Jump to content
bjshnog

⟪WN8⟫ Development / Resources

Recommended Posts

So if i am understanding this correctly, I should go back and rebuy tanks that i played in my noobish days to help my current WN8 rating more accurately reflect my current skills?  Or will it not really matter?

 

Does say for example my M3 Lee stats from like my first 400 games ~= my last 400 games in my T49 or T54 as far as my overall skill rating calculation? Do they carry the same weight?

 

Don't get me wrong, simply a question, as i have been reading through this from time to time to get a further understanding.

 

You are not understanding it correctly, although that misconception APPEARS to be pretty common. 

 

Stats are weighted by games played. My PzIII/IV has terrible stats, it was my first tier 5. My T49 has stellar stats, and is my most recent played tier 5. They have 86 and 148 games respectively. If I went back and played my PzIII/IV it would help my WN8 rating. Would it help much more than just accumulating more games on my T49? Not really, as in either case I'd be performing with my current skill level...AND most importantly, I do better in the T49 than in PzIII/IV because...skill ceiling. 

 

Now imagine someone name BeatCrabOff who has my identical abilities. Who has played (86 + 148) games in the PzIII/IV for some perverse reason. Right now WN7 would rate BeatCrabOff worse than me, because the results in that tank are going to be a bit worse, because the tank is worse. But in WN8, the standard to which the PzIII/IV will be held will not be the same as the standard as the T49 is held, it will be a little lower. So BeatCrabOff would get a slight bump from WN7 to WN8. 

 

Using a more realistic example, right now the T69 and WZ-132 have the same weight in WN7. But played by the same good player, they might have 2400 and 1800 dpg respectively. WN8 will more accurately measure an individuals tank mix.

 

But you can't just game WN8 by playing terrible tanks, you have to play those terrible tanks well. 

Share this post


Link to post
Share on other sites

Thanks for the explanation, that makes sense to me now.

 

So any future games played will be measured on their own merit so to speak.  You can go back and "fix" bad tanks, but since each game is measured independently to give your aggregate(WN) score, you are better off playing well in a tank you like/know you can perform in.

 

 

Correct?

 

Share this post


Link to post
Share on other sites

Buncha bullshit at the end of this thread. Either try to read, learn to read, or don't fucking post. We don't have to tolerate it here, thankfully. (I prefer this type of verbal warning before I delete your future drivel, it helps make it clear I'm not censoring dissent, but idiots)

 

lifeless, bjshnog's assessment stands. Are you clueless or trolling?

 

Belial - if you think people are going start "farming" WN8 on bad tanks...well you're paranoid as fuck. Sure, if I were to jump into a Super Pershing I'd be able to exceed the standards set, and it might bump my WN8. But I personally would have to be a sick fuck to play a shitty no-carry SP to 'inflate' my WN8*. Is it really inflation since I am already good at WoT? Secondly, which is is easier - exceeding the Super Pershing standards by a good margin, or playing tier 10 and tier 1 in a 2 to 1 ratio and pushing out 2200+ WN7 scores with only marginal performances at both tiers. Praetor made it very very clear why average damage per tier was insufficient.

 

*And sure, someone might do it. But we're going to keep re-norming** the performance standards per tank, so any attempts to manipulate will be disarmed...

 

I really I don't care what you do, it affects me none. My assessment remains. I have nothing else to say to you other than im right and when it goes live you will see it.

Troll-Bait

Flaming

Repeated Tinfoil Hattery Posts

If you're going to be a cancer to this community, you will be cut out.

Indefinite RO

~Kitti

Share this post


Link to post
Share on other sites

Thanks for the explanation, that makes sense to me now.

 

So any future games played will be measured on their own merit so to speak.  You can go back and "fix" bad tanks, but since each game is measured independently to give your aggregate(WN) score, you are better off playing well in a tank you like/know you can perform in.

 

 

Correct?

 

Exactly. "Fixing" bad tanks would give you the same WN8 as playing "good tanks", as long as you perform equally well in both of them. That is the whole point of WN8, maesure player skill regardless of tank selection.

Share this post


Link to post
Share on other sites

I really I don't care what you do, it affects me none. My assessment remains. I have nothing else to say to you other than im right and when it goes live you will see it.

 

I wouldn´t call it an assessment. I would call it an irrelevant, narrow-minded, tinfoil-hat fallacy.

Share this post


Link to post
Share on other sites

Oh guys, ack to important discussions, Mr. Noobmeter shared his own expected damage per tank table, please take a look at it and compare...

 

http://www.noobmeter.com/tankList

 

I think this is a very good example of why simply using the data without analyzing and processing it is a bad idea:

 

Damage expected for NC-31 is 145 and 90 for MS1? o_o

Share this post


Link to post
Share on other sites

I read a bit of this thread and my head is spinning at the absurdity I saw typed; here are my 2 cents.. if you want to devalue WN as a rating platform, do go ahead and start screwing with it based on the tanks people play.  We laughed at how irrelevant WG's rating is, yet here we you are going down the same path.

 

It is beyond arrogant to try to dictate what tanks people play in order to get ratings.. for one it will backfire, and for second, it will further devalue platoons, which are already seen as stat padding means on their own, regardless of tanks.

 

The power of WN comes from people considering it an accurate and FAIR measure ones APPROXIMATE skills. When you begin to ADJUST IT according to the NEEDS of CERTAIN players (which is what wn8 is really, lets be frank) it will hold as much weight as well.... paper.

 

feel free to neg :)

 

 

I really I don't care what you do, it affects me none. My assessment remains. I have nothing else to say to you other than im right and when it goes live you will see it.

 

I want to go on record, before someone else perma-bans you for being a snot (I make it a policy to never take an administrative action against anyone I've personally disputed with...)

 

WN8 specifically is designed to NOT dictate what tanks people should be playing. If you want to farm WN7, you play fast autoloaders, tanks with excellent premium rounds, and a mix of high damage high tiers and low tier tanks to manipulate the fact that damage does not scale linearly with tier. My 2500+ tanks are all autoloaders - batchat, t69, t54e1, 50 100. Mix any of these with a tier 1 or 2 tank and I can go over 3k WN7 any day I please. WN8 means that tanks I play well don't get measured against superior tanks in the same tier. The IS-3 I played at the same time as the 50 100, and perhaps even better, except that autoloaders get an inherent bonus in kills, as do faster tanks (beyond a certain threshold of skill). The IS-3 has 300 points lower WN7 value for me (via dossier parse). WN8 should help correct for that. 

 

Your comment about platoons is actually just nonsense. I don't understand your point, and cannot see how platoons are either valued or devalued particularly in either rating systems, and then you meander into some comment about other seemingly irrelevant views about platoon padding. 

 

The point of WN8 is an improvement in accuracy and fairness. We've adjusted it every iteration. In fact, this is the fairest adjustment yet. Who are the certain players who this is being adjust for? Aside from the people who play lots of games in a given favorite tank (which is a LOT of people). Perhaps you're insinuating that Praetor and I are giving ourselves points in WN8 because WN1-7 measure LTs poorly, and we both like LTs? Its fairly easy evidence that most of the population thinks LTs could use a little more love in the ratings, since their value doesn't generally come from kills/dmg!

 

So please, feel free to back up your assertions. Someone who doesn't have arguments, evidence or contributions is of little value here though!

Share this post


Link to post
Share on other sites

on the question asked a few days ago of rating scale: can I put a strong request in for "average" WN8 to be roughly the same value as "average" WN7: ie, preferably 1000, but 1200 will do.

 

people like metrics because they're comparable, and if you change the (roughly) expected value for average from what it is now to, well, any other value, you are going to have to re-educate the player base as well as fight against the batshit insane conspiracy theorists, who are already making a presence here. If the typical player can look at WN8 and see roughly the values they expect to see (apart from some 'purple buffing bias', which will come anyway), then the tinfoil hat wearers will be isolated and eventually ignored.

 

If you make the whole rating system incomprehensible using the framework of the existing rating, by screwing with the baseline average figure, all the people who don't understand that 25 is the new 1000, or 2500 is the new 1000, or whatever, will be inclined to listen to the tinfoil hat wearers, and the likelihood of WN8 acceptance will sharply decline.

 

There's a very good reason why IQ tests are set to a mean value of 100 and standard deviation of 15: because even if the test changes, and/or the tested population's comprehension of the test changes, the metric remains comparable, and so no-one cares. Making the mean value of WN8 look as much as possible like the mean value of WN7 will save all sorts of pain and drama not very far down the line...

Share this post


Link to post
Share on other sites

Oh guys, ack to important discussions, Mr. Noobmeter shared his own expected damage per tank table, please take a look at it and compare...

 

http://www.noobmeter.com/tankList

 

I think this is a very good example of why simply using the data without analyzing and processing it is a bad idea:

 

Damage expected for NC-31 is 145 and 90 for MS1? o_o

I am now chat with him about this data. His mistake, which he takes to the average data rather than top. Therefore tanks like KV-1S or Hellcat get underestimated requirements, because they played a lot of bad players.

I have a database of all the statistics that were in the XVM database in March of this year. And I gave him a table, where he can choose the desired percentage of the top players, or place the player on the damage. Maybe for you it will be useful, too:

7TrLaak.png

http://rghost.ru/48933887

You can choose any a percentage or an absolute position (in the tooltip can see the available options), and apply filters on the level, type and nation ...

It old, but very big statistics.

* In statistics, the only players who have spent at least 100 battles at the tank.

Share this post


Link to post
Share on other sites

on the question asked a few days ago of rating scale: can I put a strong request in for "average" WN8 to be roughly the same value as "average" WN7: ie, preferably 1000, but 1200 will do.

 

people like metrics because they're comparable, and if you change the (roughly) expected value for average from what it is now to, well, any other value, you are going to have to re-educate the player base as well as fight against the batshit insane conspiracy theorists, who are already making a presence here. If the typical player can look at WN8 and see roughly the values they expect to see (apart from some 'purple buffing bias', which will come anyway), then the tinfoil hat wearers will be isolated and eventually ignored.

 

If you make the whole rating system incomprehensible using the framework of the existing rating, by screwing with the baseline average figure, all the people who don't understand that 25 is the new 1000, or 2500 is the new 1000, or whatever, will be inclined to listen to the tinfoil hat wearers, and the likelihood of WN8 acceptance will sharply decline.

 

There's a very good reason why IQ tests are set to a mean value of 100 and standard deviation of 15: because even if the test changes, and/or the tested population's comprehension of the test changes, the metric remains comparable, and so no-one cares. Making the mean value of WN8 look as much as possible like the mean value of WN7 will save all sorts of pain and drama not very far down the line...

 

That was my thinking in setting WN8 average to be as similar as possible to WN7 average. We can discuss the scale later on, once we have the kinks worked out.

Share this post


Link to post
Share on other sites

WN8 specifically is designed to NOT dictate what tanks people should be playing. If you want to farm WN7, you play fast autoloaders, tanks with excellent premium rounds, and a mix of high damage high tiers and low tier tanks to manipulate the fact that damage does not scale linearly with tier. My 2500+ tanks are all autoloaders - batchat, t69, t54e1, 50 100. Mix any of these with a tier 1 or 2 tank and I can go over 3k WN7 any day I please. WN8 means that tanks I play well don't get measured against superior tanks in the same tier. The IS-3 I played at the same time as the 50 100, and perhaps even better, except that autoloaders get an inherent bonus in kills, as do faster tanks (beyond a certain threshold of skill). The IS-3 has 300 points lower WN7 value for me (via dossier parse). WN8 should help correct for that.

 

I'm sure that this has been covered in the past but you really want to be careful how much you want to boost or drop a players rating if they don't reach a target dpg rating.

 

ie: a player currently in a T57 can achieve xxxxx dpg. There's a nerf in 8.9 such the games played in this tank craters and the AMX 50 B becomes the new hotness. If I finally get that tank unlocked I might be doing 300 dpg based on the new rebalanced tank abilities but I would get slammed for that.

Share this post


Link to post
Share on other sites

I just wanted to drop in and say thanks to the guys working on this. I've been following this thread since its inception and really like where WN8 is going. Keep up the good work guys.

Share this post


Link to post
Share on other sites

I want to go on record, before someone else perma-bans you for being a snot (I make it a policy to never take an administrative action against anyone I've personally disputed with...)

 

WN8 specifically is designed to NOT dictate what tanks people should be playing. If you want to farm WN7, you play fast autoloaders, tanks with excellent premium rounds, and a mix of high damage high tiers and low tier tanks to manipulate the fact that damage does not scale linearly with tier. My 2500+ tanks are all autoloaders - batchat, t69, t54e1, 50 100. Mix any of these with a tier 1 or 2 tank and I can go over 3k WN7 any day I please. WN8 means that tanks I play well don't get measured against superior tanks in the same tier. The IS-3 I played at the same time as the 50 100, and perhaps even better, except that autoloaders get an inherent bonus in kills, as do faster tanks (beyond a certain threshold of skill). The IS-3 has 300 points lower WN7 value for me (via dossier parse). WN8 should help correct for that. 

 

Your comment about platoons is actually just nonsense. I don't understand your point, and cannot see how platoons are either valued or devalued particularly in either rating systems, and then you meander into some comment about other seemingly irrelevant views about platoon padding. 

 

The point of WN8 is an improvement in accuracy and fairness. We've adjusted it every iteration. In fact, this is the fairest adjustment yet. Who are the certain players who this is being adjust for? Aside from the people who play lots of games in a given favorite tank (which is a LOT of people). Perhaps you're insinuating that Praetor and I are giving ourselves points in WN8 because WN1-7 measure LTs poorly, and we both like LTs? Its fairly easy evidence that most of the population thinks LTs could use a little more love in the ratings, since their value doesn't generally come from kills/dmg!

 

So please, feel free to back up your assertions. Someone who doesn't have arguments, evidence or contributions is of little value here though!

 

You can stop being a little twat and ban me, since you've been itching for a few months now. Moving on from your little ego drama

 

What i was bringing up to people's attention is based entirely on the info provided in this thread, those are not my theories; they are yours, im just saying why they are not good.  I didn't comment on it before as I wanted to see if people were really serious about the direction they want to take wn8 in.  Currently that seems to be the matured idea of adjusting it for each tank within tier based on an arbitrary criteria of values.  

 

You want to negatively adjust the WN for certain tanks under the premise they are "padders".  That's wrong for so many reasons. But the simplest reason is this.. that is WG's job. They adjust tanks and game mechanics.  The rating is supposed to be just that, a rating, not some sort of a dictator as to what tanks people play. If you don't agree with the way tanks are, how do you see it fit to be your call how each tank is weighted?  That's not fair, that's just you waving your arms at wg being angry they don't adjust the game to your liking.

 

My point about platoons is, you can have 3 people, 2 of which use OP tanks and the 3rd farms WN8 in a "not negatively adjusted tank". They each rotate the tanks every game, padding wn8 without suffering through the agony of playing shit tanks solo, while the other 2 carry the game. is6 is6 fcm | t57 t57 fv4002 etc. They are doing it now, but wn8 will just encourage it more.  If you can't see the flaw in that, you have no right to wave a finger about my contributions. If I don't agree with your bs, it is just that, non agreement. I don't have to agree with you and suck up to you so you like me. But I expect you to sit and listen when concerns are raised, not try to dismiss them and go on the attack and wave around threats.

 

WN7 doesn't measure LTs badly, it doesn't measure spotting damage due to WG not releasing this through the api. Im not sure why you even said that? If you get the damage/kills in a LT its no different.

Share this post


Link to post
Share on other sites

I'm sure that this has been covered in the past but you really want to be careful how much you want to boost or drop a players rating if they don't reach a target dpg rating.

 

ie: a player currently in a T57 can achieve xxxxx dpg. There's a nerf in 8.9 such the games played in this tank craters and the AMX 50 B becomes the new hotness. If I finally get that tank unlocked I might be doing 300 dpg based on the new rebalanced tank abilities but I would get slammed for that.

 

Agreed, history is ALWAYS the biggest threat to validity to WN* and will continue to be so. Luckily, the averages don't move all that terribly much, even with big nerfs and buffs. There aren't incremental bonuses at given values btw, its just setting an average value, and then scaling outwards. We don't for example have per-tank standard deviations, as nice as that would be, WN8 would become a true computational nightmare. 

 

But yea, there is no correcting for re-tiered tanks, or ones that have been horribly nerfed (RIP Lorraine 40t, VK2801). 

 

Using your example, we've signed ourselves up for regular re-balancing of per-tank stats, with history in consideration. Its a thankless task. But unless the T57 got horribly nerfed AND you then immediately played say...500+ games in it, it would be barely noticeable in your account wide WN8 :-D

Share this post


Link to post
Share on other sites

Thanks seriych! :)

If it will be useful, I can do similar tables by other parameters, or make a selection of more than 100 battles on tank.

Or I can upload .csv file like this:

id;nick;battles;winrate;vname;vbattles;vwinrate;vdamage;vfrags;vspotted;vsurvived
500000094;CunningB;5556;55,62%;GB11_CAERNARVON;141;55,32%;74;0,04;0,09;1,42%
500000187;Butterz;5397;47,10%;LOWE;1090;45,32%;1377;1,12;0,90;37,61%
500000187;Butterz;5397;47,10%;M46_PATTON;492;51,63%;1678;1,23;1,67;28,86%
...

But it about ~4GB.

 

For those who don´t know seriych is one of the developers of XVM.

No, I just like to play around with the numbers sometimes :-)

Share this post


Link to post
Share on other sites

You can stop being a little twat and ban me, since you've been itching for a few months now. Moving on from your little ego drama

 

What i was bringing up to people's attention is based entirely on the info provided in this thread, those are not my theories; they are yours, im just saying why they are not good.  I didn't comment on it before as I wanted to see if people were really serious about the direction they want to take wn8 in.  Currently that seems to be the matured idea of adjusting it for each tank within tier based on an arbitrary criteria of values.  

 

You want to negatively adjust the WN for certain tanks under the premise they are "padders".  That's wrong for so many reasons. But the simplest reason is this.. that is WG's job. They adjust tanks and game mechanics.  The rating is supposed to be just that, a rating, not some sort of a dictator as to what tanks people play. If you don't agree with the way tanks are, how do you see it fit to be your call how each tank is weighted?  That's not fair, that's just you waving your arms at wg being angry they don't adjust the game to your liking.

 

My point about platoons is, you can have 3 people, 2 of which use OP tanks and the 3rd farms WN8 in a "not negatively adjusted tank". They each rotate the tanks every game, padding wn8 without suffering through the agony of playing shit tanks solo, while the other 2 carry the game. is6 is6 fcm | t57 t57 fv4002 etc. They are doing it now, but wn8 will just encourage it more.  If you can't see the flaw in that, you have no right to wave a finger about my contributions. If I don't agree with your bs, it is just that, non agreement. I don't have to agree with you and suck up to you so you like me. But I expect you to sit and listen when concerns are raised, not try to dismiss them and go on the attack and wave around threats.

 

WN7 doesn't measure LTs badly, it doesn't measure spotting damage due to WG not releasing this through the api. Im not sure why you even said that? If you get the damage/kills in a LT its no different.

 

The values are not arbitrary, they are culled from the vbaddict database, which parses masses dossiers. It might be subject to selection bias, which would be a valid criticism, but Praetor did not just start typing values into an excel spreadsheet based on how he feels a tank performs.

 

The point you are missing is that achieving a kill in the T57 isn't reflective of the same ability at WoT compared to achieving a kill in the T110E5. This isn't my opinion, its reflected in the data. The same person playing those two tanks in the same time period will average more kills in the T57. Does this mean the T57 is absolutely a better tank? We don't make that judgement, but we also know that we don't have access to data like potential damage received, which the T110E5 does better. So in measuring the guy who played 1,000 T110E5 games against the guy who played 1,000 T57 games on the basis of kills and damage, we need to normalize those values - and we've elected to do so based on the values culled from the vbaddict data, with commentary from a public panel of reviewers. And we will revise the values again, because surely we didn't get them all right the first time. 

 

This has little to do with game balance, balancing the tanks is indeed WG's job, we're not correcting for values because we think the tanks are imbalanced (although that terminology does often get used) but because the value of 1 kill isn't the same on tank X as it is on tank Y, let alone across tiers!

 

I still don't see how you pad WN8 in platoons by rotating in a shit tank. Shit tanks are shit tanks even under good conditions, and win-rate is no longer a component in WN8 scoring. If you could explain, using the formula for reference, how this works, it might be useful. Do the two people in T57s let the FV4202 get killshots? Do they help the 4202 do more damage by only shooting 3 from their clip instead of 4? I don't understand the scenarios in which it could be systematically possible to "feed" someone WN8. And anyone with the ability to manipulate the game consistently on that level is going to be top 1% anyways, where measurement always breaks down anyways (at the tails). 

 

WN7 does indeed measure LTs badly. LTs average very low damage per tier (which is what WN7 is computed on) when compared to MT/HT/TD and consistently have low kpg (we're talking tier 5+ LTs with scout MM here), with hardly anyone exceeding even 1.5 kpg. In-game this is balanced by the fact that LTs have a whole bunch of advantages that are hard to quantify (distraction, timing) or are not reported (DUD). While reporting DUD would go a long ways toward helping LT scoring,this method still means we can measure everyone's T71 against everyone else's T71, instead of against T29s and IS-2s and SU-122-44s. 

Share this post


Link to post
Share on other sites

If it will be useful, I can do similar tables by other parameters, or make a selection of more than 100 battles on tank.

Or I can upload .csv file like this:

id;nick;battles;winrate;vname;vbattles;vwinrate;vdamage;vfrags;vspotted;vsurvived
500000094;CunningB;5556;55,62%;GB11_CAERNARVON;141;55,32%;74;0,04;0,09;1,42%
500000187;Butterz;5397;47,10%;LOWE;1090;45,32%;1377;1,12;0,90;37,61%
500000187;Butterz;5397;47,10%;M46_PATTON;492;51,63%;1678;1,23;1,67;28,86%
...

But it about ~4GB.

 

No, I just like to play around with the numbers sometimes :-)

 

That would actually be a huge help... could you do mediafire, copy.com, or dropbox? that would really be grand.

 

BTW: Unsurprisingly, the thread on the official forums turned into a centrifuge of shitpoasting and unfundamented fallacies. Shnog, please don´t do that again, we have the discussion in here for a reason. What % of the players do you think understands what a ratio scale is? What percentage of players has 6ht grade level education in math, much less statistics? Im all for open discussion, but people spewing crap, outright lies, and making 100% unfundamented ignorant comments simply halts progress instead of promoting it. I have had enough of that crap on the original WN7 post in the WG forum, and moved it here for a reason...

Share this post


Link to post
Share on other sites

I should reserve judgement till I see Wn8 in action.

 

 

Hopefully you can get a program setup to check it out like you said.

 

I have a feeling my overall wn8 will be terrible because 3/4's of my tanks I played are on a blue level with only a quarter with purple levels.

 

Heavytwenty should come out good in wn8 every tank he ever plays is at the upper end of the 1.5 scale!

 

 

And your right it's just talking out my ass without numbers to back it up  :tongue:

 

 

Share this post


Link to post
Share on other sites

I think maybe the biggest problem I had with the bot baseline (third time I've said biggest problem, gah), which didn't occur to me, is that those numbers are way too high. Watching actual bot gameplay shows that they pretty much do absolutely nothing. The bot baseline of spots would/should probably be highest by far, but damage should be closer to about 10-15% of an average player with frags about the same. Defense probably in between. Neatoman made a good point on the thread I made on the official forums, which was that those 42% numbers are not bot numbers. They are just bad platoon numbers. Also, vbaddict probably doesn't have all that many bots.

Share this post


Link to post
Share on other sites

Once again, forget the "bot" stuff. These are the stats for people who can´t do enough to influence any matches they are in. You seem to be worried due to the fact that many players will get 0 WN8, or will get lower scores than WN7, when the reality is, since the multiplier is higher than before, 90% of players will get a higher WN8 rating than WN7.

Share this post


Link to post
Share on other sites

Forgive me for playing devils advocate...

 

Picture two accounts which presently are at the same skill level, but polar opposite ends of the 'how we got to where we are today' scale. Interestingly enough, for both of these people the T49 happens to be one of the more influential vehicles on their overall statistics. Lets name them Iarulla and Iarukcah for the sake of keeping track.

 

Iarulla was a players first account, he ground up from tier 1 with no prior game knowledge.

In the T49, Iarulla played ~1k games prior to the physics release, and another 2k prior to the gun sigma changes; both of these changes changed the balance of T49 considerably for the better providing large mobility and gun reliability gains with each change. Iarulla played many games in the vehicle in an aspect of discovery of best practices and teaching skills and mechanics to friends. The bulk of Iarulla's games were played during the timeo f the derp, where M4s and PzIVs were several fold more common - as the 105mm HE shell reliably 1-shot-killed a T49, this was a big deal.

 

Iarukcah is a slightly different beast, he started much more recently and had prior game knowledge. Going straight for the T49 knowing how awesome it was at present, and that it was soon to be buffed by the sigma changes. With the sigma changes the 105mm Derp was no longer nearly as effective or used by the remainder of the population - the T49 lost its only same tier competitors - life was lush. Iarukcah optimized his crews and paths taken to get the T49 and at tier 3-5 leveraged his prior game experiences to get into platoons with his skill level peers with only 150 battles under his belt. From there, with a freshly elited T49, Iarukcah clawed his way up to some quite impressive stats - certainly helped by his platoon to win, those same people were taking valuable damage and kills too. Iarukcah got back into solo games and 2 man platoons so the ceiling on individual contribution was higher. Iarukcah plays for stats and winning, and his account certainly shows that.

 

Despite both players currently being at the same skill level, Iarukcah's stats are 20-50% better that Iarulla's stats in every comparable metric. If you looked at the two players without context you would surely ascertain that one is a much better player than the other; even if one had a much smaller sample size. I f you look at the top percentile of T49 players now, Iarukcah sits among the top of them while you'll need to visit the next percentile to see where the previous era kingpin now sits.

 

The T49 was powerful pre physics, but camo and gun depression were about all it had going for it. The mobility was good, but not being able to traverse up and down zones of terrain meant that turning was much more important, and as such the vehicle was much less nimble and less able to control maps as it does now. The gun was also much less reliable, especially at longer ranges where firing at targets more than 400m away was considered a great waste of ammunition. After the sigma changes, if you can see it, you can kill it. While the relative power of the T49 increased, simultaneously the skill floor was lowered and the skill ceiling raised at each patch also.

 

If you compare the current era T49 to the introduced era T49, it's certainly more powerful and influential on each game now than it previously was; but each and every tank has also made that progression; where the top percentiles of the player base can each push higher and higher performance outliers in a vehicle while the median median performance will have only slightly crept up. The high skill floor means that its high skill ceiling isn't utilized well or at all by most of the population. It's median and mean averages are in fact lower than most vehicles of the same tier, despite the upper performance outliers being much larger and commonly occurring. The T49 to other vehicles of its era it has been slightly ahead at each step compared with its peers, but its ability to perform incredibly well has increased over time.

 

As the vehicle is considered quite OP, I assume that an increase in average tier for the T49 is coming. While Iarukcah will be normalized down, he will remain in the top percentile of T49s which will be more than comparable to other tanks of its tier and power; however the same is not true for Iarulla. With much lower initial stats, the normalization will push his head under the surface of the red sea - where the previously great performance of the T49 isn't comparable to a current era performance averages at tier 5.

 

So if these two players were to meet and determine that they are the same player skill and knowledge wise, they would certainly discover that the normalizing metric will impact every player of that vehicle, and not just the targeted players. While Iarukcah is among the causes where normalization is really required to tone him down, any action taken to do so will more severely impact Iarulla's stats and position among his peers than it will Iarukcah. We all know how WN7 will measure each of these 2 players, but with WN8 the gap between them will widen. I'm not convinced that this is a good thing.

 

Another interesting point lies in the decision of what to do from here. If the T49 is normalized harshly (and it probably should be) then Iarukcah can simply switch vehicles at a few hundred played, and pad his stats in a vehicle much more suited to the WN8 climate while Iarulla will be stuck with the several thousand games hes played already.

 

In any case, if you replace my sound reasoning with a dunning kruger substitute and mass hysteria; you're gonna be fucked.

 

 

 

Share this post


Link to post
Share on other sites

Too much supositions to be able to respond anything coherent. Just wait and see how WN8 works. The thing is, if a player was barely decent (average stats on all tanks), then overperfomred on T49, effectively raising his WN7 above what his "true" skill level is, then I think WN8 will be more fair to this player, and measure his skill better. If he has become a better player, then from the moment WN8 is implemented, and moving forward, he could play ANY tank and watch his WN8 rise... T49 or any other.

 

 

As has been said before, history, rebalancing, and metagame changes are the factors which will produce the largest noise for WN8. Nothing we can do about it though.

Share this post


Link to post
Share on other sites

So we agree that every narrative for how every player got to where he is, is going to be impacted to some degree by how WN8 is calculated. I agree with everyone that history, rebalancing and metagame changes are the factors which will produce the largest noise; all of these things can mitigated by periodic data.

 

If we captured milestones of the server averages and each players profile at each set of changes (ie. patch by WG which) then after 1-2 further patches you will have specific numbers of battles in each tank, a start of period average, and an end of period average, and a total number of battles played in each tank. This would then allow you to determine a per tank modifier specific for the period of time that the data was captured. Say if we went back and captured the snapshot for 8.8, or took it now, and then at the release of 8.9, made adjustments for the tanks existing between these those 2 times, and called it WN8a; then started capturing data for the next set of changes.

 

After however long (like 6-8 weeks, or the next patch) we can discuss and debate relative power balance and then release the WN8b formula for the period of time since 8.9 was released - as this is a captured subset of the data, you can retroactively apply the formula to the data captured since the milestone . You only need 2-3 sets of of this data to get an 60day figure from it, or well.... 60 days of data. As long as you capture the data at the right times, your analysis of it can come later on.

 

Keep recording milestones and with a years worth of data you can calculate games played between each milestone with it's individually applicable WNXx rating and include a segment of each on a ratio of games played between each period to accumulate a more comparable and accurate overall metric which has accommodated for all of the history, rebalancing and metagame changes since you started recording it.

Share this post


Link to post
Share on other sites

So we agree that every narrative for how every player got to where he is, is going to be impacted to some degree by how WN8 is calculated. I agree with everyone that history, rebalancing and metagame changes are the factors which will produce the largest noise; all of these things can mitigated by periodic data.

 

If we captured milestones of the server averages and each players profile at each set of changes (ie. patch by WG which) then after 1-2 further patches you will have specific numbers of battles in each tank, a start of period average, and an end of period average, and a total number of battles played in each tank. This would then allow you to determine a per tank modifier specific for the period of time that the data was captured. Say if we went back and captured the snapshot for 8.8, or took it now, and then at the release of 8.9, made adjustments for the tanks existing between these those 2 times, and called it WN8a; then started capturing data for the next set of changes.

 

After however long (like 6-8 weeks, or the next patch) we can discuss and debate relative power balance and then release the WN8b formula for the period of time since 8.9 was released - as this is a captured subset of the data, you can retroactively apply the formula to the data captured since the milestone . You only need 2-3 sets of of this data to get an 60day figure from it, or well.... 60 days of data. As long as you capture the data at the right times, your analysis of it can come later on.

 

Keep recording milestones and with a years worth of data you can calculate games played between each milestone with it's individually applicable WNXx rating and include a segment of each on a ratio of games played between each period to accumulate a more comparable and accurate overall metric which has accommodated for all of the history, rebalancing and metagame changes since you started recording it.

 

Absolutely correct...but the catch is, you need dossiers to do it. 

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...