Jump to content


New WG rating.


  • Please log in to reply
120 replies to this topic

brumbarr #1 Posted 12 September 2017 - 11:40 PM

    Lieutenant General

  • Player
  • 38626 battles
  • 6,326
  • Member since:
    07-30-2012

As I posted in a thread some time ago, WG is making a new rating system and hall of fame:

 

https://thedailyboun...l-of-fame-beta/

 

While it looks good overal, I do have a few problems with it:

-survivial rate should not be a factor in any rating system, how much you survive is irrelevant to how good you play. All that matters is the results, not how you got there, an aggresive player who gets like 3K dpg but only survives 20% of matches should get the exact same rating as a more passive player who also gets 3k dpg but surives 40%.

 

-same thing with blocked dmg, blocked dmg in itself is useless, blocking dmg is a tool to get a result, its a tool to get more dmg and win the game. Its not a result in itself. 

A maus player who sits in the open for no reason and gets the same dmg as a maus player who only peaks around the corner when his gun is reloaded should get the same reward.

 

- Every vehicle should get the same % weighting of factors. Imagine a scout player who  focuses more on dealing dmg.  He does 2K dmg per game and 1K spotting, then another scout player who does 1K dmg and 2K spotting. Both contribute just as much to the match, so they should get the same rating, if you play a vehicle different than how it is supposed to be played but are succesfull  in it you shouldnt be labeled lower skilled.

 

I would completely remove survival rating and heavily reduce the % of blocked dmg to the rating, at max 15% . I would make the weighting class based, not tank based, for example:

Heavy tanks: 15% blocked, 15% kills, 40% dmg, 30% assist.

Meds: 5% blocked, 15% kills, 45% kills , 35% assist.

Lights: 0% blocked, 15% kills, 45% dmg, 40% assist.

TDs:  5% blocked, 15% kills, 40% dmg, 40% assist.

Arty:  100% ammount of shots missed

 

 

 


Edited by brumbarr, 12 September 2017 - 11:43 PM.


laulaur #2 Posted 12 September 2017 - 11:50 PM

    Warrant Officer

  • Player
  • 46460 battles
  • 986
  • [-WBZ] -WBZ
  • Member since:
    08-11-2011

I agree on everything you said.

WG have so many bad ideas lately, why should anyone be rewarded for surviving the battle or blocking damage??? :amazed:

This is plain stupid, they reward the passive players.

 



brumbarr #3 Posted 12 September 2017 - 11:56 PM

    Lieutenant General

  • Player
  • 38626 battles
  • 6,326
  • Member since:
    07-30-2012

View Postlaulaur, on 12 September 2017 - 11:50 PM, said:

I agree on everything you said.

WG have so many bad ideas lately, why should anyone be rewarded for surviving the battle or blocking damage??? :amazed:

This is plain stupid, they reward the passive players.

 

 

I guess everything is better than  PR , or even wn8 at this point :p

 

 

EDIT: nvm, wn8 will still be better, since both stats are the same concept, except wn8 leaves out the nonimportant stuff.


Edited by brumbarr, 12 September 2017 - 11:58 PM.


ZlatanArKung #4 Posted 13 September 2017 - 12:03 AM

    Major General

  • Player
  • 1529 battles
  • 5,112
  • Member since:
    12-20-2014
It should include WR since every metric that doesn't include it is bad.

Btw, I prefer PR over wn8. Less padders, can't be padded at low tiers. So.

brumbarr #5 Posted 13 September 2017 - 12:10 AM

    Lieutenant General

  • Player
  • 38626 battles
  • 6,326
  • Member since:
    07-30-2012

View PostZlatanArKung, on 13 September 2017 - 12:03 AM, said:

It should include WR since every metric that doesn't include it is bad.

Btw, I prefer PR over wn8. Less padders, can't be padded at low tiers. So.

 

I disagree strongly. The whole point of making any rating at all is to make a rating independent of WR that converges much faster.   So you can compare performance after a few dozens of battles instead of a few hundred. 

And including it would be pointless,  why include it if you can jsut look at the WR  and  rating seperatly?

 

Any rating that is craphas no padders, if your rating is good, padders will arise.

 

Lets also not forget that PR  being an account overal rating makes it useless from the getgo. Noone is the smae player they once sttarted at. And the point of a rating is to look at small samples like tank performance, which PR doesnt do.

PR is pointless, if you want an account rating, look at WR.


Edited by brumbarr, 13 September 2017 - 12:11 AM.


ZlatanArKung #6 Posted 13 September 2017 - 12:13 AM

    Major General

  • Player
  • 1529 battles
  • 5,112
  • Member since:
    12-20-2014

View Postbrumbarr, on 13 September 2017 - 12:10 AM, said:

 

I disagree strongly. The whole point of making any rating at all is to make a rating independent of WR that converges much faster.   So you can compare performance after a few dozens of battles instead of a few hundred. 

And including it would be pointless,  why include it if you can jsut look at the WR  and  rating seperatly?

 

Any rating that is craphas no padders, if your rating is good, padders will arise.

 

A good rating is not paddable.

 

A guy with 3k avg dmg and low WR is a worse player then a guy with 2500 avg dmg and high WR (all else equal).

 

A rating that doesn't take the single most important stats into consideration is flawed. Damage is useless if it doesn't lead to a win.

 

If you don't include WR you will still have to look at WR to see wether the player is good or not.

 



fighting_falcon93 #7 Posted 13 September 2017 - 12:20 AM

    Lieutenant Сolonel

  • Player
  • 29717 battles
  • 3,418
  • Member since:
    02-05-2013

Well think of it this way:

 

- Player A does 3000 average damage per game and survives 50% of his battles.

- Player B does 3000 average damage per game and survives 25% of his battles.

 

Player A is obviously a better player since he manages to stay alive and keep his gun in the fight in 25% more of his battles. This basically means that if there were more HP on the enemy team, or of someone on his team messed up and did less damage, player A still has the potential to do more damage, compared to player B that has got himself killed.

 

I agree that it's basically the same thing, because both players contributed equally before the round ended. However IMHO it's still worth including survival rate in a proper rating value, because player A manages to do the same damage as player B, but still keep his gun in the fight after that damage has been done.

 

Same applies to blocked damage:

 

- Player A does 3000 average damage per game and blocks 3000 potential damage.

- Player B does 3000 average damage per game and blocks 1500 potential damage.

 

Comparing these 2 players I'd also say that player A appears to be more skilled than player B. Even if the resulting damage is the same, player A still manages to soak up twice as much damage and still pump out the same amount of damage. The blocked damage is shells that could have potentially caused damage to allied tanks, so it does actually contribute to the victory in the form of HP conservation.

 

Personally I'd actually say that WG should include even more parameters into their ranking, such as:

 

- Winrate: The main goal of the game, no point in doing damage unless you win.

- Tier: It takes more skill to play well on higher tiers.

- Accuracy: Hitting and penetrating more of your shots allows you to more reliably cause damage.

- Capture/Defense: Most boring way to win, but it's still the victories that count.

- Spotting: Lighting up targets will allow your allies to cause damage without spotting by themselves.

 

It's also worth to note that I think it's a mistake to include all these parameters for all classes. For instance, LTs should be measured mostly by spotting/assisting/survival, HTs should be measured more by blocking/damage/frags, TDs should be measured more by accuracy/damage/defense, and so on.


Edited by fighting_falcon93, 13 September 2017 - 12:24 AM.


_EXODUZ_ #8 Posted 13 September 2017 - 12:25 AM

    Lieutenant

  • Player
  • 34617 battles
  • 1,936
  • [RGT] RGT
  • Member since:
    11-05-2014
Meh, whatever... a good player will always be a good player. Numbers are quite meaningless.

laulaur #9 Posted 13 September 2017 - 12:36 AM

    Warrant Officer

  • Player
  • 46460 battles
  • 986
  • [-WBZ] -WBZ
  • Member since:
    08-11-2011

*edited*


Edited by Daxeno, 28 September 2017 - 09:36 AM.
This post has been edited by the moderation team due to inappropriate remarks.


Mike_Mckay #10 Posted 13 September 2017 - 12:48 AM

    Second Lieutenant

  • Player
  • 18668 battles
  • 1,488
  • [-AWF-] -AWF-
  • Member since:
    09-02-2015

I dunno tbh


I guess that how much each thing affects the rating is more important than whether its included or not


For example if survival rate made up 50% of the rating that would be dumb, but if it only made up say 2 or 3% that would be fine as long as other things such as damage done and assisted damage could easily outweigh the benefit of merely surviving so that just hiding would be detrimental compared to fighting. But if someone fights AND survives whilst someone else does the exact same damage etc why shouldn't that person get something extra for that? The reward itself wouldn't be the issue, just how MUCH of a reward it was

Blocked damage is a bit more tricky, a tank that is basically a block of steel on wheels doesn't really have to do much to bounce shots whereas something with angled thinner armour "could" be related to skill, or could merely be a noob like me happening to park the vehicle at a lucky angle or another noob like me not selecting the correct ammo or knowing where to shoot

So firstly the system "should" differentiate between shots that "could" have penetrated at a different angle and ones with no chance of penetrating under any circumstances and ignore the later. But how would  it then know if the remaining shots were due to skill or pure luck? And should that even matter too I guess, as making an enemy player waste shots is beneficial regardless of circumstances I suppose but ideally if there even was a way to differentiate it should only be rewarding the deliberate use of armour not pure chance

Failing that then only shots that "could" have penetrated not just all shots including ones that never could have on the part of the tank that hit

The spotting thing I do agree with to some extent, although maybe make it work like the XP/silver system where spotting gives you half and the short half, damage just gives you all of it if someone else didn't do the spotting

But like with bouncing, how far do you go? Should a tank with a 450 view range that is top tier on a battlefield where the average get the same for spotting as one with only 390 metre view range on a field with an average of 435? One of those would, or "could" have taken a lot more skill than the other so should they be rewarded equally or based on level of difficulty?

I also don't think you can really break down ratios by tank class either, some mediums, TDs and even lights have thicker armour than heavies at the same level, penetration and damage varies MASSIVELY, so should you reward 2k damage done with a gun with 240 alpha more than one with 750 alpha because the two vary greatly, view range also differs quite a bit between vehicles but to get a real idea of what work or skill was involved to "some" extent how the view range compares to the average viewranges of both teams would be worth considering and perhaps how close you were and whether you were spotted yourself to avoid doing a suicidal charge into enemy ranks paying off perhaps

And I guess something else might be what shots that "would" have hit you actually hit buildings and terrain? That "could" be due to skilful use of the surroundings or could be dumb luck too but its not really that different to bouncing shots either in some ways as playing arty safe and going hull down or peeking over ridges are "skills" too which will cause that to happen frequently
 



Jigabachi #11 Posted 13 September 2017 - 01:27 AM

    Field Marshal

  • Player
  • 17858 battles
  • 18,478
  • [T-D-U] T-D-U
  • Member since:
    08-12-2011
Every rating that tries to condense a whole set of diverse information into a single number sucks by default. That rating will be as useless as the previous one and as useless as WNx.

SuedKAT #12 Posted 13 September 2017 - 02:41 AM

    Lieutenant General

  • Player
  • 12154 battles
  • 6,284
  • [T-D-U] T-D-U
  • Member since:
    08-21-2014

So they have basically taken a bit of what third party stats utilize and thrown in stuff they feel are important parameters when determine "skill" in WoT. I kinda like the idea, but looking at what they have included and what they've left out as well as how large/small part it plays in the equation kinda makes me feel a bit confused. 

 

Having a rating built on survival rate is just wrong on so many levels (not to mention the amount, 27% for the Bat as an example), I fail to see how that can be relevant to any rating and what a high versus low survival rate would show, sure if you consider it on it's own you can draw some conclusion about how a player play his tank but that's about it. I'd replace it with Win Rate instead, then you at least measure things that directly impact a match to some extent, if it's won or not, the amount of damage and blocked damage, assists and kills.

 

I'm a bit torn on having different percentages on different tanks, since as you say it don't really matter what tank your in, if you contribute in some way it should show up in the ratings. On the other hand utilizing your tank for "what WG intended you to do" might also be nice to measure since getting a bit lower rating for not spotting in your LT or not blocking any damage in your HT most likely means you play the tank a bit "wrong" so to speak, you know those IS-7 sniping players etc.



Homer_J #13 Posted 13 September 2017 - 03:27 AM

    Field Marshal

  • Beta Tester
  • 27652 battles
  • 29,000
  • [WJDE] WJDE
  • Member since:
    09-03-2010

View PostZlatanArKung, on 13 September 2017 - 12:13 AM, said:

 

If you don't include WR you will still have to look at WR to see wether the player is good or not.

 

 

The only reason you would include win rate is to fudge the figures so the two follow.

 

A good rating should not include win rate but in a majority of cases should follow win rate.

 

As for WG's new rating, they are including blocked damage but that's a stat they have not always collected.  Same goes for spotting assist.  So what are they going to do about us old timers?  Are they going to scrap however many years worth of stats?  Can they even split them by date?  Are they going to wipe everything and start again?



Mike_Mckay #14 Posted 13 September 2017 - 06:09 AM

    Second Lieutenant

  • Player
  • 18668 battles
  • 1,488
  • [-AWF-] -AWF-
  • Member since:
    09-02-2015

View PostSuedKAT, on 13 September 2017 - 01:41 AM, said:

So they have basically taken a bit of what third party stats utilize and thrown in stuff they feel are important parameters when determine "skill" in WoT. I kinda like the idea, but looking at what they have included and what they've left out as well as how large/small part it plays in the equation kinda makes me feel a bit confused.

 

Having a rating built on survival rate is just wrong on so many levels (not to mention the amount, 27% for the Bat as an example), I fail to see how that can be relevant to any rating and what a high versus low survival rate would show, sure if you consider it on it's own you can draw some conclusion about how a player play his tank but that's about it. I'd replace it with Win Rate instead, then you at least measure things that directly impact a match to some extent, if it's won or not, the amount of damage and blocked damage, assists and kills.

 

I'm a bit torn on having different percentages on different tanks, since as you say it don't really matter what tank your in, if you contribute in some way it should show up in the ratings. On the other hand utilizing your tank for "what WG intended you to do" might also be nice to measure since getting a bit lower rating for not spotting in your LT or not blocking any damage in your HT most likely means you play the tank a bit "wrong" so to speak, you know those IS-7 sniping players etc.

 

I guess if one player did 2000 dmg and 1000 spotting and died but another did 2000 dmg and 1000 spotting and survived it would show the second was the better player


By itself you are right it doesn't show anything as without any context it could just show they got stuck on a rock at the start of the game and didn't get found before the end lol, but I don't see an issue with it contributing towards the overall figure its just how much that's the issue


as for the different amounts for each tank class it should really be per tank, some lights and meds are more like heavies at their tier, some heavies are more like mediums or TD snipers and some heavies aren't designed for brawling in the slightest so taking hits for the team isn't exactly wise without the armour to back it up


So rewards for merely "taking hits" would reward driving down the middle of the map in a lightly armoured tank and getting shot to pieces by tanks that aren't even in your view range

Considering WG couldn't even maintain balance with the tanks they had to start with and THEN started throwing in tanks outside of any balance characteristics they might have started off with meaning for the sake of a dozen tanks or so the other 200 or so then needed to be "rebalanced" rather than just balancing the extra dozen to match the existing 200 I don't have a lot of faith in their ability to balance a performance system to match each tanks characteristics in the slightest.

 

 

In fact it seems like them just creating another workload they wont be able to cope with and which will just annoy even more players than the tank imbalances already do



Long_Range_Sniper #15 Posted 13 September 2017 - 08:16 AM

    General

  • Player
  • 30494 battles
  • 8,247
  • [-MM] -MM
  • Member since:
    04-04-2011

View Postlaulaur, on 12 September 2017 - 10:50 PM, said:

WG have so many bad ideas lately, why should anyone be rewarded for surviving the battle or blocking damage??? :amazed:

 

It's obvious. You're all assuming the rating system was designed by the dev's, and in reality it was the sales and marketing department.

 


ZlatanArKung #16 Posted 13 September 2017 - 08:20 AM

    Major General

  • Player
  • 1529 battles
  • 5,112
  • Member since:
    12-20-2014

View PostHomer_J, on 13 September 2017 - 03:27 AM, said:

 

The only reason you would include win rate is to fudge the figures so the two follow.

 

A good rating should not include win rate but in a majority of cases should follow win rate.

 

As for WG's new rating, they are including blocked damage but that's a stat they have not always collected.  Same goes for spotting assist.  So what are they going to do about us old timers?  Are they going to scrap however many years worth of stats?  Can they even split them by date?  Are they going to wipe everything and start again?

 

Why shouldn't the rating follow WR for all cases?

 

I would like if the rating was tier and/or class specific.

 

WR can be padded at lower tiers, dmg can be padded without being of use to teams chance of winning.

 

Rating could be, only T5 or only T6+ tanks.

 

Damage could be rated on how early you do it combined with survival or average lifetime in battle (yolo shouldn't be rewarded, while aggressive early damage should).

Like, you have a formula like this.

(1/Time[in seconds]) × damage [gives points depending on how early you do damage].

Combine with time of death (after battle). So an early death gives low multiplication while a late death gives a larger one.

 


Edited by ZlatanArKung, 13 September 2017 - 08:29 AM.


Obsessive_Compulsive #17 Posted 13 September 2017 - 08:24 AM

    General

  • Player
  • 24245 battles
  • 8,048
  • [ADUK] ADUK
  • Member since:
    09-09-2014

View PostJigabachi, on 13 September 2017 - 01:27 AM, said:

Every rating that tries to condense a whole set of diverse information into a single number sucks by default. That rating will be as useless as the previous one and as useless as WNx.

 

Not really. WR does exactly that is a pretty good indicator and far from sucks.

tajj7 #18 Posted 13 September 2017 - 08:33 AM

    Field Marshal

  • Player
  • 24835 battles
  • 13,836
  • [-MM] -MM
  • Member since:
    03-30-2014

View Postbrumbarr, on 12 September 2017 - 10:40 PM, said:

As I posted in a thread some time ago, WG is making a new rating system and hall of fame:

 

https://thedailyboun...l-of-fame-beta/

 

While it looks good overal, I do have a few problems with it:

-survivial rate should not be a factor in any rating system, how much you survive is irrelevant to how good you play. All that matters is the results, not how you got there, an aggresive player who gets like 3K dpg but only survives 20% of matches should get the exact same rating as a more passive player who also gets 3k dpg but surives 40%.

 

-same thing with blocked dmg, blocked dmg in itself is useless, blocking dmg is a tool to get a result, its a tool to get more dmg and win the game. Its not a result in itself. 

A maus player who sits in the open for no reason and gets the same dmg as a maus player who only peaks around the corner when his gun is reloaded should get the same reward.

 

- Every vehicle should get the same % weighting of factors. Imagine a scout player who  focuses more on dealing dmg.  He does 2K dmg per game and 1K spotting, then another scout player who does 1K dmg and 2K spotting. Both contribute just as much to the match, so they should get the same rating, if you play a vehicle different than how it is supposed to be played but are succesfull  in it you shouldnt be labeled lower skilled.

 

I would completely remove survival rating and heavily reduce the % of blocked dmg to the rating, at max 15% . I would make the weighting class based, not tank based, for example:

Heavy tanks: 15% blocked, 15% kills, 40% dmg, 30% assist.

Meds: 5% blocked, 15% kills, 45% kills , 35% assist.

Lights: 0% blocked, 15% kills, 45% dmg, 40% assist.

TDs:  5% blocked, 15% kills, 40% dmg, 40% assist.

Arty:  100% ammount of shots missed

 

 

 

 

Survival rating = base campers

Damage blocked = more reward to idiot heavy tank players.

 

They are just adjusting the rating so that terrible players look less terrible. It means when people tell them they are crap for camping in base or being parked out in the open in their heavy they can point to their 'WG rating' which tells them they are not crap.

 

It's like WG's way of including a 'participation medal' into a stats rating. The fat kid at sport day gets a medal and doesn't feel as bad, WOT's is largely made up of fat kids at a sports day, that is 90% of the playerbase. 

 

They've been going down this road for ages, corridor maps so players don't have to deal with flanking and have awareness, OP base camping spots so they don't have to actually use the W key, overbuffed armour, even side armour so they can get away with their terrible plays.  More and more things to reward and placate 'Steve the IS7 driver' who has 40k battles and still hasn't learnt the game. 

 

Also another thing this rating shows is they very clearly want to shoehorn classes into narrow roles and kill diversity amongst tanks, they tried it in the first sandbox, lights must scout, TDs must camp and snipe, heavies must brawl etc. etc. Again this helps out the braindead playerbase, you know the guys who always go 'heavies must go here' 'lights must scout' etc. 



DracheimFlug #19 Posted 13 September 2017 - 08:56 AM

    Brigadier

  • Player
  • 8957 battles
  • 4,033
  • Member since:
    11-13-2014

View Posttajj7, on 13 September 2017 - 08:33 AM, said:

 

Survival rating = base campers

Damage blocked = more reward to idiot heavy tank players.

 

They are just adjusting the rating so that terrible players look less terrible. It means when people tell them they are crap for camping in base or being parked out in the open in their heavy they can point to their 'WG rating' which tells them they are not crap.

 

It's like WG's way of including a 'participation medal' into a stats rating. The fat kid at sport day gets a medal and doesn't feel as bad, WOT's is largely made up of fat kids at a sports day, that is 90% of the playerbase. 

 

They've been going down this road for ages, corridor maps so players don't have to deal with flanking and have awareness, OP base camping spots so they don't have to actually use the W key, overbuffed armour, even side armour so they can get away with their terrible plays.  More and more things to reward and placate 'Steve the IS7 driver' who has 40k battles and still hasn't learnt the game. 

 

Also another thing this rating shows is they very clearly want to shoehorn classes into narrow roles and kill diversity amongst tanks, they tried it in the first sandbox, lights must scout, TDs must camp and snipe, heavies must brawl etc. etc. Again this helps out the braindead playerbase, you know the guys who always go 'heavies must go here' 'lights must scout' etc. 

 

Again, doesn't it depend on how things are weighted? But I am sure that you are quite correct. Absorbing damage is completely useless. All armour in the game should be either removed or pen increased to make it useless, since it (apparently) requires no skill at all to use well..... everyone always places their tank at exactly the right angle in every situation... easy, right?

 

Is rating biased by tank class? If not, then how would it shoehorn anyone?



clixor #20 Posted 13 September 2017 - 09:17 AM

    Major

  • Player
  • 50380 battles
  • 2,995
  • Member since:
    08-07-2011

View Posttajj7, on 13 September 2017 - 08:33 AM, said:

 

Survival rating = base campers

Damage blocked = more reward to idiot heavy tank players.

 

They are just adjusting the rating so that terrible players look less terrible. It means when people tell them they are crap for camping in base or being parked out in the open in their heavy they can point to their 'WG rating' which tells them they are not crap.

 

It's like WG's way of including a 'participation medal' into a stats rating. The fat kid at sport day gets a medal and doesn't feel as bad, WOT's is largely made up of fat kids at a sports day, that is 90% of the playerbase. 

 

They've been going down this road for ages, corridor maps so players don't have to deal with flanking and have awareness, OP base camping spots so they don't have to actually use the W key, overbuffed armour, even side armour so they can get away with their terrible plays.  More and more things to reward and placate 'Steve the IS7 driver' who has 40k battles and still hasn't learnt the game. 

 

Also another thing this rating shows is they very clearly want to shoehorn classes into narrow roles and kill diversity amongst tanks, they tried it in the first sandbox, lights must scout, TDs must camp and snipe, heavies must brawl etc. etc. Again this helps out the braindead playerbase, you know the guys who always go 'heavies must go here' 'lights must scout' etc. 

 

It's the combination, and weight, of the factors that matter. If you have high blocked dmg, but low spot, then it's likely a heavy that yolo's (without support). But say if a tank in that situation also get's high spotting he's being very useful. And, again, it's depending on the weight, but as far as i can see the rating does represent this, so it should give some indication of how someone performs.

 

On the matter of WR, as long it's individual tank wr, i don't have a problem with it being included. But if it's overall it can be padded and shouldn't be included.






1 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users