3786f122ba7a731df473c395.png

 

TL;DR: WR is a pretty decent metric (provided you know what you're looking at; which mode are we talking about; which ship(s) and so on...), but only with a decently sized sample set.

 

 

  • Cool 1

Share this post


Link to post
Share on other sites
[CATS]
Players
29,665 posts
15,312 battles
25 minutes ago, qdfl said:

Suppose you start today for playing and your team generally has %70 win rate I assume your team win and your win rate %100 for first game. And basically it is possible, if you play 1000 games with high win rate team mates, your win rate will be also up to %70 win rate after 1000 games even if you didnt shoot or kill any enemies.

A 50% player getting 70% solo WR after 1000 battles by luck is less probable than picking up a grain of sand in the Sahara, throwing it away, and somebody else picks up the same grain the next day.

  • Cool 5
  • Funny 2
  • Bad 1

Share this post


Link to post
Share on other sites
[PIKES]
Players
512 posts
11,449 battles
16 minutes ago, qdfl said:

Use statistic is ok, but using against to other player is strange idea (while I know I can make private)

Make the statistic private is not useful for you too. If you want to see your evolution in time in that private condition, the portals which follow your performance didn't register either. I have a horrible WR, and only I improved a little when I published it and trying to follow the evolution. :Smile_sceptic:

Share this post


Link to post
Share on other sites
Beta Tester
2,189 posts
6,128 battles
24 minutes ago, qdfl said:

If I change your above word to ; If some players always perform poor, my team will stand to lose a player or have lower damage per second than they need to have. Thus we will be more likely to lose. And my win rate going to down because of other player/'s.

Thing is, everyone has those players on their teams. The only consistent player present in your teams, is you yourself.

 

What you're doing is projecting blame away from yourself. Always, ALWAYS first and foremost look at your own skills and capabilities, for if you know your weaknesses, you know what to improve upon. If you always shift blame to other players, what reason would you have to improve yourself?

 

 

Trust me in that if you are your own harshest CONSTRUCTIVE critic (i.e. don't bash yourself, but critique yourself constructively by analyzing what your mistakes were and learn from them so you'll be less likely to make the same mistake). Afterwards you'll enjoy the game much more because you'll be far more independent from the actions of other players and you'll win more.

 

 

If however, you do as you do now, blame other players, then you'll never have reason to learn and thus you won't improve. I'm not saying that other players can't drag you down, because they do all of us, as all good players drag us up. But if you don't improve yourself, you'll be dragging yourself and others down as well.

  • Cool 5

Share this post


Link to post
Share on other sites
[CHEFT]
Players
12,228 posts
9,841 battles
47 minutes ago, qdfl said:

I really wondering why some players has busy team mates or enemies win rate. When I suggest some thing while I am in game, after couple of min or after game players send a message to me that my win rate is low therefore I am noob I should not give suggestion etc.

 

Maybe people flame you because you seem to be playing extremely passive? You survive more often (>51%) than you win (47%) thus you often survive even if you lose. There is no reason to survive, you dont gain anything from it. And people get easily upset with that, if you run away instead of trying to win or help others try to win.

 

38 minutes ago, qdfl said:

If I change your above word to ; If some players always perform poor, my team will stand to lose a player or have lower damage per second than they need to have. Thus we will be more likely to lose. And my win rate going to down because of other player/'s.

 

What does prevent poor players from being in the enemy team? Nothing. What does prevent good players from being on your team? Nothing either. So in the long run, you will have very bad, bad, average, good and Unicum players on your team as often as having them on the enemy team. And since you are the only one ALWAYS being on your team, it matters what YOU are doing.

It might not matter the next game, and maybe not the game after. But it will matter in the long run.

  • Cool 8

Share this post


Link to post
Share on other sites
Beta Tester
5,562 posts
22,154 battles
1 hour ago, qdfl said:

Why players dont understand that win rate is not a PERSONAL RATE, it is numbers how many times YOUR TEAM WIN, not you alone win!

 

But Your personal skill makes the difference.

A 54% average winrate player - like... maybe... me? - getting into a battle with only 47% average winrate players - like You - should (!) be skilled enough to make an impact and help his team to win more battles than they lose.

While on the other hand a 47% average winrate player - like You - teamed up with only 54% average winrate players - like me and many others here on the forum - probably only makes an impact in a negative fashion by f*cking up and being the one little stone that starts the avalanche of loss. Which is, in the end, what leads to You having a 47% average winrate and others having 54%+

 

Or, to say it easier: You're a potato and should git gud!

:Smile_trollface:

  • Cool 2
  • Funny 1
  • Bad 1

Share this post


Link to post
Share on other sites
Players
1,956 posts
5,168 battles
2 hours ago, qdfl said:

Why players dont understand that win rate is not a PERSONAL RATE

It is a personal rate because the average winrate of your teammates over thousands of games is 50% - many don't want to hear it, it says everything about how good or bad you are.

 

-If your winrate is 40%, that means you're so terrible the odds of your team winning is reduced by 10% because you are there, despite you being only one of twelve ships on the team.

-Going below 40% is almost impossible without actively sabotaging your own team or divisioning up with other really bad players, because then one quarter of every team you are in is bad players.

-If your winrate is 60%, that means you're so great the odds of your team winning is increased by 10% because you are there, despite you being only one of twelve ships on the team.
-Going over 60% is really hard playing solo, the way to get substantially over 60% is playing with two other good players, because then one quarter of every team you are in is good players.

  • Cool 5
  • Boring 1

Share this post


Link to post
Share on other sites
[SCRUB]
Players
2,928 posts
5,330 battles
2 hours ago, qdfl said:

If I change your above word to ; If some players always perform poor, my team will stand to lose a player or have lower damage per second than they need to have. Thus we will be more likely to lose. And my win rate going to down because of other player/'s.

 

If you play thousands of games, then by probabbility, 46% of the time those bad players will be on your team, 50% the enemy team. So if over a thousand games you win or lose more, that's not luck of who is on your team, because by same luck you win or lose equally. The only factor rising above or falling below the statistical average is YOU.

  • Cool 1

Share this post


Link to post
Share on other sites
Players
1,956 posts
5,168 battles
3 hours ago, qdfl said:

(while I know I can make private)

You don't want to do that, the unwritten rule about hidden profiles is you hide your profile if your winrate is atrocious. You're at 47%, it's not terrible, just below average. People will still support you with 47%. If you get closer to 40% or hide your profile, people will assume you are useless and let you die. The way many people use the stat monitor is decide who they support and who they don't support, to increase their own chance of winning. So if you're in a DD, support cruisers won't risk HP to push up a bit and radar to help you win, if you're in a cruiser, battleships won't risk HP to help you win a fight, if you're in a battleship, DDs won't risk HP to fight off an enemy DD hunting you. Because 5K of their HP is more valuable to the team than you and your entire ship. Basically hidden profile means you are no longer worth supporting in battle and they will just let you die or use you as bait, same applies to 40%'ers

Example : If you're in a Des Moines, losing 5K HP to make a Daring back off and save your 55% WR halland is worth it. Losing 5K HP to make a Daring back off and save your 42% or hidden profile halland is not worth it. Many good players make this "calculation" all the time using monitor.

 

If you see a 39% winrate or hidden profile shimakaze with 5K average damage, for your own benefit, you should already start kiting and prepare for not having a DD on your flank, he's going to die in 90 seconds. And you certainly shouldn't push up with him or waste a radar consumable to help him win, he won't win, he has never beat the enemy DD and he never will. All this information is incredibly useful if used wisely.

  • Cool 5
  • Boring 1

Share this post


Link to post
Share on other sites
[LEEUW]
Players
2,897 posts
10,504 battles
3 hours ago, qdfl said:

Why players dont understand that win rate is not a PERSONAL RATE, it is numbers how many times YOUR TEAM WIN, not you alone win!

Man let me tell you: I had some pretty good teams who carried me to 68%wr the passed 10000 games. Not gonna lie. 

  • Funny 14

Share this post


Link to post
Share on other sites
Beta Tester
5,562 posts
22,154 battles
1 hour ago, NikolayKuznetsov said:

-If your winrate is 40%, that means you're so terrible the odds of your team winning is reduced by 10% because you are there, despite you being only one of twelve ships on the team.

-If your winrate is 60%, that means you're so great the odds of your team winning is increased by 10% because you are there, despite you being only one of twelve ships on the team.

 

Though I'm globally agreeing with You, this looks a wee bit like effed up maths to me.

I mean, there's 12 players on that team.

So if there's 11 with a 50% winrate and only 1 with 40% the totally win% for the team should (in my probable flawed and un-mathematic logic) kinda like ((11x50%)+(1x40%))/12 = 590%/12 = 49.17%

And that 60% unicorn would raise - mathematically - raise the average win% for the team to a stunning ((11x50%)+(1x60%))/12 = 610%/12 = 50.84%

 

Of course only if they use a class of ships that even is able to have enough influence on a battle.

:Smile_hiding:

Share this post


Link to post
Share on other sites
Players
1,956 posts
5,168 battles
38 minutes ago, GarrusBrutus said:

Man let me tell you: I had some pretty good teams who carried me to 68%wr the passed 10000 games. Not gonna lie. 

Exactly, you just got lucky with matchmaking, ten thousand times, in a row.

PET THE PEPEGA.gif

  • Funny 5

Share this post


Link to post
Share on other sites
Players
1,956 posts
5,168 battles
11 minutes ago, Deckeru_Maiku said:

this looks a wee bit like effed up maths to me.

You are overcomplicating the math, a 40WR player is so useless that any team he is on has a 60% chance of losing, despite him being only one of twelve ships. A 60WR player is so useful that any team he is on has a 60% chance of winning, despite him being only one of twelve ships.

You can clearly see this ingame, 40WR players are not "kinda like 60% players but they make some mistakes" - They do everything wrong. Like to get 40% you need to straight line DDs into the cap without a camo and die, you need to show flat broadside in cruisers to the first ship you see and stay broadside until you are dead.

40WR basically means the team would have been better off if you weren't there and it was 11v12, probably because you're a bigger threat to your own team than anyone on the enemy team, like shimakazes on the back line spamming 20km torps through their own team, stuff like that. It's really hard to get 40% you have to practice being terrible.

The people launching torps with 6km range at targets 15km away get at least still 45% WR

  • Cool 2

Share this post


Link to post
Share on other sites
Players
3,652 posts
14,315 battles
52 minutes ago, GarrusBrutus said:

Man let me tell you: I had some pretty good teams who carried me to 68%wr the passed 10000 games

It's only taken me 700 solo games to be carried on my new NA account :P so beat that! 

 

image.thumb.png.89ff70239e137c2a1ab918e77dc81fd2.png

  • Funny 2
  • Boring 1

Share this post


Link to post
Share on other sites
[THESO]
Players
1,452 posts
17,767 battles

Is there a specific reason why we have to go over this once a week?

 

Once a week some troll or other person that has no clue about the math of statistics, opens up one of these topics, like we have never seen it before.

And again starts the discussion about:

"Why are people telling me I am bad because of my winrate? Maybe I just had bad teams for 1000 battles while other had good teams for 1000 battles. The common denominator of all my battles is some conspiracy of Wargaming against me, cause I am so darn important to them and the world."

 

It's okay, if you're not good at something, nobody is good at everything. Just don't come in here and dump your unscientific excuses for not being good at this one thing. If you want to be good, make an effort. If you don't want to make an effort, accept you're not gonna be good. If you want to go to space, go there and don't argue that naming is just a matter of arbitrary definition and that space is wherever you are right now. That's ridiculous.

  • Cool 3
  • Funny 1

Share this post


Link to post
Share on other sites
Players
1,956 posts
5,168 battles
9 minutes ago, HMS_Kilinowski said:

Once a week some troll or other person that has no clue about the math of statistics, opens up one of these topics

Id say its two per week at least but otherwise excellent post sir

Share this post


Link to post
Share on other sites
Beta Tester
5,562 posts
22,154 battles
1 hour ago, NikolayKuznetsov said:

You can clearly see this ingame, 40WR players are not "kinda like 60% players but they make some mistakes" - They do everything wrong. Like to get 40% you need to straight line DDs into the cap without a camo and die, you need to show flat broadside in cruisers to the first ship you see and stay broadside until you are dead.

 

Honestly, I don't see that in the game.

I've got PotatoAlert running for a while and I see that there's usually abundance of 47% average winrate teams fighting against 48% average winrate teams, which end with a horrible roflstomp of the 47%s over the 48%s...

 

These teams often have players within them, that have a total average winrate above 55% or often at least about 100+ battles played in the ship in battle with such a winrate (= 60%) and to my great astonishedment these usually do not play that much better than You claim. They just get dragged down by stupid...

 

So no, no single player is influential enough to lower or raise the in chance of a team for a solid 10%.

And yes, a player with a 47% average winrate is a horrible potato who - when grouped with more of his type, which seems the current usual situation - lessens the winchance of his team by a slight amount. Unless he has a good day or doesn't play an influential class/ship, get's disco'd at the start of the battle etc. etc. etc.

Share this post


Link to post
Share on other sites
[TORAZ]
Beta Tester
14,704 posts
20,997 battles
6 hours ago, qdfl said:

Suppose you start today for playing and your team generally has %70 win rate I assume your team win and your win rate %100 for first game. And basically it is possible, if you play 1000 games with high win rate team mates, your win rate will be also up to %70 win rate after 1000 games even if you didnt shoot or kill any enemies.

 

The chances of this happening are astronomical.

In fact to say that they're astronomical would be an understatement.

 

https://forum.worldofwarships.eu/topic/76406-win-rate-and-luck/

  • Cool 5
  • Boring 1

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

×