Thoughts on player skill vs BR

[Real_K_Soze]

@AlvisWisla

@tensilaspider

There are situations where a majority of players that don’t know how to play a vehicle will get it lowered in br

Not possible. Cause those players will be equally skilled in all their vehicles thus it would be statistically identical to all other players.

You might consider to think about what you wrote.

If your statement would be true a large portion of the forum content (related to undertiered / op vehicles) would not exist.

[AlvisWisla]

Almost all under-BR’d vehicles exist due to compression above.
The statement in my post being true is why there are so much forum content.

[Real_K_Soze]

@AlvisWisla

Almost all under-BR’d vehicles exist due to compression above.

It doesn’t matter why vehicles are under-BRd - the fact that they exist shows a severe weakness in gaijin’s BR policy.

In order to show that this statement:

@tensilaspider

There are situations where a majority of players that don’t know how to play a vehicle will get it lowered in br while vehicles played by good/great players have a higher br. (Look at copy/paste vehicles)

is 100% correct go to:

StatShark.net: See All War Thunder Player, Missile, and Vehicle Statistics

and look (Air mode) just for the 3 P-51 Cs (there is also a French and SWE version) and compare the results for the US/CHN/JP versions.

You might agree that KpS and KpB vary extremely. There is no reason why all 3 have the same (= too low) BR.

[AlvisWisla]

Statshark and Gaijin are entirely different systems.
You cannot get the data that Gaijin uses for balancing from Statshark with any automated means.

Not sure why you’re citing Statshark when it proves all my posts correct.
Gaijin does not use aggregate blind data and hasn’t since the CL13 incident, which is why CL13 has never been repeated.

[MotorolaCRO]

@Real_K_Soze

That the 5-10% of the top performers in vehicle A might be totally different from the 5-10% top players of vehicle B is obvious.

Which is why I’ve said that system is flawed, as you’re here to balance vehicles A and B.
A vehicle has to be compared to it’s peers in order to see where it really belongs.

@Real_K_Soze

And as a final remark regarding skill: It doesn’t matter how you get results - just the results count.

It definitely does matter how you get the results as that will directly impact the balancing.
I already shown this to you in another thread.

@Real_K_Soze

If your statement would be true a large portion of the forum content (related to undertiered / op vehicles) would not exist.

That is under assumption that the entirety of the forum is unbiased, fact-driven community.
Just looking at the case of 2S38 proves this assumption to be wrong.

@Real_K_Soze

It doesn’t matter why vehicles are under-BRd - the fact that they exist shows a severe weakness in gaijin’s BR policy.

Not a single balancing system is enough to defeat the BR compression, where the amount of BR steps is too low to properly fit in hundreds of vehicles at the same time.

[Real_K_Soze]

@AlvisWisla

Not sure why you’re citing Statshark when it proves all my posts correct.

Just check this thread:

BI at 6.7 is April Fools event,. only it's going 24/7 for the last few years

Seriously, why is that thing 6.7. No other plane at 6.7 or even 7.3 has ANY chance in hell of ever taking it down. The best one can do is stay entire match at 700+km/h. If you ever get caught - you’re dead. You try to follow him? You’re dead. It has fuel for ages due to magical engine efficiency. Cannons are good enough to 1-click people out of existence. It turns WAY better than a Zero. Perfectly balanced. There’s literally no counter to this plane other than “have 3-4 disciplined teammat…

Regarding this:

@AlvisWisla

You cannot get the data that Gaijin uses for balancing from Statshark with any automated means.

My example with the 3 P-51Cs was clear enough and you just assume that gaijin would set BRs solely based on hard data - i wrote earlier:

@Real_K_Soze

And takes the plain average from their results. In addition BRs are used to push sales of certain premium vehicles and are (together with the MM) a tool to steer WRs on nation level.

---

@MotorolaCRO

@Real_K_Soze

That the 5-10% of the top performers in vehicle A might be totally different from the 5-10% top players of vehicle B is obvious.

Which is why I’ve said that system is flawed, as you’re here to balance vehicles A and B.
A vehicle has to be compared to it’s peers in order to see where it really belongs.

Now i understand your pov - you simply skipped the first step in a complex process.

Step 1: Ensure that the current BR is adjusted on top performing players
Step 2: Ensure that comparable vehicles can fight each other on equal terms

So whilst i was talking about Step 1 you always talked about Step 2. No wonder that we had divergent opinions. Lmao - 😂👍

@MotorolaCRO

That is under assumption that the entirety of the forum is unbiased, fact-driven community.

That should be the minimum standard for each thread.

@MotorolaCRO

@Real_K_Soze

It doesn’t matter why vehicles are under-BRd - the fact that they exist shows a severe weakness in gaijin’s BR policy.

Not a single balancing system is enough to defeat the BR compression, where the amount of BR steps is too low to properly fit in hundreds of vehicles at the same time.

Mate! You argue that BRs steps are too low due to compression. Gaijin devs are in charge of the BR steps and can change this at any point in time.

I wish both of you a great day!

[Leinadmix9_ツ]

@AlvisWisla

Gaijin does not use aggregate blind data and hasn’t since the CL13 incident, which is why CL13 has never been repeated.

Explain this one for me then:

[MotorolaCRO]

@Real_K_Soze

Step 1: Ensure that the current BR is adjusted on top performing players
Step 2: Ensure that comparable vehicles can fight each other on equal terms

So whilst i was talking about Step 1 you always talked about Step 2. No wonder that we had divergent opinions.

Those two steps are very much connected. You first have to put a vehicle somewhere in order to gauge it’s performance to it’s same BR’d peers and then make changes based on that.

@Real_K_Soze

That should be the minimum standard for each thread.

Yeah it should in a perfect world, but WT forum can get absolutely wild when talking about certain topics, so bias and subjective opinions are a big part of this community.

@Real_K_Soze

Mate! You argue that BRs steps are too low due to compression. Gaijin devs are in charge of the BR steps and can change this at any point in time.

I argue that problems with balancing lies in the fact we don’t have enough BR steps, not because our balancing system that relies on average stats is bad.

Gaijin could add more BR steps and thus solve many balancing issues, but they won’t for reasons they don’t want to disclose.

[ron_23]

@tensilaspider

Here’s a simple graph cause why not.

i also made a graph that compares playerskill vs BR

[Lauri_Heino]

@Cookies_Juicebox

There are many level 100 players too that play horribly like new players.

Honestly, this does make me wonder about how player level is / should be calculated / earned.

[Ion_492]

It’s just based on the total RP you earn.

[Lauri_Heino]

Honestly, I’m not 100% sure I like that.
For example, I’m level 26 (about a third to 27), and I really don’t think my stats reflect that.

The question, I guess, is what then could you base user level on…

[Bonrath]

The line should not go up at 13.0 btw. It’s already too high at 13.0 as well

[Beeschurger]

I dont see sharp drop to 0 or even below at 10.3/10.7.

7/10 not enough data.

[ron_23]

that is 13.7 / 14.0 where it goes up

[AlvisWisla]

@Real_K_Soze
You don’t change BRs based on players being good or not.

You only change BRs based on if that vehicle gets players more results than other vehicles; and you check this on a player-player basis as per the scientific method to mostly eliminate the skill disparity variable.

@Leinadmix9_ツ When a BR is so decompressed that similar or same-performing aircraft can be different BRs… that’s just an evidence of that BR’s health.

I see Creastroy is telling others to lie… wild.

[MapleVII-live]

This graph is accurate as past 10.0 ground is just 1.0 with more modern vehicles in terms of playstyle. (Save for CAS and SPAA)

[Real_K_Soze]

@AlvisWisla

@Real_K_Soze
You don’t change BRs based on players being good or not.

You only change BRs based on if that vehicle gets players more results than other vehicles; and you check this on a player-player basis as per the scientific method to mostly eliminate the skill disparity variable.

Mate! Respect! As usual your posts are pure gold! You should really consider a career as stand-up comedian - always a pleasure to read your posts 😂👍

Just re-read this official statement regarding the setting of BRs:

Responding To Your Feedback On Separate Battle Ratings

…entirely and you might see that gaijin itself states that the average performance (= generating player income = SLs/RPs) is major steering factor in the BR determination process - the comparison with similar aircraft plays a role too - but is a secondary factor.

A few examples:

@Stona_WT

Battle Ratings are decided based on how much a vehicle earns, but this is not purely economical. We use this metric because it’s all encompassing and considers every action a player makes with their vehicle, so this considers frags, assists, caps, and effectively every useful action a vehicle can perform all bundled into one universal metric.

@Stona_WT

This is the “Efficiency” of each vehicle and gives us a very well rounded perspective on all of its abilities and how it’s actually being used.

If a vehicle has high efficiency, it’s outperforming its contemporaries in multiple ways the majority of times it spawns on the map, and as a result may have to be increased in Battle Rating. Whereas a vehicle with low efficiency is not performing well across the board against what it fights, and may be moved down. However this is not purely a data driven process, we often consider additional factors such as the volume of players using a certain vehicle, its lineup, new features that may be altering performance in different ways etc — and this often leads us to delay a change we otherwise would have made to gather more information. Even though we do primarily go by this efficiency metric, we aren’t bound by it, and spend a lot of time each Battle Rating cycle to look over feedback for different perspectives and elements we may not have initially considered.

The translation of this is obvious:

1. Gaijin has a clear goal about the SL/RP gains for their vehicles. These goals may differ across branches (air, ground, naval) or game modes (AB/RB/SB) but these efficiency goals are a major driver for the BR.

2. If a vehicle performs on average too well - the BR will be increased - too low we see a decrease of the BR.

That’s why any discussions about BRs in this forum circle around the different understanding of efficiency.

• Players talk usually about their experiences (usually / at least often based on extensive experiences with a certain vehicle) whilst gaijin needs a cheap and easy tool to steer player income (SL/RP) which fulfills also the role of steering the game & research progress.

That’s why the expression skill (in the title of this thread) is actually rather misleading - skill is already included in the results - or “efficiency” like gaijin calls it.

So we are back at square one:

@tensilaspider

While a decent system, There are situations where a majority of players that don’t know how to play a vehicle will get it lowered in br while vehicles played by good/great players have a higher br. (Look at copy/paste vehicles)

@AlvisWisla

Gaijin does not use aggregate blind data and hasn’t since the CL13 incident, which is why CL13 has never been repeated.

Seriously???

I refresh your memory: One of the biggest seal clubber aircraft at prop BRs (the JP Bf-109 E-7) at 3.0 was for ages downtiered to 2.3 because one of their interns classified the plane as an E-3 (BR 2.3) Instead of having almost identical performance with the far superior German E-4/E-7.

@AlvisWisla

When a BR is so decompressed that similar or same-performing aircraft can be different BRs… that’s just an evidence of that BR’s health.

Seriously (Part 2)???

1. Technical performance (speed/climb/turn/firepower) in the example of @Leinadmix9_ツ prove just one thing: The JPpremium F4U-1 performed too well (SL/RP) even for a premium and received a higher BR. The US F4U-1 versions have all different SL/RP multipliers.

2. Your current example with 109s and Hurricanes BR diversity just shows different efficiencies based on different armaments and different flight characteristics…

3. My initial example with the 3 P-51 Cs ( US/CHN/JP) prove that the last two benefit from being a premium - so despite they perform way better at rank III they keep the objectively too low BRs - whilst the poor US players have the same plane just at Rank II…

Let’s get back to the topic: How can BRs be adjusted in order to avoid severely undertiered vehicles?

[Leinadmix9_ツ]

@AlvisWisla

@Leinadmix9_ツ When a BR is so decompressed that similar or same-performing aircraft can be different BRs… that’s just an evidence of that BR’s health.

That’s just an evidence of how negligent Gaijin is when it comes to BRs changes.

[tensilaspider]

@Real_K_Soze

That’s why the expression skill (in the title of this thread) is actually rather misleading - skill is already included in the results - or “efficiency” like gaijin calls it.

What would be a better name you think? Thoughts on vehicle efficiency vs br?