Is win rate really a valid estimator of a players skill?

[MajesticTwelve]

As title states, I see many people evaluating a player simply based on WLR.

Now I can see people saying that a good player should have a high average based on total of games with their skill but, we've all had teams (frequently) that are unable to perform, and me personally, I find I can do much more with a team that can hold its own vs. one that folds like laundry.

I'm neutral on the subject, but very curious to get the community opinion and information.

[Salvo_Creative]

nope not at all, the difference between a 49% W/R and a 50% W/R is usually 100-200 battles if you have played 8000 battles. 

That could be as simple as being a BB main and then playing a DD for 200 battles. 

 

Focus on having fun and not a number on the screen. Anyone wanting to have a go at you for a W/R, remember that their lives are actually quite bad comparatively and that is why they have to cling to a video game in order to say "Look look, I am good, someone appreciate me" because at the end of the day , treat others how you want to be treated and have fun and that is all you need. 

 

Just sit back and smile everytime you see someone like that because they always have a horrible life and are terrible people :)  

[Ramsalot]

8 minutes ago, MajesticTwelve said:

I find I can do much more with a team that can hold its own vs. one that folds like laundry.

This is true for anyone, but particularly skilled players can sometimes carry less capable teams to victory.

As far as win rate, it usually indicates player skill when all other factors are equal to average player base.  For example it’s easier to manipulate win rate if player mostly plays ships that are generally regarded as OP in certain tier brackets, or if player mostly plays in well organized divisions.  There are some players with very high winning rates in T1 ships, but horrible win rates at high tiers.  But since they have literally thousands of games in T1 their overall win rate is well above average.

[SireneRacker]

Overall Solo Win Rate over a large sample size (3000+) gives a decent estimate on where the player stands. No need there to split hairs over one percent, but the direction it suggests will usually hold up.

Just don't try to read too much into stats, and know that even those can at times fail.

[Ossiuum]

Win rate not necessarily, though over thousands of games SOLO winrate is a great indicator. The way I explain it to people is that you're probably going to lose 40 of every 100 and win 40 more. Those 20 missing games are where the player has an impact and can influence that stat, and that is a good indicator in long term in how well they can create wins where they otherwise would lose. 

[kingsbishop_7]

18 minutes ago, MajesticTwelve said:

As title states, I see many people evaluating a player simply based on WLR.

Now I can see people saying that a good player should have a high average based on total of games with their skill but, we've all had teams (frequently) that are unable to perform, and me personally, I find I can do much more with a team that can hold its own vs. one that folds like laundry.

I'm neutral on the subject, but very curious to get the community opinion and information.

Ah.. Isn't that what Ranked Battles is all about? You have a better Win rate vs Loss Rate. (Aren't #1 Ranked Players the Best in the Game... Get my point)

[dad003]

Damage in my opinion is a better stats a player with 55% win but 20k damage will not do much but someone with 55% win but with 80k average will surely do better and do more to actually win than the one with very little damage that got basically carried 

[MrKilmister]

Of course it is.

A better question might be how precise of an indicator of a player's skill is solo win rate?

[Salvo_Creative]

6 minutes ago, MrKilmister said:

Of course it is.

A better question might be how precise of an indicator of a player's skill is solo win rate?

There is a lot of factors to the W/R that people don't take into account also, because say that person has had a bad day, lack of sleep, not enough coffee, fired from work etc..... so the W/R is an indicator of what they have previously been through in the game as well as how their lives have also been and their over all state of mind. 

So this question is going to remain open ended with no definitive answer. 

[Yoshiblue]

14 minutes ago, MrKilmister said:

Of course it is.

A better question might be how precise of an indicator of a player's skill is solo win rate?

Don't let my WR fool you. I'm actually a garbage player. 

[Final8ty]

<---- slow down dude.

[freggo]

1 hour ago, MajesticTwelve said:

As title states, I see many people evaluating a player simply based on WLR.

Depends on what you want to 'evaluate'.

Let's say a player with 6000 battles and a 55%WR shows up in Randoms in a Gneisenau.

Is he any good in her?

You simply do not know based on WR alone.

How many battles does he have in Gneisenau?

She could be his fav ship all decked out to full secondaries with a 19pt captain with 1,000 battles in her or she may be his newest ship with a 3pt captain, base hull and no upgrades or modules on her maiden voyage.

 

 

 

 

[Camo68]

Solo win rate (provided it is impacted by a lot of low tier games)  is not a sure fire way to gauge someone's skill but it is a better indicator than any other metric currently available.

If Average XP didn't include premium bonuses it would be the best but it does so cannot be used.

[AlcatrazNC]

Yes and no.

 

Yes because 99% of the time, someone with at least 50% WR will not suicide during the first 3 min or camp and start the fight after 21 min has passed.

No because even the best unicum can sometime do some mistake. (By mistake I don't mean things like rushing a cap and get surprised radar. This is not a mistake this is a suicide).

 

Usually WR, Damage and PR are what I look to guess if a player is good or not. And funny enough I haven't met a player who has garbage WR but good-great PR and damage. The same way, I never saw a bad player gets Good-great WR with garbage PR and damage.

[Sammy_Small]

Looking at the win rate with out context is not a valid estimation of player skill.

Just like looking at player average damage as estimation of skill could be flawed. 100k damage "farmed" off BBs vs. 100k damage done to CLs/CAs and DDs.

There are other relevant stats that are not publicly displayed or may be included in "Player Rating": potential damage, spotting damage, contribution to capture or defense, etc    

[SPQQKY]

No it is not but it could be made more valid if the stats could be viewed based on chronology...i.e. what is my win rate for the last 3 months..or the last week. I don't care what the player was doing 2 years ago when he/she first started to play the game. I want to know how he is playing today.

[Final8ty]

20 minutes ago, ColdwaterCowboy_1 said:

No it is not but it could be made more valid if the stats could be viewed based on chronology...i.e. what is my win rate for the last 3 months..or the last week. I don't care what the player was doing 2 years ago when he/she first started to play the game. I want to know how he is playing today.

That's my problem when players only look at my overall, it does not matter what i did 2 years ago it only matters what i have been doing recently.

[HazeGrayUnderway]

13 hours ago, Camo68 said:

Solo win rate (provided it is impacted by a lot of low tier games)  is not a sure fire way to gauge someone's skill but it is a better indicator than any other metric currently available.

If Average XP didn't include premium bonuses it would be the best but it does so cannot be used.

I remember when the game was new, looking up stats, XP average was only accounting for BaseXPs.  Not long after launch WG allowed external modifiers to count towards XP average.  It was pretty lame, IMO.  Accounting only for BaseXP would have been great as it would be "pure" regardless if the player was Premium Acct or not, etc.

 

[Navalpride33]

No, No stats is an indicator of skill...

No stats have that lvl of complexity to "judge" a players skill...

The only one who judges is the player himself but compared to what???

In order to analyze skill of a player in the game. You have to really know what to look for, if not.  Skills will = relative, instead of being absolute. 

[kgh52]

2 hours ago, MajesticTwelve said:

As title states, I see many people evaluating a player simply based on WLR.

Now I can see people saying that a good player should have a high average based on total of games with their skill but, we've all had teams (frequently) that are unable to perform, and me personally, I find I can do much more with a team that can hold its own vs. one that folds like laundry.

I'm neutral on the subject, but very curious to get the community opinion and information.

Win rate is an individual stat that teammates play the major role in creating. 

I see win rate being used to judge if another's opinion is valid or not & as a justification why their opinion is superior far too often or to say that person has not learned the game.

To top it off stats can be padded.

I'm not down playing win rate, only relating why it should not be the king of the stats as some believe it is.

 

[Viva_Palestine]

It is about 50% of the time, for me.

[CAPTMUDDXX]

No it is not.

[Ensign_Cthulhu]

3 hours ago, freggo said:

Depends on what you want to 'evaluate'.

Let's say a player with 6000 battles and a 55%WR shows up in Randoms in a Gneisenau.

Is he any good in her?

You simply do not know based on WR alone.

How many battles does he have in Gneisenau?

She could be his fav ship all decked out to full secondaries with a 19pt captain with 1,000 battles in her or she may be his newest ship with a 3pt captain, base hull and no upgrades or modules on her maiden voyage.

Yeah, this. Win rate per ship is far more indicative of performance, but even then you need a large enough number of games in that ship. Someone like me, who works up all the lines, is not going to be as experienced in a given ship and less likely to carry a battle with it.

And to address your last sentence, I always take a new ship into battle stock and naked with a zero point captain (unless it's a premium) to see what the base performance is like, but always in co-op, never in Randoms.

[Herr_Reitz]

4 hours ago, kingsbishop_7 said:

Ah.. Isn't that what Ranked Battles is all about? You have a better Win rate vs Loss Rate. (Aren't #1 Ranked Players the Best in the Game... Get my point)

You'd like to think so but sadly no. Unless you forgo the ranked grinders. 

[CylonRed]

Not at all - especially when the WR can be seriously manipulated.