Jump to content
You need to play a total of 50 battles to post in this section.
Leo_Apollo11

Interesting post about how is WG balancing ships!

70 comments in this topic

Recommended Posts

[SCRUB]
Quality Poster
4,389 posts
15,273 battles

Hi all,

 

Interesting post about how is WG balancing ships!

 

 

For most recent "Q&A #21" on "Reddit" @Sub_Octavian posted:

 

Quote

Our methods have seriously improved since game release. We have much better tools at the moment. For example, while we still pay attention to avg. stats everyone usually discusses (WR, damage, etc) and popularity, we use MUCH more relative data nowadays, and it really helps.

 

Your YY example is actually pretty good. I know it was not a popular change, but the ship was overperforming significantly, on Daring level, and there's the question: why YY was nerfed and Daring wasn't? Well, because we pay close attention to relative stats.

 

For instance, we have a tool that allows us to see player efficiency distribution for a particular ship. Basically it looks like this (ship names specifically omitted, that's just an example of one ship group "Tier+class"): https://i.gyazo.com/e6ee5b18d21f14e70a5f52ccacc6b794.png

 

Each curve corresponds to one ship, and in this particular case we can see the account WR distribution. Do you see this one ship shifted heavily to the right? Its players are really better than other ships' players - it can be either a test ship (CCT + ST + clan test + staff generally plays better) or some hardcore ship (owners of Stalingrad, Flint, Black, etc are generally much better), for example. And when evaluating ship's efficiency we take this distribution into account. Speaking of YY and Daring, they both were overperforming, but while Daring's audience "skill curve" is significantly better, YY skill curve was more or less in line with other T10 DDs. Same players showing better results = overperforming, means the ship "buffs" its players and allows them to show much better results than they can achieve on other ships. If we theory-craft a bit more, I'm pretty sure if we replace Shimakaze (most popular T10 ship with very average players skill curve) audience with Daring audience, Shima will look OP as well - due to various reasons these guys just play better.

 

But that's not just it. We have special tools for making custom slices of all main performance metrics: winrate, frags, relative and absolute damage, lifetime and survivability, capture points in a ship group.

 

Let's take Kremlin balancing, for example. Obviously we cannot balance it by avg.stats due to the same audience difference - hence we create a slice where we can see all combat efficiency curves relative to players. So we compare Kremlin player's results with their results on other T10 battleships / results of players with similar skill. E.g. 150 k avg damage (not real stats, just an example) looks scary, but what results these players / players of the same skill show on other T10 BBs?

 

We answer such questions before we make a decision.

 

The only challenge here is to get enough battles for certain ships, as all systems like that require quite A LOT of data to be precise.

 

I'm honestly not sure we can go into such details for any announcement we make to general audience (I will think about your question and argument more, though). But what I really want to say: when we claim "this ship is overperforming" or "it is underperforming" we typically give it A LOT of research beforehand. It's not just "look up avg.numbers and scream" :-)

 

 

Leo "Apollo11"

  • Cool 11

Share this post


Link to post
Share on other sites
Players
2,169 posts
20,242 battles

Ah so all is well and perfectly legit with Kitakaze, Harugumo, Conquerer... that's really good to know. Phew. What a weight off of my chest.

 

This is the absolute Mueller report of WG balancing. No seriously.

  • Cool 4
  • Boring 3

Share this post


Link to post
Share on other sites
[BAZI]
Beta Tester
2,851 posts
10,968 battles
36 minutes ago, ColonelPete said:

And I thought they did that since ages ago.

Me too. I know they used this "mechanic" for tanks since at least 2015.

Share this post


Link to post
Share on other sites
[BYOB]
[BYOB]
Players
4,463 posts
19,573 battles

I said it elsewhere and I will say it here, this doesn't add up. If they have such improved tools to pinpoint and identify performance issues, then why are they balancing things with a sledgehammer? The YY nerf was massive, which shouldn't be necessary. They keep saying Blyskawica is alright, while people keep telling them the ship is boring and inferior to Leningrad, with public stats proving the players right. I don't believe them, either their tools aren't as good as advertised or their personnel handling them is not competent enough. Or they are straight up lying, which is also a possibility.

  • Cool 11

Share this post


Link to post
Share on other sites
[SHAFT]
Players
9,424 posts
8,803 battles
1 hour ago, Nautical_Metaphor said:

Ah so all is well and perfectly legit with Kitakaze, Harugumo, Conquerer... that's really good to know. Phew. What a weight off of my chest.

 

This is the absolute Mueller report of WG balancing. No seriously.

 

In b4 Conqueror buff? :Smile_hiding:

Most of the time, Conquerors have worse WR than corresponding TX ships... even good players.

 

 

Share this post


Link to post
Share on other sites
[TTTX]
[TTTX]
Players
4,608 posts
8,081 battles
19 minutes ago, DFens_666 said:

 

In b4 Conqueror buff? :Smile_hiding:

Most of the time, Conquerors have worse WR than corresponding TX ships... even good players.

 

 

yup, because the ship is netter than its peers at exactly one thing: farming irrelevant (because repairable) damage. That doesn't win games, it just shows up in average damage (particularly for the lower-skill "LOL HE SPAM FROM 20KM LOL" faction). So from that perspective, Conqueror isnt actually overpowered, it doesnt need a nerf based on stats/performance.

Doesn't mean it isnt crap design and sucks to fight against, mind you... And those factors don't show up in stats. What it'd need would be a bit of a redesign (of the entire line)...

  • Cool 6

Share this post


Link to post
Share on other sites
[SHAFT]
Players
9,424 posts
8,803 battles
Just now, Tyrendian89 said:

Doesn't mean it isnt crap design and sucks to fight against, mind you... And those factors don't show up in stats.

 

Thats prolly why WG doesnt understand it when we bring stuff like this up here :Smile_teethhappy:

Reminds me of the Conq we had couple of days ago... started to brag how we need 4 of us to kill him :cap_fainting:

  • Cool 1

Share this post


Link to post
Share on other sites
[FJAKA]
Players
2,975 posts
477 battles
2 hours ago, Nautical_Metaphor said:

Ah so all is well and perfectly legit with Kitakaze, Harugumo, Conquerer... that's really good to know. Phew. What a weight off of my chest.

 

This is the absolute Mueller report of WG balancing. No seriously.

finally you know why after 8k games you have 700 PR....damn those pesky OP sips that hinder your glory

  • Funny 3

Share this post


Link to post
Share on other sites
[NWP]
Players
1,331 posts
9,813 battles

Dear All,

 

51 minutes ago, Aragathor said:

I don't believe them, either their tools aren't as good as advertised or their personnel handling them is not competent enough. Or they are straight up lying, which is also a possibility.

I believe that they are putting in a serious amount of work to honestly balance the ships available to the players.

 

I will be a bit philosophical to your great dismay.

 

Unfortunately, entropy will not allow a closed system to be in absolute equilibrium. The Game (WoW) is a closed system. As such, the Game tends to maximize its entropy (the amount of disorder/lack of "balans"). It will inexorably degenerate toward a state of greater disorder, greater entropy. Locally, entropy can be lowered by external action, e.g. nerfing, buffing etc., where the entropy is being reduced (balancing one ship - local) and order is increased in one area of the game. This local increase in order is, however, only possible at the expense of an entropy increase in the surroundings; here more disorder must and will be created (un-balancing other ships). The complexity of the system is such that no matter how honestly and rigorously they try to balance everything (lower the entropy of the system) they will not have an absolutely balanced game (a system of low entropy).

 

Some ships will always be better than others.  Furthermore, any effort to balance one ship will reflect to the loss of balance of another ship. 

 

In this direction, the second law of thermodynamics, as famously enunciated by Rudolf Clausius in 1865, states that: “The entropy of the universe tends to a maximum.”

 

And just to paraphrase Fido Dido "Balans is boring"

 

245b21a928139f79eb246003f10f70ab--fido-dido-skate.jpg.17e75a3cd3c021cc3b73c1c852eed94a.jpg


Regards

Saltface

  • Cool 2

Share this post


Link to post
Share on other sites
[FJAKA]
Players
2,975 posts
477 battles
6 minutes ago, Saltface said:

Dear All,

 

I believe that they are putting in a serious amount of work to honestly balance the ships available to the players.

 

I will be a bit philosophical to your great dismay.

 

Unfortunately, entropy will not allow a closed system to be in absolute equilibrium. The Game (WoW) is a closed system. As such, the Game tends to maximize its entropy (the amount of disorder/lack of "balans"). It will inexorably degenerate toward a state of greater disorder, greater entropy. Locally, entropy can be lowered by external action, e.g. nerfing, buffing etc., where the entropy is being reduced (balancing one ship - local) and order is increased in one area of the game. This local increase in order is, however, only possible at the expense of an entropy increase in the surroundings; here more disorder must and will be created (un-balancing other ships). The complexity of the system is such that no matter how honestly and rigorously they try to balance everything (lower the entropy of the system) they will not have an absolutely balanced game (a system of low entropy).

 

Some ships will always be better than others.  Furthermore, any effort to balance one ship will reflect to the loss of balance of another ship. 

 

In this direction, the second law of thermodynamics, as famously enunciated by Rudolf Clausius in 1865, states that: “The entropy of the universe tends to a maximum.”

 

And just to paraphrase Fido Dido "Balans is boring"

 

245b21a928139f79eb246003f10f70ab--fido-dido-skate.jpg.17e75a3cd3c021cc3b73c1c852eed94a.jpg


Regards

Saltface

Also BALANCE is very stretch word. Balanced for what?

 

Lets take Yamato or Conqueror....great shisp, very strong in randoms....almost useless in any competitive play 

Grozovoi - again one of best random dds......comeptitive play....again not used very much

 

and so on

 

i must admit that WG in last releases did good job balancing ships and not creating powercreap.

 

last line where they fucked things was RN BBs, or better say conqueror. (where bigger objection was play style than relative strength) 

 

 

  • Cool 1

Share this post


Link to post
Share on other sites
Players
1,133 posts
7,085 battles
38 minutes ago, veslingr said:

Also BALANCE is very stretch word. Balanced for what?

 

Lets take Yamato or Conqueror....great shisp, very strong in randoms....almost useless in any competitive play 

Grozovoi - again one of best random dds......comeptitive play....again not used very much

 

and so on

 

i must admit that WG in last releases did good job balancing ships and not creating powercreap.

 

last line where they fucked things was RN BBs, or better say conqueror. (where bigger objection was play style than relative strength) 

 

 

well, I watch streams of competitive play when I can, and I often see Yamatos and Grozovoi.

While I'll admit Conqueror is rarely seen, it seems that for battleships it's either Yamato or Republique, and for DDs it's either Gearing or Grozovoi.

Share this post


Link to post
Share on other sites
[CATS]
Players
26,032 posts
14,064 battles

Newsflash: Competitive 7v7 =/ Random 12v12

 

You cannot balance ships for both, unless you give ships dfferent stats for each gamemode.

Share this post


Link to post
Share on other sites
[FJAKA]
Players
2,975 posts
477 battles
20 minutes ago, elblancogringo said:

well, I watch streams of competitive play when I can, and I often see Yamatos and Grozovoi.

While I'll admit Conqueror is rarely seen, it seems that for battleships it's either Yamato or Republique, and for DDs it's either Gearing or Grozovoi.

in CW (i played every season) Yamato is to slow (and his forte is big pen that mostly hurts other BBs) and Conqueror is BB killer (and with 1 BB per side rule Conq is in problem). Montana in THE choice of 99% top teams. (mobility + 12 guns that are CA killers)

 

Same is with gearing and legendary module.

 

but aside that, i think most of us agrees that if there are no best choice for tournaments, randoms and CW we have solid balance

 

 

Share this post


Link to post
Share on other sites
Players
2,169 posts
20,242 battles
Vor 2 Stunden, veslingr sagte:

finally you know why after 8k games you have 700 PR....damn those pesky OP sips that hinder your glory

No that's just because I don't give a runny excrement about stats.

It's called "normal self-esteem".

You get it only if your mummy really loves you though. If not, then you later become a stat-shamer tryer.

Share this post


Link to post
Share on other sites
[FJAKA]
Players
2,975 posts
477 battles
1 minute ago, Nautical_Metaphor said:

No that's just because I don't give a runny excrement about stats.

It's called "normal self-esteem".

You get it only if your mummy really loves you though. If not, then you later become a stat-shamer tryer.

"normal self-esteem" in your case looks like - in every thread you cry about something....OP ships, CVs, Radars, hydros and so on......ever try self criticism? 

Share this post


Link to post
Share on other sites
[GRNPA]
Beta Tester
1,296 posts

Wow, Lesta discovered WR curves, something WoT players have been doing on their own websites since 2014.

 

Applause. Their incompetence truly knows no bounds.

Share this post


Link to post
Share on other sites
Players
6,753 posts
245 battles

R.i.p poi 15km toprs 57knots and 2.5km detection.

 

and some really werid 1x1 2x2 gun layout as well.

 

yoshi is like a nerfed pre nerf azuma lol.

 

gg balans.

Share this post


Link to post
Share on other sites
Players
2,169 posts
20,242 battles
Vor 16 Minuten, veslingr sagte:

"normal self-esteem" in your case looks like - in every thread you cry about something....OP ships, CVs, Radars, hydros and so on......ever try self criticism? 

You see, of course I value the incisive input that you, one of my most valued forum members have gained by a cursory glance at my stats.

 

And I am more than grateful for you sharing it in such a well-phrased and constructive manner.

 

That's of course also because I am so highly interested in you as a person, and not just any person, but a person - dare I say it? that I so greatly respect, nay admire for their sheer wisdom, prowess, and the near erotic appeal of their polished prose. If, alas! from afar. <sigh> (not to mention for the sheer girth of their genitalia.)

  • Funny 3

Share this post


Link to post
Share on other sites
[OM-S]
Beta Tester
1,267 posts
24,445 battles

to balance on stats alone, even with this extended approach, there will be flaws that need constant adjusting to new metas.

as in one meta one ship is overperforming because of its nature. it was build to have this features. when you now balance it, it will struggle as soon the meta shifts to a different direction. means you have to balance again.....

 

and so on and so on

 

the YY example is perfect for this. i doubt the YY is overperforming now. i bet the exact opposite is the case. result is, that we dont see many YY at all atm.

abd because of the CV cancer we have a steep incline of 20km shimas on the servers. because for many it is the only way to play a DD in this meta.

 

and montanas? we did not see many in KOTS  for example. her previous good AA is no more and does not grant any protection. so why not switch to either fast or tanky BBs like repus and yamatos? and thats exactly what we saw.

you cannot prevent the OP CV strikes anyway.

 

and while we saw many zao or hindenburg in previous kots, they are now totally out of the game in competitive. hinden has CRAP AA (previously one of the best) ad zaos concealment means CRAP now with the omnipresent CVs that just f*** the zao in 2 bomb runs.

 

the max range snipers with good DPM or accuracy are the meta now. making all ships overperfom here now. henrys, yamatos (with legendary), moskvas, stalins.

as soon you balance them and the meta changes again, these will underperfom suddenly.

 

so i see the WG approach a bit ambivalent

 

so nerf these ships?! no.......nerf the ship that is causing the actual meta. in this case, the CVs in general and their potential.

  • Cool 10

Share this post


Link to post
Share on other sites
[GRNPA]
Beta Tester
1,296 posts
8 minutes ago, IceyJones said:

............ the max range snipers with good DPM or accuracy are the meta now. ................

 

Yep, WG managed to turn CVs from super OP to just OP while also turning the meta into cancer.

  • Cool 1
  • Funny 1

Share this post


Link to post
Share on other sites
Weekend Tester
42 posts

YY was pretty much a Gearing 2.0, her noticeable differences the lower detection of DW torps (while unable to hit DD) and the option to exchange Smoke with Radar. If YY was really OP looking at those would have been logical. Instead her gun and torp reload was nerfed giving her a significant disadvantage in all circumstances even if Radar or DW are not applicable. Maybe WG has the tools to find overperforming ships, yet it doesn't stop them from making poor decisions how to deal with them.

Share this post


Link to post
Share on other sites

×