Why such emphasis on win rate

[ColonelPete]

15 minutes ago, hogger4169 said:

Hello, a matchmaking based on average winrate of each player and categories of -/+ 5% will tell a lot. Players who favored from the mm glitches will have to prove themselves.

Since the current mm calculates the average winrate of the team a big bulk of bad players can "force" a good player to lose and since it is random [ call me RNGesus ] and vice versa.

my 2 cents...

A MM based on WR would render the value, it is based on, useless. When everyone has around 50% WR, how should such a MM work?

[Hogger4169]

44 minutes ago, ColonelPete said:

A MM based on WR would render the value, it is based on, useless. When everyone has around 50% WR, how should such a MM work?

You mean that a player with winrate of 60% has a plasmatic value?

 

Example:

1.) I consider meself a player who deserves a bigger winrate but bad mm grounded me to 40%.

2.) if what I wrote above was in the game mechanics...

then...

I would have to prove meself that I deserve bigger winrate. This is the goal of me life (irony included).

In game when a player is not good or very good to drag the team to victory everyone criticizes his/her winrate, saw it very often in chat. It starts to become toxic in a game when the winrate is a statistics value. And statistics can be wrong often.

 

In conclusion, we are wrong when we judge a player from the stats and not on performance ingame…

[Hanse77SWE]

22 minutes ago, hogger4169 said:

You mean that a player with winrate of 60% has a plasmatic value?

If 60% WR players meet and they are equal good their "new" WR will be 50%.

If 40% WR players meet and they are equal good thei "new" WR will be 50%.

 

In the end, everybody will be 50% and WR will be pointless.

 

Stats ARE a value of the performance ingame. 

[Hogger4169]

19 minutes ago, Hanse77SWE said:

If 60% WR players meet and they are equal good their "new" WR will be 50%.

If 40% WR players meet and they are equal good thei "new" WR will be 50%.

 

In the end, everybody will be 50% and WR will be pointless.

 

Stats ARE a value of the performance ingame. 

And it will require an amount of effort, to be even better to go to 55% winrate and above and not an easy task killing 40%'s and O.T.O.H. 40%'s will have the chance (which is sparse now due to bad mm, admit it plz)to go to 50% winrate...

and it will be more in equal terms....

another issue I would like to mention is the -/+ 2 tiers, actually 3 and reduce to 1 tier difference actually 2 but I go off topic...

[ColonelPete]

48 minutes ago, hogger4169 said:

Example:

1.) I consider meself a player who deserves a bigger winrate but bad mm grounded me to 40%.

2.) if what I wrote above was in the game mechanics...

then...

I would have to prove meself that I deserve bigger winrate. This is the goal of me life (irony included).

No. You could go AFK and your WR would rise.

To compensate for you bad performance, the MM would put better and better players in your team and worse players in the enemy team, until everyone is around 50%.

If you continue to be afk, your WR would fluctuate a bit, the MM would alway try to find player to compensate you.

 

But THANKFULLY we do not have such a MM and the 40% solo-WR player who got such a WR after a 1000 battles deserves it by being bad.

Or as I have often said, show me ONE good player with a solo winrating of 40% after 1000 battles, just one!

 

I ask this question for over three years regularly, but nobody can show me such a player.

[Hanse77SWE]

39 minutes ago, hogger4169 said:

And it will require an amount of effort, to be even better to go to 55% winrate and above and not an easy task killing 40%'s and O.T.O.H. 40%'s will have the chance (which is sparse now due to bad mm, admit it plz)to go to 50% winrate...

and it will be more in equal terms....

I have no idea what you're talking about here buddy.

[Maj_Havoc]

2 hours ago, hogger4169 said:

You mean that a player with winrate of 60% has a plasmatic value?

 

Example:

1.) I consider meself a player who deserves a bigger winrate but bad mm grounded me to 40%.

2.) if what I wrote above was in the game mechanics...

then...

I would have to prove meself that I deserve bigger winrate. This is the goal of me life (irony included).

In game when a player is not good or very good to drag the team to victory everyone criticizes his/her winrate, saw it very often in chat. It starts to become toxic in a game when the winrate is a statistics value. And statistics can be wrong often.

 

In conclusion, we are wrong when we judge a player from the stats and not on performance ingame…

You cannot blame the MM, this is a common mistake, and one many players throw out instead of looking closer to home. Statistically, over time, given enough matches, you will be on as many good teams as bad; at that point, only your efforts become the influencing factor on your W/R.

 

If it feels as though you are on too many losing teams, your W/R is less than you would like and your kill to death ratio is below 1.01. Assuming you want to improve your W/R, it is time to start looking at the way you play and what you might do differently. Of course, your preferred ship class plays into the influence you will have on your games, and you may need to consider playing a different class that can affect greater influence in matches if your preferred class is not cutting it for you.

 

Having said all that, as long as you enjoy playing, unless you have aspirations of playing competitively do not get too hung up on W/R. Having fun is far more important than any stat, and should be the number one goal playing any game.

[Hanse77SWE]

1 minute ago, Maj_Havoc said:

Having said all that, as long as you enjoy playing, unless you have aspirations of playing competitively do not get too hung up on W/R. Having fun is far more important than any stat, and should be the number one goal playing any game.

Well said but I just want to clarify that playing well is fun. At least more fun that not playing well.

[RossoneriFan2]

There is a reason that some people have Solowinrates of 50% 55% or 60% after thousandth of games: skill.

@ColonelPete

I'm not going to sit here and argue the obvious - which incidentally seems to be another favourite past time of WoWs players btw.
Haven't you had those days where you'd log into the game, start playing and for 10 straight games you simply cannot get a win because of jackass teams?
You can be the best player in the game. You can do 200k+ damage in every match. kill half the enemy team and still lose because of potato teams. That will still give you a 0% win rate out of 10 matches. Does that mean you are a bad player? That you lack skill? Of course not!
Win rate can be manipulated. You can go into a T4 Clemson and play the absolute living hell out of it, club a bunch of seals and come out with a 70% win rate. You can do that with a bunch of ships. And be a total @sshat in higher tiers.
I know. I used to be that player. That is why even though I have over 8k games I have only recently started focusing on T10 games. that and the fact I have been playing this game without premium for almost 5 years and it has taken me a lifetime to get enough T10 ships to actually have some sort of variety to play with. At first I didn't care about win rate. Only xp and to get to that precious Montana. Only after getting it did I start watching how to play better. But I digress.

 

My points is, that win rate is not a reliable indicator of how good a player is. It is a somewhat sketchy indicator. Average damage and average experience are far more reliable indicators as to how good a player really is in this game, as they both indicate the impact one has in game.

So saying things like " There is a reason that some people have Solowinrates of 50% 55% or 60% after thousandth of games: skill." does not reflect the truth when it comes to how goods a player actually is.
Now if you said that "there is a reason certain people have average damage values of 150k+ and average experience of 2k+. Skill." Then yes. I would absolutely agree with that statement.

[Verblonde]

3 hours ago, StakamanWT said:

My points is, that win rate is not a reliable indicator of how good a player is.

Sorry, but over a large enough data-set it is: over many games, a good player will make enough of a difference to win enough 'edge' cases to get their WR higher, whereas a bad player will have the reverse effect (and get a WR like mine, or - heavens forbid - even worse).

 

That's not to say it's the *only* useful measure, but it is emphatically reasonable to say that - with a big enough data set - a 40% WR means a potato and a 60% WR probably means it's time to bravely run away...

 

(There are a few exceptions of course e.g. a former unicum returning from a long layoff will be a bit rusty to start with, or someone hopeless who bought an account probably won't be as good as the account's WR suggests they should be, but these are rare)

[Johnny_Moneto]

Am 5.4.2020 um 19:19, ColonelPete sagte:

No. You could go AFK and your WR would rise.

To compensate for you bad performance, the MM would put better and better players in your team and worse players in the enemy team, until everyone is around 50%.

If you continue to be afk, your WR would fluctuate a bit, the MM would alway try to find player to compensate you.

 

But THANKFULLY we do not have such a MM and the 40% solo-WR player who got such a WR after a 1000 battles deserves it by being bad.

Or as I have often said, show me ONE good player with a solo winrating of 40% after 1000 battles, just one!

 

I ask this question for over three years regularly, but nobody can show me such a player.

While not agreeing or disagreeing with anything else posted here on this topic, I would like to respond to one point Colonel Pete makes: skill based MM means that for every good player there must be a bad player in the same team to compensate.

While that might be an interpretation of skill based MM it is not the only one: Much more reasonable would be to assume skill based MM puts players of relatively even skill in the same battle. And while it has been pointed out that if WR would be used as the only criteria for skill the system cannot work, there is the system of giving each player a "skill value". Such might include many things like how does the player do versus other players with a certain skill value, how many "positive" actions does a player perform in a battle. Such a skill value would then have to replace WR completely, of course.

 

[Johnny_Moneto]

Vor 4 Stunden, StakamanWT sagte:

Average damage and average experience are far more reliable indicators as to how good a player really is in this game, as they both indicate the impact one has in game.

 

Average damage: If I inflict 25k damage on a DD, it is gone. Big impact on the game. If I do 25k damage on a Conqueror, it will heal it back up as if nothing happened. Low impact on game but same amount of damage.

 

Average XP: I play the same battle on a F2P account and get 1000 base XP. That gives me 1000 XP avg XP. For the same battle having a WG premium account I get 1500 avg XP. For yet the same battle having a WoWs preium account I get 1650 avg XP.

 

 

You have to put many stats together to get a picture of how the player performs.

[ColonelPete]

4 hours ago, StakamanWT said:

There is a reason that some people have Solowinrates of 50% 55% or 60% after thousandth of games: skill.

@ColonelPete

I'm not going to sit here and argue the obvious - which incidentally seems to be another favourite past time of WoWs players btw.
Haven't you had those days where you'd log into the game, start playing and for 10 straight games you simply cannot get a win because of jackass teams?
You can be the best player in the game. You can do 200k+ damage in every match. kill half the enemy team and still lose because of potato teams. That will still give you a 0% win rate out of 10 matches. Does that mean you are a bad player? That you lack skill? Of course not!

• Nowadays I do not have 10 game loss streaks, that happened in the past when was a worse player, extremly good players do not even come near it
• Winrate is not based on your last 10 games
• average XP is a bad skill indicator since WG includes the premium modifier in XP, XP are also dependent on Tier
• average damage is a bad skill indicator as it depends on Tier and class, unless you want to claim that BB player are more skillful than DD player

17 minutes ago, Johnny_Moneto said:

While not agreeing or disagreeing with anything else posted here on this topic, I would like to respond to one point Colonel Pete makes: skill based MM means that for every good player there must be a bad player in the same team to compensate.

While that might be an interpretation of skill based MM it is not the only one: Much more reasonable would be to assume skill based MM puts players of relatively even skill in the same battle. And while it has been pointed out that if WR would be used as the only criteria for skill the system cannot work, there is the system of giving each player a "skill value". Such might include many things like how does the player do versus other players with a certain skill value, how many "positive" actions does a player perform in a battle. Such a skill value would then have to replace WR completely, of course.

• there are only few skilled player, waiting times for skilled players would be horrendous, for example only 3% of the playerbase have 58% or better winrating, how long should unicums wait in queue?
• yes, another metric than WR could work, but you would need that metric first, TruSkill™ is an elegant system, but I do not know how good it is with big teams

[Johnny_Moneto]

Vor 2 Minuten, ColonelPete sagte:

• there are only few skilled player, waiting times for skilled players would be horrendous, for example only 3% of the playerbase have 58% or better winrating, how long should unicums wait in queue?

Well, don't make the criteria too tight. The idea would be not to have all players to have even skills. It rather would be to avoid good players constantly having to play with bad players and vice versa.

 

Still, the long waiting time is also WG main argument against the introduction of such a system. While the player base itself might even be big enough, the players spread out to different game modes and tiers. In the end, some playerss could end up pretty lonely.

[Odo_Toothless]

24 minutes ago, ColonelPete said:

• how long should unicums wait in queue?

Not at all. Players: 60%+40% = 50% + 50% = 55%+45% etc.

Very good team of 3x60% + rest of the team worst players vs average others

 

We can almost always take group of random people and make game more or less ballanced. However WR of players probably will be going closer to 50% too, like in skill based MM.

[ColonelPete]

22 minutes ago, Odo_Toothless said:

Not at all. Players: 60%+40% = 50% + 50% = 55%+45% etc.

Very good team of 3x60% + rest of the team worst players vs average others

 

We can almost always take group of random people and make game more or less ballanced. However WR of players probably will be going closer to 50% too, like in skill based MM.

As explained...

• The better you are, the worse are your teammates
• The worse you are, the better the teammates that have to carry you --> great motivation to get better!
• after some time all players have around 50% WR, no matter if former potato or former unicum and the system is random again

[djole7]

The last week i keep getting this kinda of teams,it doesn't matter what i try or not.

It looks like a joke, as well very enoying to enjoy your game.

Is this normal staff cause for me kinda weird the hall situation.

[Witzulu]

On 11/7/2017 at 6:05 PM, Aranhawk said:

Curious does coop improve this..i play 99% random but i cannot be held responsible for the whole team when they drop like flies

Lol...agree with you there....